1 This document describes *changes* to previous versions, that might
2 affect Exim's operation, with an unchanged configuration file. For new
3 options, and new features, see the NewStuff file next to this ChangeLog.
8 JH/01 Use fewer forks & execs for sending many messages to a single host.
9 By passing back the next message-id from the transport to the delivery
10 process, we can loop there. A two-phase queue run will benefit,
11 particularly for mailinglist and smarthost cases.
16 JH/01 Support list of dkim results in the dkim_status ACL condition, making
17 it more usable in the data ACL.
19 JH/02 Bug 3040: Handle error on close of the spool data file during reception.
20 Previously This was only logged, on the assumption that errors would be
21 seen for a previous fflush(). However, a fuse filesystem has been
22 reported as showing this an error for the fclose(). The spool is now in
23 an uncertain state, and we have logged and responded acceptance. Change
24 this to respond with a temp-reject, wipe spoolfiles, and log the error
27 JH/03 Bug 3030: Fix handling of DNS servfail respons for DANE TLSA. When hit
28 during a recipient verify callout, a QUIT command was attempted on the
29 now-closed callout channel, causing a paniclog entry.
31 JH/04 Bug 3039: Fix handling of of an empty log_reject_target, with
32 a connection_reject log_selector, under tls_on_connect. Previously
33 with this combination, when the connect ACL rejected, a spurious
34 paniclog entry was made.
36 JH/05 Fix TLS resumption for TLS-on-connect. This was broken by the advent
37 of loadbalancer-detection for resumption, in 4.96 - which tries to
38 use the EHLO response. SMTPS does not have one at the time it is starting
39 TLS. Change the default for the smtp transport host_name_extract option
40 to be a static string, for TLS-on-connect cases; meaning that resumption
41 will always be attempted (unless deliberately overriden).
43 JH/06 Bug 3054: Fix dnsdb lookup for a TXT record with multiple chunks, with a
44 chunk-separator specification. This was broken by hardening introduced
47 JH/07 Bug 3050: Fix -bp for old message_id format spoolfiles. Previously it
48 included the -H with the id; this also messed up exiqgrep.
50 JH/08 Bug 3056: Tighten up parsing of DKIM DNS records. Previously, whitespace
51 was not properly skipped and empty elements would cause mis-parsing.
52 Tighten parsing of DKIM header records. Previously, all but lowercase
53 alpha chars would be ignored in potential tag names.
55 JH/09 Bug 3057: Add heuristic for spotting mistyped IPv6 addresses in lists
56 being searched. Previously we only had one for IPv4 addresses. Per the
57 documentation, the error results by default in a no-match result for the
58 list. It is logged if the unknown_in_list log_selector is used.
60 JH/10 Bug 3058: Ensure that a failing expansion in a router "set" option defers
61 the routing operation. Previously it would silently stop routing the
64 JH/11 Bug 3046: Fix queue-runs. Previously, the arrivel of a notification or
65 info-request event close in time to a scheduled run timer could result in
66 the latter being missed, and no further queue scheduled runs being
67 initiated. This ouwld be more likely on high-load systems.
69 JH/12 Refuse to accept a line "dot, LF" as end-of-DATA unless operating in
70 LF-only mode (as detected from the first header line). Previously we did
71 accept that in (normal) CRLF mode; this has been raised as a possible
72 attack scenario (under the name "smtp smuggling").
74 JH/13 Add an fdatasync call for the received message data file in spool, before
75 loggging reception and sending the SMTP ack. Previously we only flushed
76 the stdio buffer so there was still the possibility of a disk error.
78 JH/14 Bug 3061: Avoid a split log line when trying to rewrite a malformed
79 address. Previously, for the last address in a header line (commonly
80 there is only one) the terminating newline was part of the logged
83 JH/15 Bug 3061: Ensure a log line is written for a malformed address in a
84 header, when parsing for address-qualification. Previously one was only
85 written if there were rewrite rules.
87 JH/16 Two-phase queue runs are now reported in the daemon startup log line and
90 JH/17 Bug 3064: Fix combination of "-q<period> -R <recipients>". Introduction of
91 the multiple-queue-runners facility for 4.97 broke this, giving only a
92 one-time run of the queue.
94 JH/18 Bug 3068: Log a warning for use of deprecated syntax in query-style
97 JH/19 Fix TLS startup. When the last expansion done before the initiation of a
98 TLS session resulted in a forced-fail, a misleading error was logged for
99 the expansino of tls_certificates. This would affect the common case of
100 that option being set (main-section options) but not having any variable
101 parts. It could also potentially affect tls_privatekeys. The underlyding
102 coding errors go back to 4.90 but were only exposed in 4.97.
104 JH/20 Bug 3047: A recent (somewhere between 10.34 and 10.42) version of the
105 pcre2 library starting allocating 20kB rather than 112 bytes per match
106 call, which broke the 2GB total limitation on Exim's memory management
107 when a user had over 104207 messages stored and the appendfile
108 maildir_quota_directory_regex option is in use. Release the allocated
109 memory every thosand files to avoid this.
110 The same issue arises with the ACL regex condition, which is applied
111 to every line of a received message.
113 JH/21 Bug 3059: Fix crash in smtp transport. When running for a message for
114 which all recipients had been handled (itself an issue) a null-pointer
115 deref was done on trying to write a retry record. Fix that by counting
116 the outstanding recipients before trying to transmit the message.
117 The situation arose for a second MX try within a transport run, when the
118 first had perm-rejected a recipient (the only one for the connection, in
119 the case seen) during pipelining, and then closed the TCP connection.
120 The transport classified that as an I/O error, leaving the message
121 outstanding but having marked up the recipient as dealt-with. It then
122 tried another MX because of the I/O error. Fix this by converting the
123 message-level status to ok if there was a close but all recipients were
124 dealt with. Thanks to Wolfgand Breyha for debug runs.
126 JH/22 The ESMTP_LIMITS facility (RFC 9422) is promoted from experimental status
127 and is now controlled by the build-time option DISABLE_ESMTP_LIMITS.
129 JH/23 Bug 3066: Avoid leaking lookup database credentials to log.
131 JH/24 Bug 3081: Fix a delivery process crash. When the router "errors_to"
132 option specified a fixed address, later rewriting on that address would
133 trip on the configuration data being readonly. Instead of modifying
134 in-place, copy data. Found and fixed by Peter Benie.
136 JH/25 Bug 3079: Fix crash in dbmnz. When a key was present for zero-length
137 data a null pointer was followed. Find and testcase by Sebastian Bugge.
139 JH/26 Fix encoding for an AUTH parameter on a MAIL FROM command. Previously
140 decimal 127 chars were not encoded, and lowercase hex was used for
141 encoded values. Outstanding since at least 1999.
143 JH/27 Fix crash in logging. When a message with a large number of recipients
144 had been received, and logging of recipients is enabled, the buffer used
145 for logging could reach limit. A read using a null pointer would then
146 be done, resulting in a crash of the receiving process before an SMTP
147 ACK for the message was returned to the sending system. Duplicate
148 messages were created as a result.
149 Find and debug help by Mateusz Krawczyk
151 JH/28 Bug 3086: Fix exinext for ipv6. Change the format of keys in the retry
152 DB, wrapping transport record bare-ip "host names" and ipv6
153 "host addresses" in square-brackets. This makes the parsing that
154 exinext does more reliable.
156 JH/29 Bug 3087: Fix SRS encode. A zero-length quoted element in the local-part
159 JH/30 Bug 3029: Avoid feeding Resent-From: to DMARC.
161 JH/31 Bug 3027: For -bh / -bhc tests change to using the compressed form of
162 ipv6 addresses for the sender. Previously the uncompressed form was used,
163 and if used in textual form this would result in behavior difference
166 JH/32 Bug 3096: MAIL before HELO/EHLO, where required by hosts_require_helo, is
167 now classed as a protocol error and subject to smtp_max_synprot_errors.
169 JH/33 Bug 2994: A subdir dsearch lookup should permit a directory name that starts
170 ".." and has following characters.
172 JH/34 Fix delivery ordering for 2-phase queue run combined with
179 JH/01 The hosts_connection_nolog main option now also controls "no MAIL in
180 SMTP connection" log lines.
182 JH/02 Option default value updates:
183 - queue_fast_ramp (main) true (was false)
184 - remote_max_parallel (main) 4 (was 2)
186 JH/03 Cache static regex pattern compilations, for use by ACLs.
188 JH/04 Bug 2903: avoid exit on an attempt to rewrite a malformed address.
189 Make the rewrite never match and keep the logging. Trust the
190 admin to be using verify=header-syntax (to actually reject the message).
192 JH/05 Follow symlinks for placing a watch on TLS creds files. This means
193 (under Linux) we watch the dir containing the final file; previously
194 it would be the dir with the first symlink. We still do not monitor
197 JH/06 Check for bad chars in rDNS for sender_host_name. The OpenBSD (at least)
198 dn_expand() is happy to pass them through.
200 JH/07 OpenSSL Fix auto-reload of changed server OCSP proof. Previously, if
201 the file with the proof had an unchanged name, the new proof(s) were
202 loaded on top of the old ones (and nover used; the old ones were stapled).
204 JH/08 Bug 2915: Fix use-after-free for $regex<n> variables. Previously when
205 more than one message arrived in a single connection a reference from
206 the earlier message could be re-used. Often a sigsegv resulted.
207 These variables were introduced in Exim 4.87.
208 Debug help from Graeme Fowler.
210 JH/09 Fix ${filter } for conditions that modify $value. Previously the
211 modified version would be used in construction the result, and a memory
214 JH/10 GnuTLS: fix for (IOT?) clients offering no TLS extensions at all.
215 Find and fix by Jasen Betts.
217 JH/11 OpenSSL: fix for ancient clients needing TLS support for versions earlier
218 than TLSv1,2, Previously, more-recent versions of OpenSSL were permitting
219 the systemwide configuration to override the Exim config.
221 HS/01 Bug 2728: Introduce EDITME option "DMARC_API" to work around incompatible
222 API changes in libopendmarc.
224 JH/12 Bug 2930: Fix daemon startup. When started from any process apart from
225 pid 1, in the normal "background daemon" mode, having to drop process-
226 group leadership also lost track of needing to create listener sockets.
228 JH/13 Bug 2929: Fix using $recipients after ${run...}. A change made for 4.96
229 resulted in the variable appearing empty. Find and fix by Ruben Jenster.
231 JH/14 Bug 2933: Fix regex substring match variables for null matches. Since 4.96
232 a capture group which obtained no text (eg. "(abc)*" matching zero
233 occurrences) could cause a segfault if the corresponding $<n> was
236 JH/15 Fix argument parsing for ${run } expansion. Previously, when an argument
237 included a close-brace character (eg. it itself used an expansion) an
240 JH/16 Move running the smtp connect ACL to before, for TLS-on-connect ports,
241 starting TLS. Previously it was after, meaning that attackers on such
242 ports had to be screened using the host_reject_connection main config
243 option. The new sequence aligns better with the STARTTLS behaviour, and
244 permits defences against crypto-processing load attacks, even though it
245 is strictly an incompatible change.
246 Also, avoid sending any SMTP fail response for either the connect ACL
247 or host_reject_connection, for TLS-on-connect ports.
249 JH/17 Permit the ACL "encrypted" condition to be used in a HELO/EHLO ACL,
250 Previously this was not permitted, but it makes reasonable sense.
251 While there, restore a restriction on using it from a connect ACL; given
252 the change JH/16 it could only return false (and before 4.91 was not
255 JH/18 Fix a fencepost error in logging. Previously (since 4.92) when a log line
256 was exactly sized compared to the log buffer, a crash occurred with the
257 misleading message "bad memory reference; pool not found".
258 Found and traced by Jasen Betts.
260 JH/19 Bug 2911: Fix a recursion in DNS lookups. Previously, if the main option
261 dns_again_means_nonexist included an element causing a DNS lookup which
262 itself returned DNS_AGAIN, unbounded recursion occurred. Possible results
263 included (though probably not limited to) a process crash from stack
264 memory limit, or from excessive open files. Replace this with a paniclog
265 whine (as this is likely a configuration error), and returning
268 JH/20 Bug 2954: (OpenSSL) Fix setting of explicit EC curve/group. Previously
269 this always failed, probably leading to the usual downgrade to in-clear
272 JH/21 Fix TLSA lookups. Previously dns_again_means_nonexist would affect
273 SERVFAIL results, which breaks the downgrade resistance of DANE. Change
274 to not checking that list for these lookups.
276 JH/22 Bug 2434: Add connection-elapsed "D=" element to more connection
279 JH/23 Fix crash in string expansions. Previously, if an empty variable was
280 immediately followed by an expansion operator, a null-indirection read
281 was done, killing the process.
283 JH/24 Bug 2997: When built with EXPERIMENTAL_DSN_INFO, bounce messages can
284 include an SMTP response string which is longer than that supported
285 by the delivering transport. Alleviate by wrapping such lines before
288 JH/25 Bug 2827: Restrict size of References: header in bounce messages to 998
289 chars (RFC limit). Previously a limit of 12 items was made, which with
290 a not-impossible References: in the message being bounced could still
291 be over-large and get stopped in the transport.
293 JH/26 For a ${readsocket } in TLS mode, send a TLS Close Alert before the TCP
294 close. Previously a bare socket close was done.
296 JH/27 Fix ${srs_encode ..}. Previously it would give a bad result for one day
299 JH/28 Bug 2996: Fix a crash in the smtp transport. When finding that the
300 message being considered for delivery was already being handled by
301 another process, and having an SMTP connection already open, the function
302 to close it tried to use an uninitialized variable. This would afftect
303 high-volume sites more, especially when running mailing-list-style loads.
304 Pollution of logs was the major effect, as the other process delivered
305 the message. Found and partly investigated by Graeme Fowler.
307 JH/29 Change format of the internal ID used for message identification. The old
308 version only supported 31 bits for a PID element; the new 64 (on systems
309 which can use Base-62 encoding, which is all currently supported ones
310 but not Darwin (MacOS) or Cygwin, which have case-insensitive filesystems
311 and must use Base-36). The new ID is 23 characters rather than 16, and is
312 visible in various places - notably logs, message headers, and spool file
313 names. Various of the ancillary utilities also have to know the format.
314 As well as the expanded PID portion, the sub-second part of the time
315 recorded in the ID is expanded to support finer precision. Theoretically
316 this permits a receive rate from a single comms channel of better than the
318 The major timestamp part of the ID is not changed; at 6 characters it is
319 usable until about year 3700.
320 Updating from previously releases is fully supported: old-format spool
321 files are still usable, and the utilities support both formats. New
322 message will use the new format. The one hints-DB file type which uses
323 message-IDs (the transport wait- DB) will be discarded if an old-format ID
324 is seen; new ones will be built with only new-format IDs.
325 Optionally, a utility can be used to convert spool files from old to new,
326 but this is only an efficiency measure not a requirement for operation
327 Downgrading from new to old requires running a provided utility, having
328 first stopped all operations. This will convert any spool files from new
329 back to old (losing time-precision and PID information) and remove any
330 wait- hints databases.
332 JH/30 Bug 3006: Fix handling of JSON strings having embedded commas. Previously
333 we treated them as item separators when parsing for a list item, but they
334 need to be protected by the doublequotes. While there, add handling for
337 JH/31 Bug 2998: Fix ${utf8clean:...} to disallow UTF-16 surrogate codepoints.
338 Found and fixed by Jasen Betts. No testcase for this as my usual text
339 editor insists on emitting only valid UTF-8.
341 JH/32 Fix "tls_dhparam = none" under GnuTLS. At least with 3.7.9 this gave
342 a null-indirection SIGSEGV for the receive process.
344 JH/33 Fix free for live variable $value created by a ${run ...} expansion during
345 -bh use. Internal checking would spot this and take a panic.
347 JH/34 Bug 3013: Fix use of $recipients within arguments for ${run...}.
348 In 4.96 this would expand to empty.
350 JH/35 Bug 3014: GnuTLS: fix expiry date for an auto-generated server
351 certificate. Find and fix by Andreas Metzler.
353 JH/36 Add ARC info to DMARC hostory records.
355 JH/37 Bug 3016: Avoid sending DSN when message was accepted under fakereject
356 or fakedefer. Previously the sender could discover that the message
357 had in fact been accepted.
359 JH/38 Taint-track intermediate values from the peer in multi-stage authentation
360 sequences. Previously the input was not noted as being tainted; notably
361 this resulted in behaviour of LOGIN vs. PLAIN being inconsistent under
362 bad coding of authenticators.
364 JH/39 Bug 3023: Fix crash induced by some combinations of zero-length strings
365 and ${tr...}. Found and diagnosed by Heiko Schlichting.
367 JH/40 Bug 2999: Fix a possible OOB write in the external authenticator, which
368 could be triggered by externally-supplied input. Found by Trend Micro.
371 JH/41 Bug 3000: Fix a possible OOB write in the SPA authenticator, which could
372 be triggered by externally-controlled input. Found by Trend Micro.
375 JH/42 Bug 3001: Fix a possible OOB read in the SPA authenticator, which could
376 be triggered by externally-controlled input. Found by Trend Micro.
379 JH/43 Bug 2903: avoid exit on an attempt to rewrite a malformed address.
380 Make the rewrite never match and keep the logging. Trust the
381 admin to be using verify=header-syntax (to actually reject the message).
383 JH/44 Bug 3033: Harden dnsdb lookups against crafted DNS responses.
386 HS/02 Fix string_is_ip_address() CVE-2023-42117 (Bug 3031)
392 JH/01 Move the wait-for-next-tick (needed for unique message IDs) from
393 after reception to before a subsequent reception. This should
394 mean slightly faster delivery, and also confirmation of reception
397 JH/02 Move from using the pcre library to pcre2. The former is no longer
398 being developed or supported (by the original developer).
400 JH/03 Constification work in the filters module required a major version
401 bump for the local-scan API. Specifically, the "headers_charset"
402 global which is visible via the API is now const and may therefore
403 not be modified by local-scan code.
405 JH/04 Fix ClamAV TCP use under FreeBSD. Previously the OS-specific shim for
406 sendfile() didi not account for the way the ClamAV driver code called it.
408 JH/05 Bug 2819: speed up command-line messages being read in. Previously a
409 time check was being done for every character; replace that with one
412 JH/06 Bug 2815: Fix ALPN sent by server under OpenSSL. Previously the string
413 sent was prefixed with a length byte.
415 JH/07 Change the SMTP feature name for pipelining connect to be compliant with
416 RFC 5321. Previously Dovecot (at least) would log errors during
419 JH/08 Remove stripping of the binaries from the FreeBSD build. This was added
420 in 4.61 without a reason logged. Binaries will be bigger, which might
421 matter on diskspace-constrained systems, but debug is easier.
423 JH/09 Fix macro-definition during "-be" expansion testing. The move to
424 write-protected store for macros had not accounted for these runtime
425 additions; fix by removing this protection for "-be" mode.
427 JH/10 Convert all uses of select() to poll(). FreeBSD 12.2 was found to be
428 handing out large-numbered file descriptors, violating the usual Unix
429 assumption (and required by Posix) that the lowest possible number will be
430 allocated by the kernel when a new one is needed. In the daemon, and any
431 child procesees, values higher than 1024 (being bigger than FD_SETSIZE)
432 are not useable for FD_SET() [and hence select()] and overwrite the stack.
433 Assorted crashes happen.
435 JH/11 Fix use of $sender_host_name in daemon process. When used in certain
436 main-section options or in a connect ACL, the value from the first ever
437 connection was never replaced for subsequent connections. Found by
440 JH/12 Bug 2838: Fix for i32lp64 hard-align platforms. Found for SPARC Linux,
441 though only once PCRE2 was introduced: the memory accounting used under
442 debug offset allocations by an int, giving a hard trap in early startup.
443 Change to using a size_t. Debug and fix by John Paul Adrian Glaubitz.
445 JH/13 Bug 2845: Fix handling of tls_require_ciphers for OpenSSL when a value
446 with underbars is given. The write-protection of configuration introduced
447 in 4.95 trapped when normalisation was applied to an option not needing
450 JH/14 Bug 1895: TLS: Deprecate RFC 5114 Diffie-Hellman parameters.
452 JH/15 Fix a resource leak in *BSD. An off-by-one error resulted in the daemon
453 failing to close the certificates directory, every hour or any time it
456 JH/16 Debugging initiated by an ACL control now continues through into routing
457 and transport processes. Previously debugging stopped any time Exim
458 re-execs, or for processing a queued message.
460 JH/17 The "expand" debug selector now gives more detail, specifically on the
461 result of expansion operators and items.
463 JH/18 Bug 2751: Fix include_directory in redirect routers. Previously a
464 bad comparison between the option value and the name of the file to
465 be included was done, and a mismatch was wrongly identified.
466 4.88 to 4.95 are affected.
468 JH/19 Support for Berkeley DB versions 1 and 2 is withdrawn.
470 JH/20 When built with NDBM for hints DB's check for nonexistence of a name
471 supplied as the db file-pair basename. Previously, if a directory
472 path was given, for example via the autoreply "once" option, the DB
473 file.pag and file.dir files would be created in that directory's
476 JH/21 Remove the "allow_insecure_tainted_data" main config option and the
477 "taint" log_selector. These were previously deprecated.
479 JH/22 Fix static address-list lookups to properly return the matched item.
480 Previously only the domain part was returned.
482 JH/23 Bug 2864: FreeBSD: fix transport hang after 4xx/5xx response. Previously
483 the call into OpenSSL to send a TLS Close was being repeated; this
484 resulted in the library waiting for the peer's Close. If that was never
485 sent we waited forever. Fix by tracking send calls.
487 JH/24 The ${run} expansion item now expands its command string elements after
488 splitting. Previously it was before; the new ordering makes handling
489 zero-length arguments simpler. The old ordering can be obtained by
490 appending a new option "preexpand", after a comma, to the "run".
492 JH/25 Taint-check exec arguments for transport-initiated external processes.
493 Previously, tainted values could be used. This affects "pipe", "lmtp" and
494 "queryprogram" transport, transport-filter, and ETRN commands.
495 The ${run} expansion is also affected: in "preexpand" mode no part of
496 the command line may be tainted, in default mode the executable name
499 JH/26 Fix CHUNKING on a continued-transport. Previously the usabliility of
500 the the facility was not passed across execs, and only the first message
501 passed over a connection could use BDAT; any further ones using DATA.
503 JH/27 Support the PIPECONNECT facility in the smtp transport when the helo_data
504 uses $sending_ip_address and an interface is specified.
505 Previously any use of the local address in the EHLO name disabled
506 PIPECONNECT, the common case being to use the rDNS of it.
508 JH/28 OpenSSL: fix transport-required OCSP stapling verification under session
509 resumption. Previously verify failed because no certificate status is
510 passed on the wire for the restarted session. Fix by using the recorded
511 ocsp status of the stored session for the new connection.
513 JH/29 TLS resumption: the key for session lookup in the client now includes
514 more info that a server could potentially use in configuring a TLS
515 session, avoiding oferring mismatching sessions to such a server.
516 Previously only the server IP was used.
518 JH/30 Fix string_copyn() for limit greater than actual string length.
519 Previously the copied amount was the limit, which could result in a
520 overlapping memcpy for newly allocated destination soon after a
521 source string shorter than the limit. Found/investigated by KM.
523 JH/31 Bug 2886: GnuTLS: Do not free the cached creds on transport connection
524 close; it may be needed for a subsequent connection. This caused a
525 SEGV on primary-MX defer. Found/investigated by Gedalya & Andreas.
527 JH/32 Fix CHUNKING for a second message on a connection when the first was
528 rejected. Previously we did not reset the chunking-offered state, and
529 erroneously rejected the BDAT command. Investigation help from
532 JH/33 Fis ${srs_encode ...} to handle an empty sender address, now returning
533 an empty address. Previously the expansion returned an error.
535 HS/01 Bug 2855: Handle a v4mapped sender address given us by a frontending
536 proxy. Previously these were misparsed, leading to paniclog entries.
542 JH/01 Bug 1329: Fix format of Maildir-format filenames to match other mail-
543 related applications. Previously an "H" was used where available info
544 says that "M" should be, so change to match.
546 JH/02 Bug 2587: Fix pam expansion condition. Tainted values are commonly used
547 as arguments, so an implementation trying to copy these into a local
548 buffer was taking a taint-enforcement trap. Fix by using dynamically
549 created buffers. Similar fix for radius expansion condition.
551 JH/03 Bug 2586: Fix listcount expansion operator. Using tainted arguments is
552 reasonable, eg. to count headers. Fix by using dynamically created
553 buffers rather than a local. Do similar fixes for ACL actions "dcc",
554 "log_reject_target", "malware" and "spam"; the arguments are expanded
555 so could be handling tainted values.
557 JH/04 Bug 2590: Fix -bi (newaliases). A previous code rearrangement had
558 broken the (no-op) support for this sendmail command. Restore it
559 to doing nothing, silently, and returning good status.
561 JH/05 Bug 2593: Fix "vacation" in Exim filter. Previously, when a "once"
562 record path was given (or the default used) without a leading directory
563 path, an error occurred on trying to open it. Use the transport's working
566 JH/06 Bug 2594: Change the name used for certificate name checks in the smtp
567 transport. Previously it was the name on the DNS A-record; use instead
568 the head of the CNAME chain leading there (if there is one). This seems
569 to align better with RFC 6125.
571 JH/07 Bug 2597: Fix a resource leak. Using a lookup in obtaining a value for
572 smtp_accept_max_per_host allocated resources which were not released
573 when the limit was exceeded. This eventually crashed the daemon. Fix
574 by adding a release action in that path.
576 JH/08 Bug 2598: Fix verify ACL condition. The options for the condition are
577 expanded; previously using tainted values was rejected. Fix by using
578 dynamically-created buffers.
580 JH/09 Relax restrictions on ACL verify condition needing access to message
581 headers. Previously they were only permitted in data and non-smtp ACLs;
582 permit also mime, dkim, prdr quit and notquit. Applies to header-syntax,
583 not_blind, header_sender and header_names_ascii verification.
585 JH/10 Bug 2603: Fix coding of string copying to only evaluate arguments once.
586 Previously a macro used one argument twice; when called with the
587 argument as an expression having side-effects, incorrect operation
588 resulted. Use an inlineable function.
590 JH/11 Bug 2604: Fix request to cutthrough-deliver when a connection is already
591 held open for a verify callout. Previously this wan not accounted for
592 and a corrupt onward SMTP conversation resulted.
594 JH/12 Bug 2607: Fix the ${srs_encode } expansion to handle quoted local_parts.
595 Previously they were embedded naively in the constructed address; when
596 needed, strip the quoting and quote the entire local_part.
597 Also make the inbound_srs expansion condition handle quoting.
599 JH/13 Fix dsearch "subdir" filter to ignore ".". Previously only ".." was
600 excluded, not matching the documentation.
602 JH/14 Bug 2606: Fix a segfault in sqlite lookups. When no, or a bad, filename
603 was given for the sqlite_dbfile a trap resulted.
605 JH/15 Bug 2620: Fix "spam" ACL condition. Previously, tainted values for the
606 "name" argument resulted in a trap. There is no reason to disallow such;
607 this was a coding error.
609 JH/16 Bug 2615: Fix pause during message reception, on systems that have been
610 suspended/resumed. The Linux CLOCK_MONOTONIC does not account for time
611 spent suspended, ignoring the POSIX definition. Previously we assumed
612 it did and a constant offset from real time could be used as a correction.
613 Change to using the same clock source for the start-of-message and the
614 post-message next-tick-wait. Also change to using CLOCK_BOOTTIME if it
615 exists, just to get a clock slightly more aligned to reality.
617 JH/17 Bug 2295: Fix DKIM signing to always semicolon-terminate. Although the
618 RFC says it is optional some validators care. The missing char was not
619 intended but triggered by a line-wrap alignment. Discovery and fix by
620 Guillaume Outters, hacked on by JH.
622 JH/18 Bug 2617: Fix a taint trap in parse_fix_phrase(). Previously when the
623 name being quoted was tainted a trap would be taken. Fix by using
624 dynamically created buffers. The routine could have been called by a
625 rewrite with the "h" flag, by using the "-F" command-line option, or
626 by using a "name=" option on a control=submission ACL modifier.
628 JH/19 SPF: change the Authentication-Results expansion component to give
629 smtp.helo when the sender domain is empty. Previously it gave
632 JH/20 Bug 2631: ACL dnslist conditions now ignore and log any lookups returns
633 not in 127.0.0.0/8 to help in spotting list domains taken over by a
634 domain-parking registrar.
636 JH/21 Bug 2630: Fix eol-replacement string for the ${readsocket } expansion.
637 Previously when a whitespace character was specified it was not inserted
638 after removing the newline.
640 JH/22 Bug 2265: Force SNI usage for smtp transport DANE'd connections, to be
641 the domain part of the recipient address. This overrides any tls_sni
642 option set, which was previously used.
644 JH/23 Logging: with the +tls_sni log_selector, do not wrap the received SNI
647 JH/24 Bug 2634: Fix a taint trap seen on NetBSD: the testing coded for
648 is_tainted() had an off-by-one error in the overenthusiastic direction.
649 Find and fix by Gavan. Although NetBSD is not a supported platform for
650 4.94 this bug could affect other platforms.
652 PP/01 Fix default prime selection to be consistent.
653 One path used ike23 still, instead of exim.dev.20160529.3; now both
654 execution flows will use the same DH primes (currently
655 exim.dev.20160529.3).
657 JH/25 OpenSSL: Fix back-compatibility behaviour surrounding tls_certificates
658 option in smtp transport, to match the documentation. Previously
659 verification was not being done in some cases where it should have been.
661 JH/26 Bug 2646: fix a memory usage issue in ldap lookups. Previously, when more
662 than one server was defined and depending on the platform memory layout
663 details, an internal consistency trap could be hit while walking the list
666 JH/27 Bug 2648: fix the passing of an authenticator public-name through spool
667 files. The value is used by the authresults expansion item. Previously
668 if this was used in a router or transport, a crash could result.
670 JH/28 Fix spurious logging of select error. Some platforms, notably FreeBSD,
671 have a sufficient incidence of EINTR returns from select that an
672 interaction with other operations done by the main daemon loop exposed
673 a bug in the error-handling. This was benign apart from the log
676 JH/29 Bug 2675: add outgoing-interface I= element to deferred "==" log lines,
677 for consistency with delivered "=>" and failed "**" lines. While we're
678 there, handle PRX and TFO.
680 JH/30 Bug 2677: fix matching of long addresses. Since 4.93 a limit of 256 was
681 applied. This resulted, if any header-line rewrite rules were configured,
682 in a panic-log triggerable by sending a message with a long address in
683 a header. Fix by increasing the arbitrary limit to larger than a single
684 (dewrapped) 5322 header line maximum size.
686 JH/31 The ESMTP option name advertised for the SUPPORT_EARLY_PIPE build option
687 is changed from X_PIPE_CONNECT to PIPE_CONNECT. This is in line with
688 RFC 6648 which deprecates X- options in protocols as a general practice.
689 Changeover between the implementations is handled by the mechanisms
692 JH/32 Bug 2599: fix delay of delivery to a local address where there is also
693 a remote which uses callout/hold. Previously the local was queued.
695 JH/33 Fix a taint trap in the ${listextract } expansion when the source data
698 JH/34 Fix the placement of a multiple-message delivery marker in the delivery
699 log line. The asterisk is now consistently appended to the remote IP
700 (and port, if given), and will also be provided on defer and fail log
701 lines. Previously it could be placed on the local IP if that was being
702 logged, and was only provided on delivery lines.
704 JH/35 Bug 2343: Harden exim_tidydb against corrupt wait- files.
706 JH/36 Bug 2687: Fix interpretation of multiple ^ chars in a plaintext
707 authenticator client_send option. Previously the next char, after a pair
708 was collapsed, was taken verbatim (so ^^^foo became ^^foo; ^^^^foo became
709 ^^\x00foo). Fixed to get ^\x00foo and ^^foo respectively to match the
710 documentation. There is still no way to get a leading ^ immediately
711 after a NUL (ie. for the password of a PLAIN method authenticator.
713 JH/37 Enforce the expected size, for fixed-size records read from hints-DB
714 files. For bad sizes read, delete the record and whine to paniclog.
716 JH/38 When logging an AUTH failure, as server, do not include sensitive
717 information. Previously, the credentials would be included if given
718 as part of the AUTH command line and an ACL denied authentication.
720 JH/39 Bug 2691: fix $local_part_data. When the matching list element
721 referred to a file, bad data was returned. This likely also affected
724 JH/40 The gsasl authenticator now supports caching of the salted password
725 generated by the client-side implementation. This required the addition
726 of a new variable: $auth4.
728 JH/41 Fix daemon SIGHUP on FreeBSD. Previously, a named socket for IPC was
729 left undeleted; the attempt to re-create it then failed - resulting in
730 the usual "SIGHUP tp have daemon reload configuration" to not work.
731 This affected any platform not supporting "abstract" Unix-domain
732 sockets (i.e. not Linux).
734 JH/42 Bug 2693: Harden against a peer which reneges on a 452 "too many
735 recipients" response to RCPT in a later response, with a 250. The
736 previous coding assumed this would not happen, and under PIPELINING
737 would result in both lost and duplicate recipients for a message.
739 JH/43 Bug 2694: Fix weighted distribution of work to multiple spamd servers.
740 Previously the weighting was incorrectly applied. Similar fix for socks
741 proxies. Found and fixed by Heiko Schlichting.
743 JH/44 Bug 2701: Fix list-expansion of dns_ipv4_lookup. Previously, it did
744 not handle sub-lists included using the +namedlist syntax. While
745 investigating, the same found for dns_trust_aa, dns_again_means_nonexist,
746 dnssec_require_domains, dnssec_request_domains, srv_fail_domains,
749 JH/45 Use a (new) separate store pool-pair for DKIM verify working data.
750 Previously the permanent pool was used, so the store could not be freed.
751 This meant a connection with many messages would use continually-growing
754 JH/46 Use an exponentially-increasing block size when malloc'ing store. Do it
755 per-pool so as not to waste too much space. Previously a constant size
756 was used which resulted in O(n^2) behaviour; now we get O(n log n) making
757 DOS attacks harder. The cost is wasted memory use in the larger blocks.
759 JH/47 Use explicit alloc/free for DNS lookup workspace. This permits using the
760 same space repeatedly, and a smaller process footprint.
762 JH/48 Use a less bogus-looking filename for a temporary used for DH-parameters
763 for GnuTLS. Previously the name started "%s" which, while not a bug,
764 looked as if if might be one.
766 JH/49 Bug 2710: when using SOCKS for additional messages after the first (a
767 "continued connection") make the $proxy_* variables available. Previously
768 the information was not passed across the exec() call for subsequent
769 transport executions. This also mean that the log lines for the
770 messages can show the proxy information.
772 JH/50 Bug 2672: QT elements in log lines, unless disabled, now exclude the
773 receive time. With modern systems the difference is significant.
774 The historical behaviour can be restored by disabling (a new) log_selector
775 "queue_time_exclusive".
777 JH/51 Taint-check ACL line. Previously, only filenames (for out-of-line ACL
778 content) were specifically tested for. Now, also cover expansions
779 resulting in ACL names and inline ACL content.
781 JH/52 Fix ${ip6norm:} operator. Previously, any trailing line text was dropped,
782 making it unusable in complex expressions.
784 JH/53 Bug 2743: fix immediate-delivery via named queue. Previously this would
785 fail with a taint-check on the spoolfile name, and leave the message
788 HS/01 Enforce absolute PID file path name.
790 HS/02 Handle SIGINT as we handle SIGTERM: terminate the Exim process.
792 PP/01 Add a too-many-bad-recipients guard to the default config's RCPT ACL.
794 PP/02 Bug 2643: Correct TLS DH constants.
795 A missing NUL termination in our code-generation tool had led to some
796 incorrect Diffie-Hellman constants in the Exim source.
797 Reported by kylon94, code-gen tool fix by Simon Arlott.
799 PP/03 Impose security length checks on various command-line options.
800 Fixes CVE-2020-SPRSS reported by Qualys.
802 PP/04 Fix Linux security issue CVE-2020-SLCWD and guard against PATH_MAX
803 better. Reported by Qualys.
805 PP/05 Fix security issue CVE-2020-PFPSN and guard against cmdline invoker
806 providing a particularly obnoxious sender full name.
809 PP/06 Fix CVE-2020-28016 (PFPZA): Heap out-of-bounds write in parse_fix_phrase()
811 PP/07 Refuse to allocate too little memory, block negative/zero allocations.
814 PP/08 Change default for recipients_max from unlimited to 50,000.
816 PP/09 Fix security issue with too many recipients on a message (to remove a
817 known security problem if someone does set recipients_max to unlimited,
818 or if local additions add to the recipient list).
819 Fixes CVE-2020-RCPTL reported by Qualys.
821 PP/10 Fix security issue in SMTP verb option parsing
822 Fixes CVE-2020-EXOPT reported by Qualys.
824 PP/11 Fix security issue in BDAT state confusion.
825 Ensure we reset known-good where we know we need to not be reading BDAT
826 data, as a general case fix, and move the places where we switch to BDAT
827 mode until after various protocol state checks.
828 Fixes CVE-2020-BDATA reported by Qualys.
830 HS/03 Die on "/../" in msglog file names
832 QS/01 Creation of (database) files in $spool_dir: only uid=0 or the uid of
833 the Exim runtime user are allowed to create files.
835 QS/02 PID file creation/deletion: only possible if uid=0 or uid is the Exim
838 QS/03 When reading the output from interpreted forward files we do not
839 pass the pipe between the parent and the interpreting process to
840 executed child processes (if any).
842 QS/04 Always die if requested from internal logging, even is logging is
845 JH/54 DMARC: recent versions of the OpenDMARC library appear to have broken
846 the API; compilation noo longer completes with DMARC support included.
847 This affects 1.4.1-1 on Fedora 33 (1.3.2-3 is functional); and has
848 been reported on other platforms.
850 JH/55 TLS: as server, reject connections with ALPN indicating non-smtp use.
852 JH/56 Make the majority of info read from config files readonly, for defence-in-
853 depth against exploits. Suggestion by Qualys.
854 Not supported on Solaris 10.
856 JH/57 Fix control=fakreject for a custom message containing tainted data.
857 Previously this resulted in a log complaint, due to a re-expansion present
858 since fakereject was originally introduced.
860 JH/58 GnuTLS: Fix certextract expansion. If a second modifier after a tag
861 modifier was given, a loop resulted.
863 JH/59 DKIM: Fix small-message verification under TLS with chunking. If a
864 pipelined SMTP command followed the BDAT LAST then it would be
865 incorrectly treated as part of the message body, causing a verification
868 JH/60 Bug 2805: Fix logging of domain-literals in Message_ID: headers. They
869 require looser validation rules than those for 821-level addresses,
870 which only permit IP addresses.
876 JH/01 Avoid costly startup code when not strictly needed. This reduces time
877 for some exim process initialisations. It does mean that the logging
878 of TLS configuration problems is only done for the daemon startup.
880 JH/02 Early-pipelining support code is now included unless disabled in Makefile.
882 JH/03 DKIM verification defaults no long accept sha1 hashes, to conform to
883 RFC 8301. They can still be enabled, using the dkim_verify_hashes main
886 JH/04 Support CHUNKING from an smtp transport using a transport_filter, when
887 DKIM signing is being done. Previously a transport_filter would always
888 disable CHUNKING, falling back to traditional DATA.
890 JH/05 Regard command-line recipients as tainted.
892 JH/06 Bug 340: Remove the daemon pid file on exit, when due to SIGTERM.
894 JH/07 Bug 2489: Fix crash in the "pam" expansion condition. It seems that the
895 PAM library frees one of the arguments given to it, despite the
896 documentation. Therefore a plain malloc must be used.
898 JH/08 Bug 2491: Use tainted buffers for the transport smtp context. Previously
899 on-stack buffers were used, resulting in a taint trap when DSN information
900 copied from a received message was written into the buffer.
902 JH/09 Bug 2493: Harden ARC verify against Outlook, whick has been seen to mix
903 the ordering of its ARC headers. This caused a crash.
905 JH/10 Bug 2492: Use tainted memory for retry record when needed. Previously when
906 a new record was being constructed with information from the peer, a trap
909 JH/11 Bug 2494: Unset the default for dmarc_tld_file. Previously a naiive
910 installation would get error messages from DMARC verify, when it hit the
911 nonexistent file indicated by the default. Distros wanting DMARC enabled
912 should both provide the file and set the option.
913 Also enforce no DMARC verification for command-line sourced messages.
915 JH/12 Fix an uninitialised flag in early-pipelining. Previously connections
916 could, depending on the platform, hang at the STARTTLS response.
918 JH/13 Bug 2498: Reset a counter used for ARC verify before handling another
919 message on a connection. Previously if one message had ARC headers and
920 the following one did not, a crash could result when adding an
921 Authentication-Results: header.
923 JH/14 Bug 2500: Rewind some of the common-coding in string handling between the
924 Exim main code and Exim-related utities. The introduction of taint
925 tracking also did many adjustments to string handling. Since then, eximon
926 frequently terminated with an assert failure.
928 JH/15 When PIPELINING, synch after every hundred or so RCPT commands sent and
929 check for 452 responses. This slightly helps the inefficieny of doing
930 a large alias-expansion into a recipient-limited target. The max_rcpt
931 transport option still applies (and at the current default, will override
932 the new feature). The check is done for either cause of synch, and forces
933 a fast-retry of all 452'd recipients using a new MAIL FROM on the same
934 connection. The new facility is not tunable at this time.
936 JH/16 Fix the variables set by the gsasl authenticator. Previously a pointer to
937 library live data was being used, so the results became garbage. Make
938 copies while it is still usable.
940 JH/17 Logging: when the deliver_time selector ise set, include the DT= field
941 on delivery deferred (==) and failed (**) lines (if a delivery was
942 attemtped). Previously it was only on completion (=>) lines.
944 JH/18 Authentication: the gsasl driver not provides the $authN variables in time
945 for the expansion of the server_scram_iter and server_scram_salt options.
947 WB/01 SPF: DNS lookups for the obsolete SPF RR type done by the libspf2 library
948 are now specifically given a NO_DATA response without hitting the system
949 resolver. The library goes on to do the now-standard TXT lookup.
950 Use of dnsdb lookups is not affected.
952 JH/19 Bug 2507: Modules: on handling a dynamic-module (lookups) open failure,
953 only retrieve the errormessage once. Previously two calls to dlerror()
954 were used, and the second one (for mainlog/paniclog) retrieved null
957 JH/20 Taint checking: disallow use of tainted data for
958 - the appendfile transport file and directory options
959 - the pipe transport command
960 - the autoreply transport file, log and once options
961 - file names used by the redirect router (including filter files)
963 - paths used by single-key lookups
964 Previously this was permitted.
966 JH/21 Bug 2501: Fix init call in the heimdal authenticator. Previously it
967 adjusted the size of a major service buffer; this failed because the
968 buffer was in use at the time. Change to a compile-time increase in the
969 buffer size, when this authenticator is compiled into exim.
971 JH/22 Taint-checking: move to safe-mode taint checking on all platforms. The
972 previous fast-mode was untenable in the face of glibs using mmap to
973 support larger malloc requests.
975 PP/01 Update the openssl_options possible values through OpenSSL 1.1.1c.
976 New values supported, if defined on system where compiled:
977 allow_no_dhe_kex, cryptopro_tlsext_bug, enable_middlebox_compat,
978 no_anti_replay, no_encrypt_then_mac, prioritize_chacha, tlsext_padding
980 JH/23 Performance improvement in the initial phase of a two-pass queue run. By
981 running a limited number of proceses in parallel, a benefit is gained. The
982 amount varies with the platform hardware and load. The use of the option
983 queue_run_in_order means we cannot do this, as ordering becomes
986 JH/24 Bug 2524: fix the cyrus_sasl auth driver gssapi usage. A previous fix
987 had introduced a string-copy (for ensuring NUL-termination) which was not
988 appropriate for that case, which can include embedded NUL bytes in the
989 block of data. Investigation showed the copy to actually be needless, the
990 data being length-specified.
992 JH/25 Fix use of concurrent TLS connections under GnuTLS. When a callout was
993 done during a receiving connection, and both used TLS, global info was
994 used rather than per-connection info for tracking the state of data
995 queued for transmission. This could result in a connection hang.
997 JH/26 Fix use of the SIZE parameter on MAIL commands, on continued connections.
998 Previously, when delivering serveral messages down a single connection
999 only the first would provide a SIZE. This was due to the size information
1000 not being properly tracked.
1002 JH/27 Bug 2530: When operating in a timezone with sub-minute offset, such as
1003 TAI (at 37 seconds currently), pretend to be in UTC for time-related
1004 expansion and logging. Previously, spurious values such as a future
1005 minute could be seen.
1007 JH/28 Bug 2533: Fix expansion of ${tr } item. When called in some situations
1008 it could crash from a null-deref. This could also affect the
1009 ${addresses: } operator and ${readsock } item.
1011 JH/29 Bug 2537: Fix $mime_part_count. When a single connection had a non-mime
1012 message following a mime one, the variable was not reset.
1014 JH/30 When an pipelined-connect fails at the first response, assume incorrect
1015 cached capability (perhaps the peer reneged?) and immediately retry in
1018 JH/31 Fix spurious detection of timeout while writing to transport filter.
1020 JH/32 Bug 2541: Fix segfault on bad cmdline -f (sender) argument. Previously
1021 an attempt to copy the string was made before checking it.
1023 JH/33 Fix the dsearch lookup to return an untainted result. Previously the
1024 taint of the lookup key was maintained; we now regard the presence in the
1025 filesystem as sufficient validation.
1027 JH/34 Fix the readsocket expansion to not segfault when an empty "options"
1028 argument is supplied.
1030 JH/35 The dsearch lookup now requires that the directory is an absolute path.
1031 Previously this was not checked, and nonempty relative paths made an
1032 access under Exim's current working directory.
1034 JH/36 Bug 2554: Fix msg:defer event for the hosts_max_try_hardlimit case.
1035 Previously no event was raised.
1037 JH/37 Bug 2552: Fix the check on spool space during reception to use the SIZE
1038 parameter supplied by the sender MAIL FROM command. Previously it was
1039 ignored, and only the check_spool_space option value for the required
1042 JH/38 Fix $dkim_key_length. This should, after a DKIM verification, present
1043 the size of the signing public-key. Previously it was instead giving
1044 the size of the signature hash.
1046 JH/39 DKIM verification: the RFC 8301 restriction on sizes of RSA keys is now
1047 the default. See the (new) dkim_verify_min_keysizes option.
1049 JH/40 Fix a memory-handling bug: when a connection carried multiple messages
1050 and an ACL use a lookup for checking either the local_part or domain,
1051 stale data could be accessed. Ensure that variable references are
1052 dropped between messages.
1054 JH/41 Bug 2571: Fix SPA authenticator. Running as a server, an offset supplied
1055 by the client was not checked as pointing within response data before
1056 being used. A malicious client could thus cause an out-of-bounds read and
1057 possibly gain authentication. Fix by adding the check.
1059 JH/42 Internationalisation: change the default for downconversion in the smtp
1060 transport to be "if needed". Previously it was "as previously set" for
1061 the message, which usually meant "if needed" for message-submission but
1062 "no" for everything else. However, MTAs have been seen using SMTPUTF8
1063 even when the envelope addresses did not need it, resulting in forwarding
1064 failures to non-supporting MTAs. A downconvert in such cases will be
1065 a no-op on the addresses, merely dropping the use of SMTPUTF8 by the
1066 transport. The change does mean that addresses needing conversion will
1067 be converted when previously a delivery failure would occur.
1069 JH/43 Fix possible long line in DSN. Previously when a very long SMTP error
1070 response was received it would be used unchecked in a fail-DSN, violating
1071 standards on line-length limits. Truncate if needed.
1073 HS/01 Remove parameters of the link to www.open-spf.org. The linked form
1074 doesn't work. (Additionally add a new main config option to configure the
1081 JH/01 OpenSSL: With debug enabled output keying information sufficient, server
1082 side, to decode a TLS 1.3 packet capture.
1084 JH/02 OpenSSL: Suppress the sending of (stateful) TLS1.3 session tickets.
1085 Previously the default library behaviour applied, sending two, each in
1086 its own TCP segment.
1088 JH/03 Debug output for ACL now gives the config file name and line number for
1091 JH/04 The default received_header_text now uses the RFC 8314 tls cipher clause.
1093 JH/05 DKIM: ensure that dkim_domain elements are lowercased before use.
1095 JH/06 Fix buggy handling of autoreply bounce_return_size_limit, and a possible
1096 buffer overrun for (non-chunking) other transports.
1098 JH/07 GnuTLS: Our use of late (post-handshake) certificate verification, under
1099 TLS1.3, means that a server rejecting a client certificate is not visible
1100 to the client until the first read of encrypted data (typically the
1101 response to EHLO). Add detection for that case and treat it as a failed
1102 TLS connection attempt, so that the normal retry-in-clear can work (if
1103 suitably configured).
1105 JB/01 Bug 2375: fix expansions of 822 addresses having comments in local-part
1106 and/or domain. Found and fixed by Jason Betts.
1108 JH/08 Add hardening against SRV & TLSA lookups the hit CNAMEs (a nonvalid
1109 configuration). If a CNAME target was not a wellformed name pattern, a
1112 JH/09 Logging: Fix initial listening-on line for multiple ports for an IP when
1113 the OS reports them interleaved with other addresses.
1115 JH/10 OpenSSL: Fix aggregation of messages. Previously, when PIPELINING was
1116 used both for input and for a verify callout, both encrypted, SMTP
1117 responses being sent by the server could be lost. This resulted in
1118 dropped connections and sometimes bounces generated by a peer sending
1121 JH/11 Harden plaintext authenticator against a badly misconfigured client-send
1122 string. Previously it was possible to cause undefined behaviour in a
1123 library routine (usually a crash). Found by "zerons".
1125 JH/12 Bug 2384: fix "-bP smtp_receive_timeout". Previously it returned no
1128 JH/13 Bug 2386: Fix builds with Dane under LibreSSL 2.9.0 onward. Some old
1129 API was removed, so update to use the newer ones.
1131 JH/14 Bug 1891: Close the log file if receiving a non-smtp message, without
1132 any timeout set, is taking a long time. Previously we would hang on to a
1133 rotated logfile "forever" if the input was arriving with long gaps
1134 (a previous attempt to fix addressed lack, for a long time, of initial
1137 HS/01 Bug 2390: Use message_id for tempfile creation to avoid races in a
1138 shared (NFS) environment. The length of the tempfile name is now
1139 4 + 16 ("hdr.$message_exim_id") which might break on file
1140 systems which restrict the file name length to lower values.
1141 (It was "hdr.$pid".)
1143 HS/02 Bug 2390: Use message_id for tempfile creation to avoid races in a
1144 shared (NFS) environment.
1146 HS/03 Bug 2392: exigrep does case sensitive *option* processing (as it
1147 did for all versions <4.90). Notably -M, -m, --invert, -I may be
1150 JH/15 Use unsigned when creating bitmasks in macros, to avoid build errors
1151 on some platforms for bit 31.
1153 JH/16 GnuTLS: rework ciphersuite strings under recent library versions. Thanks
1154 to changes apparently associated with TLS1.3 handling some of the APIs
1155 previously used were either nonfunctional or inappropriate. Strings
1156 like TLS1.3:ECDHE_SECP256R1__RSA_PSS_RSAE_SHA256__AES_256_GCM__AEAD:256
1157 and TLS1.2:ECDHE_SECP256R1__RSA_SHA256__AES_128_CBC__SHA256:128 replace
1158 the previous TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256 .
1159 This affects log line X= elements, the $tls_{in,out}_cipher variables,
1160 and the use of specific cipher names in the encrypted= ACL condition.
1162 JH/17 OpenSSL: the default openssl_options now disables ssl_v3.
1164 JH/18 GnuTLS: fix $tls_out_ocsp under hosts_request_ocsp. Previously the
1165 verification result was not updated unless hosts_require_ocsp applied.
1167 JH/19 Bug 2398: fix listing of a named-queue. Previously, even with the option
1168 queue_list_requires_admin set to false, non-admin users were denied the
1171 JH/20 Bug 2389: fix server advertising of usable certificates, under GnuTLS in
1172 directory-of-certs mode. Previously they were advertised despite the
1175 JH/21 The smtp transport option "hosts_noproxy_tls" is now unset by default.
1176 A single TCP connection by a client will now hold a TLS connection open
1177 for multiple message deliveries, by default. Previously the default was to
1180 JH/22 The smtp transport option "hosts_try_dane" now enables all hosts by
1181 default. If built with the facility, DANE will be used. The facility
1182 SUPPORT_DANE is now enabled in the prototype build Makefile "EDITME".
1184 JH/23 The build default is now for TLS to be included; the SUPPORT_TLS define
1185 is replaced with DISABLE_TLS. Either USE_GNUTLS or (the new) USE_OPENSSL
1186 must be defined and you must still, unless you define DISABLE_TLS, manage
1187 the the include-dir and library-file requirements that go with that
1188 choice. Non-TLS builds are still supported.
1190 JH/24 Fix duplicated logging of peer name/address, on a transport connection-
1193 JH/25 The smtp transport option "hosts_try_fastopen" now enables all hosts by
1194 default. If the platform supports and has the facility enabled, it will
1195 be requested on all coneections.
1197 JH/26 The PIPE_CONNECT facility is promoted from experimental status and is now
1198 controlled by the build-time option SUPPORT_PIPE_CONNECT.
1200 PP/01 Unbreak heimdal_gssapi, broken in 4.92.
1202 JH/27 Bug 2404: Use the main-section configuration option "dsn_from" for
1203 success-DSN messages. Previously the From: header was always the default
1204 one for these; the option was ignored.
1206 JH/28 Fix the timeout on smtp response to apply to the whole response.
1207 Previously it was reset for every read, so a teergrubing peer sending
1208 single bytes within the time limit could extend the connection for a
1209 long time. Credit to Qualsys Security Advisory Team for the discovery.
1211 JH/29 Fix DSN Final-Recipient: field. Previously it was the post-routing
1212 delivery address, which leaked information of the results of local
1213 forwarding. Change to the original envelope recipient address, per
1216 JH/30 Bug 2411: Fix DSN generation when RFC 3461 failure notification is
1217 requested. Previously not bounce was generated and a log entry of
1218 error ignored was made.
1220 JH/31 Avoid re-expansion in ${sort } expansion. (CVE-2019-13917)
1222 JH/32 Introduce a general tainting mechanism for values read from the input
1223 channel, and values derived from them. Refuse to expand any tainted
1224 values, to catch one form of exploit.
1226 JH/33 Bug 2413: Fix dkim_strict option. Previously the expansion result
1227 was unused and the unexpanded text used for the test. Found and
1228 fixed by Ruben Jenster.
1230 JH/34 Fix crash after TLS shutdown. When the TCP/SMTP channel was left open,
1231 an attempt to use a TLS library read routine dereffed a nul pointer,
1234 JH/35 Bug 2409: filter out-of-spec chars from callout response before using
1235 them in our smtp response.
1237 JH/36 Have the general router option retry_use_local_part default to true when
1238 any of the restrictive preconditions are set (to anything). Previously it
1239 was only for check_local user. The change removes one item of manual
1240 configuration which is required for proper retries when a remote router
1241 handles a subset of addresses for a domain.
1243 JH/37 Appendfile: when evaluating quota use (non-quota_size_regex) take the file
1244 link count into consideration.
1246 HS/04 Fix handling of very log lines in -H files. If a -<key> <value> line
1247 caused the extension of big_buffer, the following lines were ignored.
1249 JH/38 Bug 1395: Teach the DNS negative-cache about TTL value from the SOA in
1250 accordance with RFC 2308. Previously there was no expiry, so a longlived
1251 receive process (eg. due to ACL delays) versus a short SOA value could
1254 HS/05 Handle trailing backslash gracefully. (CVE-2019-15846)
1256 JH/39 Promote DMARC support to mainline.
1258 JH/40 Bug 2452: Add a References: header to DSNs.
1260 JH/41 With GnuTLS 3.6.0 (and later) do not attempt to manage Diffie-Hellman
1261 parameters. The relevant library call is documented as "Deprecated: This
1262 function is unnecessary and discouraged on GnuTLS 3.6.0 or later. Since
1263 3.6.0, DH parameters are negotiated following RFC7919."
1265 HS/06 Change the default of dnssec_request_domains to "*"
1267 JH/42 Bug 2545: Fix CHUNKING for all RCPT commands rejected. Previously we
1268 carried on and emitted a BDAT command, even when PIPELINING was not
1271 JH/43 Bug 2465: Fix taint-handling in dsearch lookup. Previously a nontainted
1272 buffer was used for the filename, resulting in a trap when tainted
1273 arguments (eg. $domain) were used.
1275 JH/44 With OpenSSL 1.1.1 (onwards) disable renegotiation for TLS1.2 and below;
1276 recommended to avoid a possible server-load attack. The feature can be
1277 re-enabled via the openssl_options main cofiguration option.
1279 JH/45 local_scan API: documented the current smtp_printf() call. This changed
1280 for version 4.90 - adding a "more data" boolean to the arguments.
1281 Bumped the ABI version number also, this having been missed previously;
1282 release versions 4.90 to 4.92.3 inclusive were effectively broken in
1283 respect of usage of smtp_printf() by either local_scan code or libraries
1284 accessed via the ${dlfunc } expansion item. Both will need coding
1285 adjustment for any calls to smtp_printf() to match the new function
1286 signature; a FALSE value for the new argument is always safe.
1288 JH/46 FreeBSD: fix use of the sendfile() syscall. The shim was not updating
1289 the file-offset (which the Linux syscall does, and exim expects); this
1290 resulted in an indefinite loop.
1292 JH/47 ARC: fix crash in signing, triggered when a configuration error failed
1293 to do ARC verification. The Authentication-Results: header line added
1294 by the configuration then had no ARC item.
1296 JH/48 Bug 2784: fix shutdown=no in the ${readsocket) expansion item. Previously
1297 an incorrect mode was used for reading the result, resulting in it being
1304 JH/01 Remove code calling the customisable local_scan function, unless a new
1305 definition "HAVE_LOCAL_SCAN=yes" is present in the Local/Makefile.
1307 JH/02 Bug 1007: Avoid doing logging from signal-handlers, as that can result in
1308 non-signal-safe functions being used.
1310 JH/03 Bug 2269: When presented with a received message having a stupidly large
1311 number of DKIM-Signature headers, disable DKIM verification to avoid
1312 a resource-consumption attack. The limit is set at twenty.
1314 JH/04 Add variables $arc_domains, $arc_oldest_pass for ARC verify. Fix the
1315 report of oldest_pass in ${authres } in consequence, and separate out
1316 some descriptions of reasons for verification fail.
1318 JH/05 Bug 2273: Cutthrough delivery left a window where the received messsage
1319 files in the spool were present and unlocked. A queue-runner could spot
1320 them, resulting in a duplicate delivery. Fix that by doing the unlock
1321 after the unlink. Investigation by Tim Stewart. Take the opportunity to
1322 add more error-checking on spoolfile handling while that code is being
1325 PP/01 Refuse to open a spool data file (*-D) if it's a symlink.
1326 No known attacks, no CVE, this is defensive hardening.
1328 JH/06 Bug 2275: The MIME ACL unlocked the received message files early, and
1329 a queue-runner could start a delivery while other operations were ongoing.
1330 Cutthrough delivery was a common victim, resulting in duplicate delivery.
1331 Found and investigated by Tim Stewart. Fix by using the open message data
1332 file handle rather than opening another, and not locally closing it (which
1333 releases a lock) for that case, while creating the temporary .eml format
1334 file for the MIME ACL. Also applies to "regex" and "spam" ACL conditions.
1336 JH/07 Bug 177: Make a random-recipient callout success visible in ACL, by setting
1337 $sender_verify_failure/$recipient_verify_failure to "random".
1339 JH/08 When generating a selfsigned cert, use serial number 1 since zero is not
1342 JH/09 Bug 2274: Fix logging of cmdline args when starting in an unlinked cwd.
1343 Previously this would segfault.
1345 JH/10 Fix ARC signing for case when DKIM signing failed. Previously this would
1348 JH/11 Bug 2264: Exim now only follows CNAME chains one step by default. We'd
1349 like zero, since the resolver should be doing this for us, But we need one
1350 as a CNAME but no MX presence gets the CNAME returned; we need to check
1351 that doesn't point to an MX to declare it "no MX returned" rather than
1352 "error, loop". A new main option is added so the older capability of
1353 following some limited number of chain links is maintained.
1355 JH/12 Add client-ip info to non-pass iprev ${authres } lines.
1357 JH/13 For receent Openssl versions (1.1 onward) use modern generic protocol
1358 methods. These should support TLS 1.3; they arrived with TLS 1.3 and the
1359 now-deprecated earlier definitions used only specified the range up to TLS
1360 1.2 (in the older-version library docs).
1362 JH/14 Bug 2284: Fix DKIM signing for body lines starting with a pair of dots.
1364 JH/15 Rework TLS client-side context management. Stop using a global, and
1365 explicitly pass a context around. This enables future use of TLS for
1366 connections to service-daemons (eg. malware scanning) while a client smtp
1367 connection is using TLS; with cutthrough connections this is quite likely.
1369 JH/16 Fix ARC verification to do AS checks in reverse order.
1371 JH/17 Support a "tls" option on the ${readsocket } expansion item.
1373 JH/18 Bug 2287: Fix the protocol name (eg utf8esmtp) for multiple messages
1374 using the SMTPUTF8 option on their MAIL FROM commands, in one connection.
1375 Previously the "utf8" would be re-prepended for every additional message.
1377 JH/19 Reject MAIL FROM commands with SMTPUTF8 when the facility was not advertised.
1378 Previously thery were accepted, resulting in issues when attempting to
1379 forward messages to a non-supporting MTA.
1381 PP/02 Let -n work with printing macros too, not just options.
1383 JH/20 Bug 2296: Fix cutthrough for >1 address redirection. Previously only
1384 one parent address was copied, and bogus data was used at delivery-logging
1385 time. Either a crash (after delivery) or bogus log data could result.
1386 Discovery and analysis by Tim Stewart.
1388 PP/03 Make ${utf8clean:} expansion operator detect incomplete final character.
1389 Previously if the string ended mid-character, we did not insert the
1390 promised '?' replacement.
1392 PP/04 Documentation: current string operators work on bytes, not codepoints.
1394 JH/21 Change as many as possible of the global flags into one-bit bitfields; these
1395 should pack well giving a smaller memory footprint so better caching and
1396 therefore performance. Group the declarations where this can't be done so
1397 that the byte-sized flag variables are not interspersed among pointer
1398 variables, giving a better chance of good packing by the compiler.
1400 JH/22 Bug 1896: Fix the envelope from for DMARC forensic reports to be possibly
1401 non-null, to avoid issues with sites running BATV. Previously reports were
1402 sent with an empty envelope sender so looked like bounces.
1404 JH/23 Bug 2318: Fix the noerror command within filters. It wasn't working.
1405 The ignore_error flag wasn't being returned from the filter subprocess so
1406 was not set for later routers. Investigation and fix by Matthias Kurz.
1408 JH/24 Bug 2310: Raise a msg:fail:internal event for each undelivered recipient,
1409 and a msg:complete for the whole, when a message is manually removed using
1410 -Mrm. Developement by Matthias Kurz, hacked on by JH.
1412 JH/25 Avoid fixed-size buffers for pathnames in DB access. This required using
1413 a "Gnu special" function, asprintf() in the DB utility binary builds; I
1414 hope that is portable enough.
1416 JH/26 Bug 2311: Fix DANE-TA verification under GnuTLS. Previously it was also
1417 requiring a known-CA anchor certificate; make it now rely entirely on the
1418 TLSA as an anchor. Checking the name on the leaf cert against the name
1419 on the A-record for the host is still done for TA (but not for EE mode).
1421 JH/27 Fix logging of proxy address. Previously, a pointless "PRX=[]:0" would be
1422 included in delivery lines for non-proxied connections, when compiled with
1423 SUPPORT_SOCKS and running with proxy logging enabled.
1425 JH/28 Bug 2314: Fire msg:fail:delivery event even when error is being ignored.
1426 Developement by Matthias Kurz, tweaked by JH. While in that bit of code,
1427 move the existing event to fire before the normal logging of message
1428 failure so that custom logging is bracketed by normal logging.
1430 JH/29 Bug 2322: A "fail" command in a non-system filter (file) now fires the
1431 msg:fail:internal event. Developement by Matthias Kurz.
1433 JH/30 Bug 2329: Increase buffer size used for dns lookup from 2k, which was
1434 far too small for todays use of crypto signatures stored there. Go all
1435 the way to the max DNS message size of 64kB, even though this might be
1436 overmuch for IOT constrained device use.
1438 JH/31 Fix a bad use of a copy function, which could be used to pointlessly
1439 copy a string over itself. The library routine is documented as not
1440 supporting overlapping copies, and on MacOS it actually raised a SIGABRT.
1442 JH/32 For main options check_spool_space and check_inode_space, where the
1443 platform supports 64b integers, support more than the previous 2^31 kB
1444 (i.e. more than 2 TB). Accept E, P and T multipliers in addition to
1445 the previous G, M, k.
1447 JH/33 Bug 2338: Fix the cyrus-sasl authenticator to fill in the
1448 $authenticated_fail_id variable on authentication failure. Previously
1451 JH/34 Increase RSA keysize of autogen selfsign cert from 1024 to 2048. RHEL 8.0
1452 OpenSSL didn't want to use such a weak key. Do for GnuTLS also, and for
1453 more-modern GnuTLS move from GNUTLS_SEC_PARAM_LOW to
1454 GNUTLS_SEC_PARAM_MEDIUM.
1456 JH/35 OpenSSL: fail the handshake when SNI processing hits a problem, server
1457 side. Previously we would continue as if no SNI had been received.
1459 JH/36 Harden the handling of string-lists. When a list consisted of a sole
1460 "<" character, which should be a list-separator specification, we walked
1461 off past the nul-terimation.
1463 JH/37 Bug 2341: Send "message delayed" warning MDNs (restricted to external
1464 causes) even when the retry time is not yet met. Previously they were
1465 not, meaning that when (say) an account was over-quota and temp-rejecting,
1466 and multiple senders' messages were queued, only one sender would get
1467 notified on each configured delay_warning cycle.
1469 JH/38 Bug 2351: Log failures to extract envelope addresses from message headers.
1471 JH/39 OpenSSL: clear the error stack after an SSL_accept(). With anon-auth
1472 cipher-suites, an error can be left on the stack even for a succeeding
1473 accept; this results in impossible error messages when a later operation
1476 AM/01 Bug 2359: GnuTLS: repeat lowlevel read and write operations while they
1477 return error codes indicating retry. Under TLS1.3 this becomes required.
1479 JH/40 Fix the feature-cache refresh for EXPERIMENTAL_PIPE_CONNECT. Previously
1480 it only wrote the new authenticators, resulting in a lack of tracking of
1481 peer changes of ESMTP extensions until the next cache flush.
1483 JH/41 Fix the loop reading a message header line to check for integer overflow,
1484 and more-often against header_maxsize. Previously a crafted message could
1485 induce a crash of the recive process; now the message is cleanly rejected.
1487 JH/42 Bug 2366: Fix the behaviour of the dkim_verify_signers option. It had
1488 been totally disabled for all of 4.91. Discovery and fix by "Mad Alex".
1494 GF/01 DEFER rather than ERROR on redis cluster MOVED response.
1495 When redis_servers is set to a list of > 1 element, and the Redis servers
1496 in that list are in cluster configuration, convert the REDIS_REPLY_ERROR
1497 case of MOVED into a DEFER case instead, thus moving the query onto the
1498 next server in the list. For a cluster of N elements, all N servers must
1499 be defined in redis_servers.
1501 GF/02 Catch and remove uninitialized value warning in exiqsumm
1502 Check for existence of @ARGV before looking at $ARGV[0]
1504 JH/01 Replace the store_release() internal interface with store_newblock(),
1505 which internalises the check required to safely use the old one, plus
1506 the allocate and data copy operations duplicated in both (!) of the
1507 extant use locations.
1509 JH/02 Disallow '/' characters in queue names specified for the "queue=" ACL
1510 modifier. This matches the restriction on the commandline.
1512 JH/03 Fix pgsql lookup for multiple result-tuples with a single column.
1513 Previously only the last row was returned.
1515 JH/04 Bug 2217: Tighten up the parsing of DKIM signature headers. Previously
1516 we assumed that tags in the header were well-formed, and parsed the
1517 element content after inspecting only the first char of the tag.
1518 Assumptions at that stage could crash the receive process on malformed
1521 JH/05 Bug 2215: Fix crash associated with dnsdb lookup done from DKIM ACL.
1522 While running the DKIM ACL we operate on the Permanent memory pool so that
1523 variables created with "set" persist to the DATA ACL. Also (at any time)
1524 DNS lookups that fail create cache records using the Permanent pool. But
1525 expansions release any allocations made on the current pool - so a dnsdb
1526 lookup expansion done in the DKIM ACL releases the memory used for the
1527 DNS negative-cache, and bad things result. Solution is to switch to the
1528 Main pool for expansions.
1529 While we're in that code, add checks on the DNS cache during store_reset,
1530 active in the testsuite.
1531 Problem spotted, and debugging aided, by Wolfgang Breyha.
1533 JH/06 Fix issue with continued-connections when the DNS shifts unreliably.
1534 When none of the hosts presented to a transport match an already-open
1535 connection, close it and proceed with the list. Previously we would
1536 queue the message. Spotted by Lena with Yahoo, probably involving
1539 JH/07 Bug 2214: Fix SMTP responses resulting from non-accept result of MIME ACL.
1540 Previously a spurious "250 OK id=" response was appended to the proper
1543 JH/08 The "support for" informational output now, which built with Content
1544 Scanning support, has a line for the malware scanner interfaces compiled
1545 in. Interface can be individually included or not at build time.
1547 JH/09 The "aveserver", "kavdaemon" and "mksd" interfaces are now not included
1548 by the template makefile "src/EDITME". The "STREAM" support for an older
1549 ClamAV interface method is removed.
1551 JH/10 Bug 2223: Fix mysql lookup returns for the no-data case (when the number of
1552 rows affected is given instead).
1554 JH/11 The runtime Berkeley DB library version is now additionally output by
1555 "exim -d -bV". Previously only the compile-time version was shown.
1557 JH/12 Bug 2230: Fix cutthrough routing for nonfirst messages in an initiating
1558 SMTP connection. Previously, when one had more recipients than the
1559 first, an abortive onward connection was made. Move to full support for
1560 multiple onward connections in sequence, handling cutthrough connection
1561 for all multi-message initiating connections.
1563 JH/13 Bug 2229: Fix cutthrough routing for nonstandard port numbers defined by
1564 routers. Previously, a multi-recipient message would fail to match the
1565 onward-connection opened for the first recipient, and cause its closure.
1567 JH/14 Bug 2174: A timeout on connect for a callout was also erroneously seen as
1568 a timeout on read on a GnuTLS initiating connection, resulting in the
1569 initiating connection being dropped. This mattered most when the callout
1570 was marked defer_ok. Fix to keep the two timeout-detection methods
1573 JH/15 Relax results from ACL control request to enable cutthrough, in
1574 unsupported situations, from error to silently (except under debug)
1575 ignoring. This covers use with PRDR, frozen messages, queue-only and
1578 HS/01 Fix Buffer overflow in base64d() (CVE-2018-6789)
1580 JH/16 Fix bug in DKIM verify: a buffer overflow could corrupt the malloc
1581 metadata, resulting in a crash in free().
1583 PP/01 Fix broken Heimdal GSSAPI authenticator integration.
1584 Broken in f2ed27cf5, missing an equals sign for specified-initialisers.
1585 Broken also in d185889f4, with init system revamp.
1587 JH/17 Bug 2113: Fix conversation closedown with the Avast malware scanner.
1588 Previously we abruptly closed the connection after reading a malware-
1589 found indication; now we go on to read the "scan ok" response line,
1592 JH/18 Bug 2239: Enforce non-usability of control=utf8_downconvert in the mail
1593 ACL. Previously, a crash would result.
1595 JH/19 Speed up macro lookups during configuration file read, by skipping non-
1596 macro text after a replacement (previously it was only once per line) and
1597 by skipping builtin macros when searching for an uppercase lead character.
1599 JH/20 DANE support moved from Experimental to mainline. The Makefile control
1600 for the build is renamed.
1602 JH/21 Fix memory leak during multi-message connections using STARTTLS. A buffer
1603 was allocated for every new TLS startup, meaning one per message. Fix
1604 by only allocating once (OpenSSL) or freeing on TLS-close (GnuTLS).
1606 JH/22 Bug 2236: When a DKIM verification result is overridden by ACL, DMARC
1607 reported the original. Fix to report (as far as possible) the ACL
1608 result replacing the original.
1610 JH/23 Fix memory leak during multi-message connections using STARTTLS under
1611 OpenSSL. Certificate information is loaded for every new TLS startup,
1612 and the resources needed to be freed.
1614 JH/24 Bug 2242: Fix exim_dbmbuild to permit directoryless filenames.
1616 JH/25 Fix utf8_downconvert propagation through a redirect router. Previously it
1619 JH/26 Bug 2253: For logging delivery lines under PRDR, append the overall
1620 DATA response info to the (existing) per-recipient response info for
1621 the "C=" log element. It can have useful tracking info from the
1622 destination system. Patch from Simon Arlott.
1624 JH/27 Bug 2251: Fix ldap lookups that return a single attribute having zero-
1625 length value. Previously this would segfault.
1627 HS/02 Support Avast multiline protoocol, this allows passing flags to
1628 newer versions of the scanner.
1630 JH/28 Ensure that variables possibly set during message acceptance are marked
1631 dead before release of memory in the daemon loop. This stops complaints
1632 about them when the debug_store option is enabled. Discovered specifically
1633 for sender_rate_period, but applies to a whole set of variables.
1634 Do the same for the queue-runner and queue-list loops, for variables set
1635 from spool message files. Do the same for the SMTP per-message loop, for
1636 certain variables indirectly set in ACL operations.
1638 JH/29 Bug 2250: Fix a longstanding bug in heavily-pipelined SMTP input (such
1639 as a multi-recipient message from a mailinglist manager). The coding had
1640 an arbitrary cutoff number of characters while checking for more input;
1641 enforced by writing a NUL into the buffer. This corrupted long / fast
1642 input. The problem was exposed more widely when more pipelineing of SMTP
1643 responses was introduced, and one Exim system was feeding another.
1644 The symptom is log complaints of SMTP syntax error (NUL chars) on the
1645 receiving system, and refused recipients seen by the sending system
1646 (propating to people being dropped from mailing lists).
1647 Discovered and pinpointed by David Carter.
1649 JH/30 The (EXPERIMENTAL_DMARC) variable $dmarc_ar_header is withdrawn, being
1650 replaced by the ${authresults } expansion.
1652 JH/31 Bug 2257: Fix pipe transport to not use a socket-only syscall.
1654 HS/03 Set a handler for SIGTERM and call exit(3) if running as PID 1. This
1655 allows proper process termination in container environments.
1657 JH/32 Bug 2258: Fix spool_wireformat in combination with LMTP transport.
1658 Previously the "final dot" had a newline after it; ensure it is CR,LF.
1660 JH/33 SPF: remove support for the "spf" ACL condition outcome values "err_temp"
1661 and "err_perm", deprecated since 4.83 when the RFC-defined words
1662 "temperror" and "permerror" were introduced.
1664 JH/34 Re-introduce enforcement of no cutthrough delivery on transports having
1665 transport-filters or DKIM-signing. The restriction was lost in the
1666 consolidation of verify-callout and delivery SMTP handling.
1667 Extend the restriction to also cover ARC-signing.
1669 JH/35 Cutthrough: for a final-dot response timeout (and nonunderstood responses)
1670 in defer=pass mode supply a 450 to the initiator. Previously the message
1673 PP/02 DANE: add dane_require_tls_ciphers SMTP Transport option; if unset,
1674 tls_require_ciphers is used as before.
1676 HS/03 Malware Avast: Better match the Avast multiline protocol. Add
1677 "pass_unscanned". Only tmpfails from the scanner are written to
1678 the paniclog, as they may require admin intervention (permission
1679 denied, license issues). Other scanner errors (like decompression
1680 bombs) do not cause a paniclog entry.
1682 JH/36 Fix reinitialisation of DKIM logging variable between messages.
1683 Previously it was possible to log spurious information in receive log
1686 JH/37 Bug 2255: Revert the disable of the OpenSSL session caching. This
1687 triggered odd behaviour from Outlook Express clients.
1689 PP/03 Add util/renew-opendmarc-tlds.sh script for safe renewal of public
1692 JH/38 DKIM: accept Ed25519 pubkeys in SubjectPublicKeyInfo-wrapped form,
1693 since the IETF WG has not yet settled on that versus the original
1694 "bare" representation.
1696 JH/39 Fix syslog logging for syslog_timestamp=no and log_selector +millisec.
1697 Previously the millisecond value corrupted the output.
1698 Fix also for syslog_pid=no and log_selector +pid, for which the pid
1699 corrupted the output.
1705 JH/01 Rework error string handling in TLS interface so that the caller in
1706 more cases is responsible for logging. This permits library-sourced
1707 string to be attached to addresses during delivery, and collapses
1708 pairs of long lines into single ones.
1710 PP/01 Allow PKG_CONFIG_PATH to be set in Local/Makefile and use it correctly
1711 during configuration. Wildcards are allowed and expanded.
1713 JH/02 Rework error string handling in DKIM to pass more info back to callers.
1714 This permits better logging.
1716 JH/03 Rework the transport continued-connection mechanism: when TLS is active,
1717 do not close it down and have the child transport start it up again on
1718 the passed-on TCP connection. Instead, proxy the child (and any
1719 subsequent ones) for TLS via a unix-domain socket channel. Logging is
1720 affected: the continued delivery log lines do not have any DNSSEC, TLS
1721 Certificate or OCSP information. TLS cipher information is still logged.
1723 JH/04 Shorten the log line for daemon startup by collapsing adjacent sets of
1724 identical IP addresses on different listening ports. Will also affect
1727 PP/02 Bug 2070: uClibc defines __GLIBC__ without providing glibc headers;
1728 add noisy ifdef guards to special-case this sillyness.
1729 Patch from Bernd Kuhls.
1731 JH/05 Tighten up the checking in isip4 (et al): dotted-quad components larger
1732 than 255 are no longer allowed.
1734 JH/06 Default openssl_options to include +no_ticket, to reduce load on peers.
1735 Disable the session-cache too, which might reduce our load. Since we
1736 currrectly use a new context for every connection, both as server and
1737 client, there is no benefit for these.
1738 GnuTLS appears to not support tickets server-side by default (we don't
1739 call gnutls_session_ticket_enable_server()) but client side is enabled
1740 by default on recent versions (3.1.3 +) unless the PFS priority string
1743 PP/03 Add $SOURCE_DATE_EPOCH support for reproducible builds, per spec at
1744 <https://reproducible-builds.org/specs/source-date-epoch/>.
1746 JH/07 Fix smtp transport use of limited max_rcpt under mua_wrapper. Previously
1747 the check for any unsuccessful recipients did not notice the limit, and
1748 erroneously found still-pending ones.
1750 JH/08 Pipeline CHUNKING command and data together, on kernels that support
1751 MSG_MORE. Only in-clear (not on TLS connections).
1753 JH/09 Avoid using a temporary file during transport using dkim. Unless a
1754 transport-filter is involved we can buffer the headers in memory for
1755 creating the signature, and read the spool data file once for the
1756 signature and again for transmission.
1758 JH/10 Enable use of sendfile in Linux builds as default. It was disabled in
1759 4.77 as the kernel support then wasn't solid, having issues in 64bit
1760 mode. Now, it's been long enough. Add support for FreeBSD also.
1762 JH/11 Bug 2104: Fix continued use of a transport connection with TLS. In the
1763 case where the routing stage had gathered several addresses to send to
1764 a host before calling the transport for the first, we previously failed
1765 to close down TLS in the old transport process before passing the TCP
1766 connection to the new process. The new one sent a STARTTLS command
1767 which naturally failed, giving a failed delivery and bloating the retry
1768 database. Investigation and fix prototype from Wolfgang Breyha.
1770 JH/12 Fix check on SMTP command input synchronisation. Previously there were
1771 false-negatives in the check that the sender had not preempted a response
1772 or prompt from Exim (running as a server), due to that code's lack of
1773 awareness of the SMTP input buffering.
1775 PP/04 Add commandline_checks_require_admin option.
1776 Exim drops privileges sanely, various checks such as -be aren't a
1777 security problem, as long as you trust local users with access to their
1778 own account. When invoked by services which pass untrusted data to
1779 Exim, this might be an issue. Set this option in main configuration
1780 AND make fixes to the calling application, such as using `--` to stop
1783 JH/13 Do pipelining under TLS. Previously, although safe, no advantage was
1784 taken. Now take care to pack both (client) MAIL,RCPT,DATA, and (server)
1785 responses to those, into a single TLS record each way (this usually means
1786 a single packet). As a side issue, smtp_enforce_sync now works on TLS
1789 PP/05 OpenSSL/1.1: use DH_bits() for more accurate DH param sizes. This
1790 affects you only if you're dancing at the edge of the param size limits.
1791 If you are, and this message makes sense to you, then: raise the
1792 configured limit or use OpenSSL 1.1. Nothing we can do for older
1795 JH/14 For the "sock" variant of the malware scanner interface, accept an empty
1796 cmdline element to get the documented default one. Previously it was
1799 JH/15 Fix a crash in the smtp transport caused when two hosts in succession
1800 are unsuable for non-message-specific reasons - eg. connection timeout,
1801 banner-time rejection.
1803 JH/16 Fix logging of delivery remote port, when specified by router, under
1806 PP/06 Repair manualroute's ability to take options in any order, even if one
1807 is the name of a transport.
1810 HS/01 Cleanup, prevent repeated use of -p/-oMr (CVE-2017-1000369)
1812 JH/17 Change the list-building routines interface to use the expanding-string
1813 triplet model, for better allocation and copying behaviour.
1815 JH/18 Prebuild the data-structure for "builtin" macros, for faster startup.
1816 Previously it was constructed the first time a possibly-matching string
1817 was met in the configuration file input during startup; now it is done
1820 JH/19 Bug 2141: Use the full-complex API for Berkeley DB rather than the legacy-
1821 compatible one, to avoid the (poorly documented) possibility of a config
1822 file in the working directory redirecting the DB files, possibly correpting
1823 some existing file. CVE-2017-10140 assigned for BDB.
1825 JH/20 Bug 2147: Do not defer for a verify-with-callout-and-random which is not
1826 cache-hot. Previously, although the result was properly cached, the
1827 initial verify call returned a defer.
1829 JH/21 Bug 2151: Avoid using SIZE on the MAIL for a callout verify, on any but
1830 the main verify for receipient in uncached-mode.
1832 JH/22 Retire historical build files to an "unsupported" subdir. These are
1833 defined as "ones for which we have no current evidence of testing".
1835 JH/23 DKIM: enforce the DNS pubkey record "h" permitted-hashes optional field,
1836 if present. Previously it was ignored.
1838 JH/24 Start using specified-initialisers in C structure init coding. This is
1839 a C99 feature (it's 2017, so now considered safe).
1841 JH/25 Use one-bit bitfields for flags in the "addr" data structure. Previously
1842 if was a fixed-sized field and bitmask ops via macros; it is now more
1845 PP/07 GitHub PR 56: Apply MariaDB build fix.
1846 Patch provided by Jaroslav Škarvada.
1848 PP/08 Bug 2161: Fix regression in sieve quoted-printable handling introduced
1849 during Coverity cleanups [4.87 JH/47]
1850 Diagnosis and fix provided by Michael Fischer v. Mollard.
1852 JH/26 Fix DKIM bug: when the pseudoheader generated for signing was exactly
1853 the right size to place the terminating semicolon on its own folded
1854 line, the header hash was calculated to an incorrect value thanks to
1855 the (relaxed) space the fold became.
1857 HS/02 Fix Bug 2130: large writes from the transport subprocess were chunked
1858 and confused the parent.
1860 JH/27 Fix SOCKS bug: an unitialized pointer was deref'd by the transport process
1861 which could crash as a result. This could lead to undeliverable messages.
1863 JH/28 Logging: "next input sent too soon" now shows where input was truncated
1866 JH/29 Fix queue_run_in_order to ignore the PID portion of the message ID. This
1867 matters on fast-turnover and PID-randomising systems, which were getting
1868 out-of-order delivery.
1870 JH/30 Fix a logging bug on aarch64: an unsafe routine was previously used for
1871 a possibly-overlapping copy. The symptom was that "Remote host closed
1872 connection in response to HELO" was logged instead of the actual 4xx
1875 JH/31 Fix CHUNKING code to properly flush the unwanted chunk after an error.
1876 Previously only that bufferd was discarded, resulting in SYMTP command
1879 JH/32 DKIM: when a message has multiple signatures matching an identity given
1880 in dkim_verify_signers, run the dkim acl once for each. Previously only
1881 one run was done. Bug 2189.
1883 JH/33 Downgrade an unfound-list name (usually a typo in the config file) from
1884 "panic the current process" to "deliberately defer". The panic log is
1885 still written with the problem list name; the mail and reject logs now
1886 get a temp-reject line for the message that was being handled, saying
1887 something like "domains check lookup or other defer". The SMTP 451
1888 message is still "Temporary local problem".
1890 JH/34 Bug 2199: Fix a use-after-free while reading smtp input for header lines.
1891 A crafted sequence of BDAT commands could result in in-use memory beeing
1892 freed. CVE-2017-16943.
1894 HS/03 Bug 2201: Fix checking for leading-dot on a line during headers reading
1895 from SMTP input. Previously it was always done; now only done for DATA
1896 and not BDAT commands. CVE-2017-16944.
1898 JH/35 Bug 2201: Flush received data in BDAT mode after detecting an error fatal
1899 to the message (such as an overlong header line). Previously this was
1900 not done and we did not exit BDAT mode. Followon from the previous item
1901 though a different problem.
1907 JH/01 Bug 1922: Support IDNA2008. This has slightly different conversion rules
1908 than -2003 did; needs libidn2 in addition to libidn.
1910 JH/02 The path option on a pipe transport is now expanded before use.
1912 PP/01 GitHub PR 50: Do not call ldap_start_tls_s on ldapi:// connections.
1913 Patch provided by "Björn", documentation fix added too.
1915 JH/03 Bug 2003: fix Proxy Protocol v2 handling: the address size field was
1916 missing a wire-to-host endian conversion.
1918 JH/04 Bug 2004: fix CHUNKING in non-PIPELINEING mode. Chunk data following
1919 close after a BDAT command line could be taken as a following command,
1920 giving a synch failure. Fix by only checking for synch immediately
1921 before acknowledging the chunk.
1923 PP/02 GitHub PR 52: many spelling fixes, which include fixing parsing of
1924 no_require_dnssec option and creation of _HAVE_TRANSPORT_APPEND_MAILDIR
1925 macro. Patches provided by Josh Soref.
1927 JH/05 Have the EHLO response advertise VRFY, if there is a vrfy ACL defined.
1928 Previously we did not; the RFC seems ambiguous and VRFY is not listed
1929 by IANA as a service extension. However, John Klensin suggests that we
1932 JH/06 Bug 2017: Fix DKIM verification in -bh test mode. The data feed into
1933 the dkim code may be unix-mode line endings rather than smtp wire-format
1934 CRLF, so prepend a CR to any bare LF.
1936 JH/07 Rationalise the coding for callout smtp conversations and transport ones.
1937 As a side-benfit, callouts can now use PIPELINING hence fewer round-trips.
1939 JH/08 Bug 2016: Fix DKIM verification vs. CHUNKING. Any BDAT commands after
1940 the first were themselves being wrongly included in the feed into dkim
1941 processing; with most chunk sizes in use this resulted in an incorrect
1942 body hash calculated value.
1944 JH/09 Bug 2014: permit inclusion of a DKIM-Signature header in a received
1945 DKIM signature block, for verification. Although advised against by
1946 standards it is specifically not ruled illegal.
1948 JH/10 Bug 2025: Fix reception of (quoted) local-parts with embedded spaces.
1950 JH/11 Bug 2029: Fix crash in DKIM verification when a message signature block is
1951 missing a body hash (the bh= tag).
1953 JH/12 Bug 2018: Re-order Proxy Protocol startup versus TLS-on-connect startup.
1954 It seems that HAProxy sends the Proxy Protocol information in clear and
1955 only then does a TLS startup, so do the same.
1957 JH/13 Bug 2027: Avoid attempting to use TCP Fast Open for non-transport client
1958 TCP connections (such as for Spamd) unless the daemon successfully set
1959 Fast Open mode on its listening sockets. This fixes breakage seen on
1960 too-old kernels or those not configured for Fast Open, at the cost of
1961 requiring both directions being enabled for TFO, and TFO never being used
1962 by non-daemon-related Exim processes.
1964 JH/14 Bug 2000: Reject messages recieved with CHUNKING but with malformed line
1965 endings, at least on the first header line. Try to canonify any that get
1966 past that check, despite the cost.
1968 JH/15 Angle-bracket nesting (an error inserted by broken sendmails) levels are
1969 now limited to an arbitrary five deep, while parsing addresses with the
1970 strip_excess_angle_brackets option enabled.
1972 PP/03 Bug 2018: For Proxy Protocol and TLS-on-connect, do not over-read and
1973 instead leave the unprompted TLS handshake in socket buffer for the
1974 TLS library to consume.
1976 PP/04 Bug 2018: Also handle Proxy Protocol v2 safely.
1978 PP/05 FreeBSD compat: handle that Ports no longer create /usr/bin/perl
1980 JH/16 Drop variables when they go out of scope. Memory management drops a whole
1981 region in one operation, for speed, and this leaves assigned pointers
1982 dangling. Add checks run only under the testsuite which checks all
1983 variables at a store-reset and panics on a dangling pointer; add code
1984 explicitly nulling out all the variables discovered. Fixes one known
1985 bug: a transport crash, where a dangling pointer for $sending_ip_address
1986 originally assigned in a verify callout, is re-used.
1988 PP/06 Drop '.' from @INC in various Perl scripts.
1990 PP/07 Switch FreeBSD iconv to always use the base-system libc functions.
1992 PP/08 Reduce a number of compilation warnings under clang; building with
1993 CC=clang CFLAGS+=-Wno-dangling-else -Wno-logical-op-parentheses
1994 should be warning-free.
1996 JH/17 Fix inbound CHUNKING when DKIM disabled at runtime.
1998 HS/01 Fix portability problems introduced by PP/08 for platforms where
1999 realloc(NULL) is not equivalent to malloc() [SunOS et al].
2001 HS/02 Bug 1974: Fix missing line terminator on the last received BDAT
2002 chunk. This allows us to accept broken chunked messages. We need a more
2003 general solution here.
2005 PP/09 Wrote util/chunking_fixqueue_finalnewlines.pl to help recover
2006 already-broken messages in the queue.
2008 JH/18 Bug 2061: Fix ${extract } corrupting an enclosing ${reduce } $value.
2010 JH/19 Fix reference counting bug in routing-generated-address tracking.
2016 JH/01 Use SIZE on MAIL FROM in a cutthrough connection, if the destination
2017 supports it and a size is available (ie. the sending peer gave us one).
2019 JH/02 The obsolete acl condition "demime" is removed (finally, after ten
2020 years of being deprecated). The replacements are the ACLs
2021 acl_smtp_mime and acl_not_smtp_mime.
2023 JH/03 Upgrade security requirements imposed for hosts_try_dane: previously
2024 a downgraded non-dane trust-anchor for the TLS connection (CA-style)
2025 or even an in-clear connection were permitted. Now, if the host lookup
2026 was dnssec and dane was requested then the host is only used if the
2027 TLSA lookup succeeds and is dnssec. Further hosts (eg. lower priority
2028 MXs) will be tried (for hosts_try_dane though not for hosts_require_dane)
2029 if one fails this test.
2030 This means that a poorly-configured remote DNS will make it incommunicado;
2031 but it protects against a DNS-interception attack on it.
2033 JH/04 Bug 1810: make continued-use of an open smtp transport connection
2034 non-noisy when a race steals the message being considered.
2036 JH/05 If main configuration option tls_certificate is unset, generate a
2037 self-signed certificate for inbound TLS connections.
2039 JH/06 Bug 165: hide more cases of password exposure - this time in expansions
2040 in rewrites and routers.
2042 JH/07 Retire gnutls_require_mac et.al. These were nonfunctional since 4.80
2043 and logged a warning sing 4.83; now they are a configuration file error.
2045 JH/08 Bug 1836: Fix crash in VRFY handling when handed an unqualified name
2046 (lacking @domain). Apply the same qualification processing as RCPT.
2048 JH/09 Bug 1804: Avoid writing msglog files when in -bh or -bhc mode.
2050 JH/10 Support ${sha256:} applied to a string (as well as the previous
2053 JH/11 Cutthrough: avoid using the callout hints db on a verify callout when
2054 a cutthrough deliver is pending, as we always want to make a connection.
2055 This also avoids re-routing the message when later placing the cutthrough
2056 connection after a verify cache hit.
2057 Do not update it with the verify result either.
2059 JH/12 Cutthrough: disable when verify option success_on_redirect is used, and
2060 when routing results in more than one destination address.
2062 JH/13 Cutthrough: expand transport dkim_domain option when testing for dkim
2063 signing (which inhibits the cutthrough capability). Previously only
2064 the presence of an option was tested; now an expansion evaluating as
2065 empty is permissible (obviously it should depend only on data available
2066 when the cutthrough connection is made).
2068 JH/14 Fix logging of errors under PIPELINING. Previously the log line giving
2069 the relevant preceding SMTP command did not note the pipelining mode.
2071 JH/15 Fix counting of empty lines in $body_linecount and $message_linecount.
2072 Previously they were not counted.
2074 JH/16 DANE: treat a TLSA lookup response having all non-TLSA RRs, the same
2075 as one having no matching records. Previously we deferred the message
2076 that needed the lookup.
2078 JH/17 Fakereject: previously logged as a normal message arrival "<="; now
2079 distinguished as "(=".
2081 JH/18 Bug 1867: make the fail_defer_domains option on a dnslookup router work
2082 for missing MX records. Previously it only worked for missing A records.
2084 JH/19 Bug 1850: support Radius libraries that return REJECT_RC.
2086 JH/20 Bug 1872: Ensure that acl_smtp_notquit is run when the connection drops
2087 after the data-go-ahead and data-ack. Patch from Jason Betts.
2089 JH/21 Bug 1846: Send DMARC forensic reports for reject and quarantine results,
2090 even for a "none" policy. Patch from Tony Meyer.
2092 JH/22 Fix continued use of a connection for further deliveries. If a port was
2093 specified by a router, it must also match for the delivery to be
2096 JH/23 Bug 1874: fix continued use of a connection for further deliveries.
2097 When one of the recipients of a message was unsuitable for the connection
2098 (has no matching addresses), we lost track of needing to mark it
2099 deferred. As a result mail would be lost.
2101 JH/24 Bug 1832: Log EHLO response on getting conn-close response for HELO.
2103 JH/25 Decoding ACL controls is now done using a binary search; the source code
2104 takes up less space and should be simpler to maintain. Merge the ACL
2105 condition decode tables also, with similar effect.
2107 JH/26 Fix problem with one_time used on a redirect router which returned the
2108 parent address unchanged. A retry would see the parent address marked as
2109 delivered, so not attempt the (identical) child. As a result mail would
2112 JH/27 Fix a possible security hole, wherein a process operating with the Exim
2113 UID can gain a root shell. Credit to http://www.halfdog.net/ for
2114 discovery and writeup. Ubuntu bug 1580454; no bug raised against Exim
2117 JH/28 Enable {spool,log} filesystem space and inode checks as default.
2118 Main config options check_{log,spool}_{inodes,space} are now
2119 100 inodes, 10MB unless set otherwise in the configuration.
2121 JH/29 Fix the connection_reject log selector to apply to the connect ACL.
2122 Previously it only applied to the main-section connection policy
2125 JH/30 Bug 1897: fix callouts connection fallback from TLS to cleartext.
2127 PP/01 Changed default Diffie-Hellman parameters to be Exim-specific, created
2128 by me. Added RFC7919 DH primes as an alternative.
2130 PP/02 Unbreak build via pkg-config with new hash support when crypto headers
2131 are not in the system include path.
2133 JH/31 Fix longstanding bug with aborted TLS server connection handling. Under
2134 GnuTLS, when a session startup failed (eg because the client disconnected)
2135 Exim did stdio operations after fclose. This was exposed by a recent
2136 change which nulled out the file handle after the fclose.
2138 JH/32 Bug 1909: Fix OCSP proof verification for cases where the proof is
2139 signed directly by the cert-signing cert, rather than an intermediate
2140 OCSP-signing cert. This is the model used by LetsEncrypt.
2142 JH/33 Bug 1914: Ensure socket is nonblocking before draining after SMTP QUIT.
2144 HS/01 Fix leak in verify callout under GnuTLS, about 3MB per recipient on
2145 an incoming connection.
2147 HS/02 Bug 1802: Do not half-close the connection after sending a request
2150 HS/03 Use "auto" as the default EC curve parameter. For OpenSSL < 1.0.2
2151 fallback to "prime256v1".
2153 JH/34 SECURITY: Use proper copy of DATA command in error message.
2154 Could leak key material. Remotely exploitable. CVE-2016-9963.
2160 JH/01 Bug 1664: Disable OCSP for GnuTLS library versions at/before 3.3.16
2161 and 3.4.4 - once the server is enabled to respond to an OCSP request
2162 it does even when not requested, resulting in a stapling non-aware
2163 client dropping the TLS connection.
2165 TF/01 Code cleanup: Overhaul the debug_selector and log_selector machinery to
2166 support variable-length bit vectors. No functional change.
2168 TF/02 Improve the consistency of logging incoming and outgoing interfaces.
2169 The I= interface field on outgoing lines is now after the H= remote
2170 host field, same as incoming lines. There is a separate
2171 outgoing_interface log selector which allows you to disable the
2174 JH/02 Bug 728: Close logfiles after a daemon-process "exceptional" log write.
2175 If not running log_selector +smtp_connection the mainlog would be held
2176 open indefinitely after a "too many connections" event, including to a
2177 deleted file after a log rotate. Leave the per net connection logging
2178 leaving it open for efficiency as that will be quickly detected by the
2179 check on the next write.
2181 HS/01 Bug 1671: Fix post transport crash.
2182 Processing the wait-<transport> messages could crash the delivery
2183 process if the message IDs didn't exist for some reason. When
2184 using 'split_spool_directory=yes' the construction of the spool
2185 file name failed already, exposing the same netto behaviour.
2187 JH/03 Bug 425: Capture substrings in $regex1, $regex2 etc from regex &
2188 mime_regex ACL conditions.
2190 JH/04 Bug 1686: When compiled with EXPERIMENTAL_DSN_INFO: Add extra information
2191 to DSN fail messages (bounces): remote IP, remote greeting, remote response
2192 to HELO, local diagnostic string.
2194 JH/05 Downgrade message for a TLS-certificate-based authentication fail from
2195 log line to debug. Even when configured with a tls authenticator many
2196 client connections are expected to not authenticate in this way, so
2197 an authenticate fail is not an error.
2199 HS/02 Add the Exim version string to the process info. This way exiwhat
2200 gives some more detail about the running daemon.
2202 JH/06 Bug 1395: time-limit caching of DNS lookups, to the TTL value. This may
2203 matter for fast-change records such as DNSBLs.
2205 JH/07 Bug 1678: Always record an interface option value, if set, as part of a
2206 retry record, even if constant. There may be multiple transports with
2207 different interface settings and the retry behaviour needs to be kept
2210 JH/08 Bug 1586: exiqgrep now refuses to run if there are unexpected arguments.
2212 JH/09 Bug 1700: ignore space & tab embedded in base64 during decode.
2214 JH/10 Bug 840: fix log_defer_output option of pipe transport
2216 JH/11 Bug 830: use same host for all RCPTS of a message, even under
2217 hosts_randomize. This matters a lot when combined with mua_wrapper.
2219 JH/12 Bug 1706: percent and underbar characters are no longer escaped by the
2220 ${quote_pgsql:<string>} operator.
2222 JH/13 Bug 1708: avoid misaligned access in cached lookup.
2224 JH/14 Change header file name for freeradius-client. Relevant if compiling
2225 with Radius support; from the Gentoo tree and checked under Fedora.
2227 JH/15 Bug 1712: Introduce $prdr_requested flag variable
2229 JH/16 Bug 1714: Permit an empty string as expansion result for transport
2230 option transport_filter, meaning no filtering.
2232 JH/17 Bug 1713: Fix non-PDKIM_DEBUG build. Patch from Jasen Betts.
2234 JH/18 Bug 1709: When built with TLS support, the tls_advertise_hosts option now
2235 defaults to "*" (all hosts). The variable is now available when not built
2236 with TLS, default unset, mainly to enable keeping the testsuite sane.
2237 If a server certificate is not supplied (via tls_certificate) an error is
2238 logged, and clients will find TLS connections fail on startup. Presumably
2239 they will retry in-clear.
2240 Packagers of Exim are strongly encouraged to create a server certificate
2241 at installation time.
2243 HS/03 Add -bP config_file as a synonym for -bP configure_file, for consistency
2244 with the $config_file variable.
2246 JH/19 Two additional event types: msg:rcpt:defer and msg:rcpt:host:defer. Both
2247 in transport context, after the attempt, and per-recipient. The latter type
2248 is per host attempted. The event data is the error message, and the errno
2249 information encodes the lookup type (A vs. MX) used for the (first) host,
2250 and the trailing two digits of the smtp 4xx response.
2252 GF/01 Bug 1715: Fix for race condition in exicyclog, where exim could attempt
2253 to write to mainlog (or rejectlog, paniclog) in the window between file
2254 creation and permissions/ownership being changed. Particularly affects
2255 installations where exicyclog is run as root, rather than exim user;
2256 result is that the running daemon panics and dies.
2258 JH/20 Bug 1701: For MySQL lookups, support MySQL config file option group names.
2260 JH/21 Bug 1720: Add support for priority groups and weighted-random proxy
2261 selection for the EXPERIMENTAL_SOCKS feature, via new per-proxy options
2262 "pri" and "weight". Note that the previous implicit priority given by the
2263 list order is no longer honoured.
2265 JH/22 Bugs 963, 1721: Fix some corner cases in message body canonicalization
2266 for DKIM processing.
2268 JH/23 Move SOCKS5 support from Experimental to mainline, enabled for a build
2269 by defining SUPPORT_SOCKS.
2271 JH/26 Move PROXY support from Experimental to mainline, enabled for a build
2272 by defining SUPPORT_PROXY. Note that the proxy_required_hosts option
2273 is renamed to hosts_proxy, and the proxy_{host,target}_{address,port}.
2274 variables are renamed to proxy_{local,external}_{address,port}.
2276 JH/27 Move Internationalisation support from Experimental to mainline, enabled
2277 for a build by defining SUPPORT_I18N
2279 JH/28 Bug 1745: Fix redis lookups to handle (quoted) spaces embedded in parts
2280 of the query string, and make ${quote_redis:} do that quoting.
2282 JH/29 Move Events support from Experimental to mainline, enabled by default
2283 and removable for a build by defining DISABLE_EVENT.
2285 JH/30 Updated DANE implementation code to current from Viktor Dukhovni.
2287 JH/31 Fix bug with hosts_connection_nolog and named-lists which were wrongly
2288 cached by the daemon.
2290 JH/32 Move Redis support from Experimental to mainline, enabled for a build
2291 by defining LOOKUP_REDIS. The libhiredis library is required.
2293 JH/33 Bug 1748: Permit ACL dnslists= condition in non-smtp ACLs if explicit
2294 keys are given for lookup.
2296 JH/34 Bug 1192: replace the embedded copy of PolarSSL RSA routines in the DKIM
2297 support, by using OpenSSL or GnuTLS library ones. This means DKIM is
2298 only supported when built with TLS support. The PolarSSL SHA routines
2299 are still used when the TLS library is too old for convenient support.
2301 JH/35 Require SINGLE_DH_USE by default in OpenSSL (main config option
2302 openssl_options), for security. OpenSSL forces this from version 1.1.0
2303 server-side so match that on older versions.
2305 JH/36 Bug 1778: longstanding bug in memory use by the ${run } expansion: A fresh
2306 allocation for $value could be released as the expansion processing
2307 concluded, but leaving the global pointer active for it.
2309 JH/37 Bug 1769: Permit a VRFY ACL to override the default 252 response,
2310 and to use the domains and local_parts ACL conditions.
2312 JH/38 Fix cutthrough bug with body lines having a single dot. The dot was
2313 incorrectly not doubled on cutthrough transmission, hence seen as a
2314 body-termination at the receiving system - resulting in truncated mails.
2315 Commonly the sender saw a TCP-level error, and retransmitted the message
2316 via the normal store-and-forward channel. This could result in duplicates
2317 received - but deduplicating mailstores were liable to retain only the
2318 initial truncated version.
2320 JH/39 Bug 1781: Fix use of DKIM private-keys having trailing '=' in the base-64.
2322 JH/40 Fix crash in queryprogram router when compiled with EXPERIMENTAL_SRS.
2324 JH/41 Bug 1792: Fix selection of headers to sign for DKIM: bottom-up. While
2325 we're in there, support oversigning also; bug 1309.
2327 JH/42 Bug 1796: Fix error logged on a malware scanner connection failure.
2329 HS/04 Add support for keep_environment and add_environment options.
2331 JH/43 Tidy coding issues detected by gcc --fsanitize=undefined. Some remain;
2332 either intentional arithmetic overflow during PRNG, or testing config-
2335 JH/44 Bug 1800: The combination of a -bhc commandline option and cutthrough
2336 delivery resulted in actual delivery. Cancel cutthrough before DATA
2339 JH/45 Fix cutthrough, when connection not opened by verify and target hard-
2340 rejects a recipient: pass the reject to the originator.
2342 JH/46 Multiple issues raised by Coverity. Some were obvious or plausible bugs.
2343 Many were false-positives and ignorable, but it's worth fixing the
2346 JH/47 Fix build on HP-UX and older Solaris, which need (un)setenv now also
2347 for the new environment-manipulation done at startup. Move the routines
2348 from being local to tls.c to being global via the os.c file.
2350 JH/48 Bug 1807: Fix ${extract } for the numeric/3-string case. While preparsing
2351 an extract embedded as result-arg for a map, the first arg for extract
2352 is unavailable so we cannot tell if this is a numbered or keyed
2353 extraction. Accept either.
2359 JH/01 Bug 1545: The smtp transport option "retry_include_ip_address" is now
2362 JH/02 The smtp transport option "multi_domain" is now expanded.
2364 JH/03 The smtp transport now requests PRDR by default, if the server offers
2367 JH/04 Certificate name checking on server certificates, when exim is a client,
2368 is now done by default. The transport option tls_verify_cert_hostnames
2369 can be used to disable this per-host. The build option
2370 EXPERIMENTAL_CERTNAMES is withdrawn.
2372 JH/05 The value of the tls_verify_certificates smtp transport and main options
2373 default to the word "system" to access the system default CA bundle.
2374 For GnuTLS, only version 3.0.20 or later.
2376 JH/06 Verification of the server certificate for a TLS connection is now tried
2377 (but not required) by default. The verification status is now logged by
2378 default, for both outbound TLS and client-certificate supplying inbound
2381 JH/07 Changed the default rfc1413 lookup settings to disable calls. Few
2384 JH/08 The EXPERIMENTAL_DSN compile option is no longer needed; all Delivery
2385 Status Notification (bounce) messages are now MIME format per RFC 3464.
2386 Support for RFC 3461 DSN options NOTIFY,ENVID,RET,ORCPT can be advertised
2387 under the control of the dsn_advertise_hosts option, and routers may
2388 have a dsn_lasthop option.
2390 JH/09 A timeout of 2 minutes is now applied to all malware scanner types by
2391 default, modifiable by a malware= option. The list separator for
2392 the options can now be changed in the usual way. Bug 68.
2394 JH/10 The smtp_receive_timeout main option is now expanded before use.
2396 JH/11 The incoming_interface log option now also enables logging of the
2397 local interface on delivery outgoing connections.
2399 JH/12 The cutthrough-routing facility now supports multi-recipient mails,
2400 if the interface and destination host and port all match.
2402 JH/13 Bug 344: The verify = reverse_host_lookup ACL condition now accepts a
2405 JH/14 Bug 1573: The spam= ACL condition now additionally supports Rspamd.
2406 Patch from Andrew Lewis.
2408 JH/15 Bug 670: The spamd_address main option (for the spam= ACL condition)
2409 now supports optional time-restrictions, weighting, and priority
2410 modifiers per server. Patch originally by <rommer@active.by>.
2412 JH/16 The spamd_address main option now supports a mixed list of local
2413 and remote servers. Remote servers can be IPv6 addresses, and
2414 specify a port-range.
2416 JH/17 Bug 68: The spamd_address main option now supports an optional
2417 timeout value per server.
2419 JH/18 Bug 1581: Router and transport options headers_add/remove can
2420 now have the list separator specified.
2422 JH/19 Bug 392: spamd_address, and clamd av_scanner, now support retry
2425 JH/20 Bug 1571: Ensure that $tls_in_peerdn is set, when verification fails
2428 JH/21 Support for the A6 type of dns record is withdrawn.
2430 JH/22 Bug 608: The result of a QUIT or not-QUIT toplevel ACL now matters
2431 rather than the verbs used.
2433 JH/23 Bug 1572: Increase limit on SMTP confirmation message copy size
2434 from 255 to 1024 chars.
2436 JH/24 Verification callouts now attempt to use TLS by default.
2438 HS/01 DNSSEC options (dnssec_require_domains, dnssec_request_domains)
2439 are generic router options now. The defaults didn't change.
2441 JH/25 Bug 466: Add RFC2322 support for MIME attachment filenames.
2442 Original patch from Alexander Shikoff, worked over by JH.
2444 HS/02 Bug 1575: exigrep falls back to autodetection of compressed
2445 files if ZCAT_COMMAND is not executable.
2447 JH/26 Bug 1539: Add timeout/retry options on dnsdb lookups.
2449 JH/27 Bug 286: Support SOA lookup in dnsdb lookups.
2451 JH/28 Bug 1588: Do not use the A lookup following an AAAA for setting the FQDN.
2452 Normally benign, it bites when the pair was led to by a CNAME;
2453 modern usage is to not canonicalize the domain to a CNAME target
2454 (and we were inconsistent anyway for A-only vs AAAA+A).
2456 JH/29 Bug 1632: Removed the word "rejected" from line logged for ACL discards.
2458 JH/30 Check the forward DNS lookup for DNSSEC, in addition to the reverse,
2459 when evaluating $sender_host_dnssec.
2461 JH/31 Check the HELO verification lookup for DNSSEC, adding new
2462 $sender_helo_dnssec variable.
2464 JH/32 Bug 1397: Enable ECDHE on OpenSSL, just the NIST P-256 curve.
2466 JH/33 Bug 1346: Note MAIL cmd seen in -bS batch, to avoid smtp_no_mail log.
2468 JH/34 Bug 1648: Fix a memory leak seen with "mailq" and large queues.
2470 JH/35 Bug 1642: Fix support of $spam_ variables at delivery time. Was
2471 documented as working, but never had. Support all but $spam_report.
2473 JH/36 Bug 1659: Guard checking of input smtp commands again pseudo-command
2474 added for tls authenticator.
2476 HS/03 Add perl_taintmode main config option
2482 TL/01 When running the test suite, the README says that variables such as
2483 no_msglog_check are global and can be placed anywhere in a specific
2484 test's script, however it was observed that placement needed to be near
2485 the beginning for it to behave that way. Changed the runtest perl
2486 script to read through the entire script once to detect and set these
2487 variables, reset to the beginning of the script, and then run through
2488 the script parsing/test process like normal.
2490 TL/02 The BSD's have an arc4random API. One of the functions to induce
2491 adding randomness was arc4random_stir(), but it has been removed in
2492 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
2493 function when detected.
2495 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
2496 cause callback expansion.
2498 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
2499 syntax errors in an expansion can be treated as a string instead of
2500 logging or causing an error, due to the internal use of bool_lax
2501 instead of bool when processing it.
2503 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
2504 server certificates when making smtp deliveries.
2506 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
2508 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
2510 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
2512 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
2513 Merged patch from Sebastian Wiedenroth.
2515 JH/05 Fix results-pipe from transport process. Several recipients, combined
2516 with certificate use, exposed issues where response data items split
2517 over buffer boundaries were not parsed properly. This eventually
2518 resulted in duplicates being sent. This issue only became common enough
2519 to notice due to the introduction of connection certificate information,
2520 the item size being so much larger. Found and fixed by Wolfgang Breyha.
2522 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
2523 size buffer was used, resulting in syntax errors when an expansion
2526 JH/07 Add support for directories of certificates when compiled with a GnuTLS
2527 version 3.3.6 or later.
2529 JH/08 Rename the TPDA experimental facility to Event Actions. The #ifdef
2530 is EXPERIMENTAL_EVENT, the main-configuration and transport options
2531 both become "event_action", the variables become $event_name, $event_data
2532 and $event_defer_errno. There is a new variable $verify_mode, usable in
2533 routers, transports and related events. The tls:cert event is now also
2534 raised for inbound connections, if the main configuration event_action
2537 TL/06 In test suite, disable OCSP for old versions of openssl which contained
2538 early OCSP support, but no stapling (appears to be less than 1.0.0).
2540 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
2541 server certificate names available under the smtp transport option
2542 "tls_verify_cert_hostname" now do not permit multi-component wildcard
2545 JH/10 Time-related extraction expansions from certificates now use the main
2546 option "timezone" setting for output formatting, and are consistent
2547 between OpenSSL and GnuTLS compilations. Bug 1541.
2549 JH/11 Fix a crash in mime ACL when meeting a zero-length, quoted or RFC2047-
2550 encoded parameter in the incoming message. Bug 1558.
2552 JH/12 Bug 1527: Autogrow buffer used in reading spool files. Since they now
2553 include certificate info, eximon was claiming there were spoolfile
2556 JH/13 Bug 1521: Fix ldap lookup for single-attr request, multiple-attr return.
2558 JH/14 Log delivery-related information more consistently, using the sequence
2559 "H=<name> [<ip>]" wherever possible.
2561 TL/07 Bug 1547: Omit RFCs from release. Draft and RFCs have licenses which
2562 are problematic for Debian distribution, omit them from the release
2565 JH/15 Updates and fixes to the EXPERIMENTAL_DSN feature.
2567 JH/16 Fix string representation of time values on 64bit time_t architectures.
2570 JH/17 Fix a null-indirection in certextract expansions when a nondefault
2571 output list separator was used.
2576 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
2577 checkers that were complaining about end of non-void function with no
2580 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
2581 This was a regression introduced in 4.83 by another bugfix.
2583 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
2585 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
2586 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
2592 TF/01 Correctly close the server side of TLS when forking for delivery.
2594 When a message was received over SMTP with TLS, Exim failed to clear up
2595 the incoming connection properly after forking off the child process to
2596 deliver the message. In some situations the subsequent outgoing
2597 delivery connection happened to have the same fd number as the incoming
2598 connection previously had. Exim would try to use TLS and fail, logging
2599 a "Bad file descriptor" error.
2601 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
2602 utilities have not been installed.
2604 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
2605 temporary space as the ACL may create new global variables.
2607 TL/01 LDAP support uses per connection or global context settings, depending
2608 upon the detected version of the libraries at build time.
2610 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
2611 to extract and use the src ip:port in logging and expansions as if it
2612 were a direct connection from the outside internet. PPv2 support was
2613 updated based on HAProxy spec change in May 2014.
2615 JH/02 Add ${listextract {number}{list}{success}{fail}}.
2617 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
2618 Properly escape header and check for NULL return.
2620 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
2623 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
2625 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
2626 characters in header names, implemented as a verify condition.
2627 Contributed by Michael Fischer v. Mollard.
2629 TL/05 Rename SPF condition results err_perm and err_temp to standardized
2630 results permerror and temperror. Previous values are deprecated but
2631 still accepted. In a future release, err_perm and err_temp will be
2632 completely removed, which will be a backward incompatibility if the
2633 ACL tests for either of these two old results. Patch contributed by
2634 user bes-internal on the mailing list.
2636 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
2638 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
2639 selectors, in both main and reject logs.
2641 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
2644 JH/07 Add malware type "sock" for talking to simple daemon.
2646 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
2648 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
2649 routers/transports under cutthrough routing.
2651 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
2652 numbers. Touch up "bool" conditional to keep the same definition.
2654 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
2656 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
2658 JH/12 Expand items in router/transport headers_add or headers_remove lists
2659 individually rather than the list as a whole. Bug 1452.
2661 Required for reasonable handling of multiple headers_ options when
2662 they may be empty; requires that headers_remove items with embedded
2663 colons must have them doubled (or the list-separator changed).
2665 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
2666 view the policy declared in the DMARC record. Currently, $dmarc_status
2667 is a combined value of both the record presence and the result of the
2670 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
2672 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
2673 dnslookup router and the smtp transport (applying to the forward
2676 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
2677 of ldap servers used for a specific lookup. Patch provided by Heiko
2680 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
2681 New variable $lookup_dnssec_authenticated for observability.
2683 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
2684 Patch submitted by Lars Timman.
2686 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
2688 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
2689 Requires trusted mode and valid format message id, aborts otherwise.
2690 Patch contributed by Heiko Schlichting.
2692 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
2693 certextract with support for various fields. Bug 1358.
2695 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
2696 is requested by default, modifiable by smtp transport option
2699 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
2700 operate on certificate variables to give certificate fingerprints
2701 Also new ${sha256:cert_variable}.
2703 JH/23 The PRDR feature is moved from being Experimental into the mainline.
2705 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
2706 Christian Aistleitner.
2708 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
2710 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
2711 file. Patch from Wolfgang Breyha.
2713 JH/25 Expand the coverage of the delivery $host and $host_address to
2714 client authenticators run in verify callout. Bug 1476.
2716 JH/26 Port service names are now accepted for tls_on_connect_ports, to
2717 align with daemon_smtp_ports. Bug 72.
2719 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
2720 support and error reporting did not work properly.
2722 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
2723 and is readable. Patch from Andrew Colin Kissa.
2725 TL/14 Enhance documentation of ${run expansion and how it parses the
2726 commandline after expansion, particularly in the case when an
2727 unquoted variable expansion results in an empty value.
2729 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
2731 PP/02 Fix internal collision of T_APL on systems which support RFC3123
2732 by renaming away from it. Addresses GH issue 15, reported by
2735 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
2737 TL/15 SECURITY: prevent double expansion in math comparison functions
2738 (can expand unsanitized data). Not remotely exploitable.
2745 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
2747 PP/02 Make -n do something, by making it not do something.
2748 When combined with -bP, the name of an option is not output.
2750 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
2753 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
2754 $sender_host_name and config options to manage this, and basic check
2757 PP/05 DSCP support for outbound connections and control modifier for inbound.
2759 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
2760 (Only plugin which currently uses this is kerberos4, which nobody should
2761 be using, but we should make it available and other future plugins might
2762 conceivably use it, even though it would break NAT; stuff *should* be
2763 using channel bindings instead).
2765 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
2766 name; added for Sendmail compatibility; requires admin caller.
2767 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
2768 just ignore it); requires trusted caller.
2769 Also parse but ignore: -Ac -Am -X<logfile>
2772 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
2774 TL/02 Add +smtp_confirmation as a default logging option.
2776 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
2777 Patch by Magnus Holmgren from 2007-02-20.
2779 TL/04 Bugzilla 1281 - Spec typo.
2780 Bugzilla 1283 - Spec typo.
2781 Bugzilla 1290 - Spec grammar fixes.
2783 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
2785 TL/06 Add Experimental DMARC support using libopendmarc libraries.
2787 TL/07 Fix an out of order global option causing a segfault. Reported to dev
2788 mailing list by by Dmitry Isaikin.
2790 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
2792 JH/02 Support "G" suffix to numbers in ${if comparisons.
2794 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
2796 NM/01 Bugzilla 1197 - Spec typo
2797 Bugzilla 1196 - Spec examples corrections
2799 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
2801 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
2802 gnutls_enable_pkcs11, but renamed to more accurately indicate its
2805 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
2806 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
2808 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
2809 "acl {{name}{arg}...}", and optional args on acl condition
2812 JH/05 Permit multiple router/transport headers_add/remove lines.
2814 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
2816 JH/07 Avoid using a waiting database for a single-message-only transport.
2817 Performance patch from Paul Fisher. Bugzilla 1262.
2819 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
2822 JH/09 Add $headers_added variable, with content from use of ACL modifier
2823 add_header (but not yet added to the message). Bugzilla 199.
2825 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
2826 Pulled from Bugzilla 817 by Wolfgang Breyha.
2828 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
2830 (nb: this is the same fix as in Exim 4.80.1)
2832 JH/11 Add A= logging on delivery lines, and a client_set_id option on
2835 JH/12 Add optional authenticated_sender logging to A= and a log_selector
2838 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
2840 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
2841 advertise SMTP AUTH mechanism to us, instead of a generic
2842 protocol violation error. Also, make Exim more robust to bad
2843 data from the Dovecot auth socket.
2845 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
2847 When a queue runner is handling a message, Exim first routes the
2848 recipient addresses, during which it prunes them based on the retry
2849 hints database. After that it attempts to deliver the message to
2850 any remaining recipients. It then updates the hints database using
2853 So if a recipient address works intermittently, it can get repeatedly
2854 deferred at routing time. The retry hints record remains fresh so the
2855 address never reaches the final cutoff time.
2857 This is a fairly common occurrence when a user is bumping up against
2858 their storage quota. Exim had some logic in its local delivery code
2859 to deal with this. However it did not apply to per-recipient defers
2860 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
2862 This change adds a proper retry rule check during routing so that the
2863 final cutoff time is checked against the message's age. We only do
2864 this check if there is an address retry record and there is not a
2865 domain retry record; this implies that previous attempts to handle
2866 the address had the retry_use_local_parts option turned on. We use
2867 this as an approximation for the destination being like a local
2868 delivery, as in LMTP.
2870 I suspect this new check makes the old local delivery cutoff check
2871 redundant, but I have not verified this so I left the code in place.
2873 TF/02 Correct gecos expansion when From: is a prefix of the username.
2875 Test 0254 submits a message to Exim with the header
2879 When I ran the test suite under the user fanf2, Exim expanded
2880 the header to contain my full name, whereas it should have added
2881 a Resent-Sender: header. It erroneously treats any prefix of the
2882 username as equal to the username.
2884 This change corrects that bug.
2886 GF/01 DCC debug and logging tidyup
2887 Error conditions log to paniclog rather than rejectlog.
2888 Debug lines prefixed by "DCC: " to remove any ambiguity.
2890 TF/03 Avoid unnecessary rebuilds of lookup-related code.
2892 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
2893 Bug spotted by Jeremy Harris; was flawed since initial commit.
2894 Would have resulted in OCSP responses post-SNI triggering an Exim
2895 NULL dereference and crash.
2897 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
2899 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
2900 Bug detection, analysis and fix by Samuel Thibault.
2901 Bugzilla 1331, Debian bug #698092.
2903 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
2905 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
2906 Server implementation by Todd Lyons, client by JH.
2907 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
2908 config variable "prdr_enable" controls whether the server
2909 advertises the facility. If the client requests PRDR a new
2910 acl_data_smtp_prdr ACL is called once for each recipient, after
2911 the body content is received and before the acl_smtp_data ACL.
2912 The client is controlled by both of: a hosts_try_prdr option
2913 on the smtp transport, and the server advertisement.
2914 Default client logging of deliveries and rejections involving
2915 PRDR are flagged with the string "PRDR".
2917 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
2918 fclose(). Diagnosis by Todd Lyons.
2920 PP/17 Update configure.default to handle IPv6 localhost better.
2921 Patch by Alain Williams (plus minor tweaks).
2924 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
2925 This is now consistent with GnuTLS, and is now documented: the
2926 previous undocumented portable approach to treating the option as
2927 unset was to force an expansion failure. That still works, and
2928 an empty string is now equivalent.
2930 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
2931 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
2932 not performing validation itself.
2934 PP/20 Added force_command boolean option to pipe transport.
2935 Patch from Nick Koston, of cPanel Inc.
2937 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
2940 TF/04 Added udpsend ACL modifier and hexquote expansion operator
2942 PP/21 Fix eximon continuous updating with timestamped log-files.
2943 Broken in a format-string cleanup in 4.80, missed when I repaired the
2944 other false fix of the same issue.
2945 Report and fix from Heiko Schlichting.
2948 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
2949 Report from Prashanth Katuri.
2951 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
2952 It's SecureTransport, so affects any MacOS clients which use the
2953 system-integrated TLS libraries, including email clients.
2955 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
2956 using a MIME ACL for non-SMTP local injection.
2957 Report and assistance in diagnosis by Warren Baker.
2959 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
2961 JH/16 Fix comparisons for 64b. Bugzilla 1385.
2963 TL/09 Add expansion variable $authenticated_fail_id to keep track of
2964 last id that failed so it may be referenced in subsequent ACL's.
2966 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
2969 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
2970 ldap library initialization, allowing self-signed CA's to be
2971 used. Also properly sets require_cert option later in code by
2972 using NULL (global ldap config) instead of ldap handle (per
2973 session). Bug diagnosis and testing by alxgomz.
2975 TL/12 Enhanced documentation in the ratelimit.pl script provided in
2976 the src/util/ subdirectory.
2978 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
2979 renamed to Transport Post Delivery Action by Jeremy Harris, as
2982 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
2983 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
2984 redis_servers = needs to be configured which will be used by the redis
2985 lookup. Patch from Warren Baker, of The Packet Hub.
2987 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
2989 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
2990 hostname or reverse DNS when processing a host list. Used suggestions
2991 from multiple comments on this bug.
2993 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
2995 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
2996 Missed a few lines, added it to make the runtest require no keyboard
2999 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
3000 contains upper case chars. Make router use caseful_local_part.
3002 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
3003 support when GnuTLS has been built with p11-kit.
3009 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
3011 This, or similar/improved, will also be change PP/11 of 4.82.
3017 PP/01 Handle short writes when writing local log-files.
3018 In practice, only affects FreeBSD (8 onwards).
3019 Bugzilla 1053, with thanks to Dmitry Isaikin.
3021 NM/01 Bugzilla 949 - Documentation tweak
3023 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
3026 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
3028 PP/02 Implemented gsasl authenticator.
3030 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
3032 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
3033 `pkg-config foo` for cflags/libs.
3035 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
3036 with rest of GSASL and with heimdal_gssapi.
3038 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
3039 `pkg-config foo` for cflags/libs for the TLS implementation.
3041 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
3042 properties get this fed in as external SSF. A number of robustness
3043 and debugging improvements to the cyrus_sasl authenticator.
3045 PP/08 cyrus_sasl server now expands the server_realm option.
3047 PP/09 Bugzilla 1214 - Log authentication information in reject log.
3048 Patch by Jeremy Harris.
3050 PP/10 Added dbmjz lookup type.
3052 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
3054 PP/12 MAIL args handles TAB as well as SP, for better interop with
3055 non-compliant senders.
3056 Analysis and variant patch by Todd Lyons.
3058 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
3059 Bug report from Lars Müller <lars@samba.org> (via SUSE),
3060 Patch from Dirk Mueller <dmueller@suse.com>
3062 PP/13 tls_peerdn now print-escaped for spool files.
3063 Observed some $tls_peerdn in wild which contained \n, which resulted
3064 in spool file corruption.
3066 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
3067 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
3068 or write after TLS renegotiation, which otherwise led to messages
3071 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
3072 as a tracking header (ie: a signed header comes before the signature).
3073 Patch from Wolfgang Breyha.
3075 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
3076 comma-sep list; embedded commas doubled.
3078 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
3080 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
3082 Report and patch from Dmitry Banschikov.
3084 PP/16 Removed "dont_insert_empty_fragments" from "openssl_options".
3085 Removed SSL_clear() after SSL_new() which led to protocol negotiation
3086 failures. We appear to now support TLS1.1+ with Exim.
3088 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
3089 lets Exim select keys and certificates based upon TLS SNI from client.
3090 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
3091 before an outbound SMTP session. New log_selector, +tls_sni.
3093 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
3094 NULL dereference. Report and patch from Alun Jones.
3096 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
3097 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
3098 Not seeing resolver debug output on NetBSD, but suspect this is a
3099 resolver implementation change.
3101 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
3102 Left warnings. Added "eximon gdb" invocation mode.
3104 PP/21 Defaulting "accept_8bitmime" to true, not false.
3106 PP/22 Added -bw for inetd wait mode support.
3108 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
3109 locate the relevant includes and libraries. Made this the default.
3111 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
3112 Bugzilla 1246, report and most of solution from Tomasz Kusy.
3114 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
3115 This may cause build issues on older platforms.
3117 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
3118 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
3119 gnutls_require_mac & gnutls_require_protocols (no longer supported).
3120 Added SNI support via GnuTLS too.
3121 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
3123 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
3125 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
3126 Applied second patch from Janne, implementing suggestion to default
3127 multiple-strings-in-record handling to match SPF spec.
3129 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
3131 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
3132 read-only, out of scope).
3133 Patch from Wolfgang Breyha, report from Stuart Northfield.
3135 PP/29 Fix three issues highlighted by clang analyser static analysis.
3136 Only crash-plausible issue would require the Cambridge-specific
3137 iplookup router and a misconfiguration.
3138 Report from Marcin Mirosław.
3140 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
3142 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
3143 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
3144 As part of this, removing so much warning spew let me fix some minor
3145 real issues in debug logging.
3147 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
3148 assignment on my part. Fixed.
3150 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
3151 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
3152 Janne Snabb (who went above and beyond: thank you).
3154 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
3155 string otherwise requires a connection and a bunch more work and it's
3156 relatively easy to get wrong. Should also expose TLS library linkage
3159 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
3160 64-bit ${eval} (JH/03).
3162 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
3163 GNU libc to support some of the 64-bit stuff, should not lead to
3164 conflicts. Defined before os.h is pulled in, so if a given platform
3165 needs to override this, it can.
3167 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
3168 protection layer was required, which is not implemented.
3169 Bugzilla 1254, patch from Wolfgang Breyha.
3171 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
3172 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
3173 tls_dhparam take prime identifiers. Also unbreak combination of
3174 OpenSSL+DH_params+TLSSNI.
3176 PP/39 Disable SSLv2 by default in OpenSSL support.
3182 PP/01 Solaris build fix for Oracle's LDAP libraries.
3183 Bugzilla 1109, patch from Stephen Usher.
3185 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
3187 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
3190 TF/02 Fix a couple more cases where we did not log the error message
3191 when unlink() failed. See also change 4.74-TF/03.
3193 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
3194 lock up or crash if it happened to be inside a call to libc when it
3195 got a SIGUSR1 from exiwhat.
3197 The SIGUSR1 handler appends the current process status to the process
3198 log which is later printed by exiwhat. It used to use the general
3199 purpose logging code to do this, but several functions it calls are
3200 not safe for signals.
3202 The new output code in the SIGUSR1 handler is specific to the process
3203 log, and simple enough that it's easy to inspect for signal safety.
3204 Removing some special cases also simplifies the general logging code.
3205 Removing the spurious timestamps from the process log simplifies
3208 TF/04 Improved ratelimit ACL condition.
3210 The /noupdate option has been deprecated in favour of /readonly which
3211 has clearer semantics. The /leaky, /strict, and /readonly update modes
3212 are mutually exclusive. The update mode is no longer included in the
3213 database key; it just determines when the database is updated. (This
3214 means that when you upgrade Exim will forget old rate measurements.)
3216 Exim now checks that the per_* options are used with an update mode that
3217 makes sense for the current ACL. For example, when Exim is processing a
3218 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
3219 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
3220 must specify per_mail/readonly. If you omit the update mode it defaults to
3221 /leaky where that makes sense (as before) or /readonly where required.
3223 The /noupdate option is now undocumented but still supported for
3224 backwards compatibility. It is equivalent to /readonly except that in
3225 ACLs where /readonly is required you may specify /leaky/noupdate or
3226 /strict/noupdate which are treated the same as /readonly.
3228 A useful new feature is the /count= option. This is a generalization
3229 of the per_byte option, so that you can measure the throughput of other
3230 aggregate values. For example, the per_byte option is now equivalent
3231 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
3233 The per_rcpt option has been generalized using the /count= mechanism
3234 (though it's more complicated than the per_byte equivalence). When it is
3235 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
3236 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
3237 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
3238 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
3239 Note that using per_rcpt with a non-readonly update mode in more than
3240 one ACL will cause the recipients to be double-counted. (The per_mail
3241 and per_byte options don't have this problem.)
3243 The handling of very low rates has changed slightly. If the computed rate
3244 is less than the event's count (usually one) then this event is the first
3245 after a long gap. In this case the rate is set to the same as this event's
3246 count, so that the first message of a spam run is counted properly.
3248 The major new feature is a mechanism for counting the rate of unique
3249 events. The new per_addr option counts the number of different
3250 recipients that someone has sent messages to in the last time period. It
3251 behaves like per_rcpt if all the recipient addresses are different, but
3252 duplicate recipient addresses do not increase the measured rate. Like
3253 the /count= option this is a general mechanism, so the per_addr option
3254 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
3255 example, measure the rate that a client uses different sender addresses
3256 with the options per_mail/unique=$sender_address. There are further
3257 details in the main documentation.
3259 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
3261 TF/06 Removed a few PCRE remnants.
3263 TF/07 Automatically extract Exim's version number from tags in the git
3264 repository when doing development or release builds.
3266 PP/02 Raise smtp_cmd_buffer_size to 16kB.
3267 Bugzilla 879. Patch from Paul Fisher.
3269 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
3270 Heavily based on revision 40f9a89a from Simon Arlott's tree.
3273 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
3275 PP/05 Variable $av_failed, true if the AV scanner deferred.
3276 Bugzilla 1078. Patch from John Horne.
3278 PP/06 Stop make process more reliably on build failure.
3279 Bugzilla 1087. Patch from Heiko Schlittermann.
3281 PP/07 Make maildir_use_size_file an _expandable_ boolean.
3282 Bugzilla 1089. Patch from Heiko Schlittermann.
3284 PP/08 Handle ${run} returning more data than OS pipe buffer size.
3285 Bugzilla 1131. Patch from Holger Weiß.
3287 PP/09 Handle IPv6 addresses with SPF.
3288 Bugzilla 860. Patch from Wolfgang Breyha.
3290 PP/10 GnuTLS: support TLS 1.2 & 1.1.
3292 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
3295 PP/11 match_* no longer expand right-hand-side by default.
3296 New compile-time build option, EXPAND_LISTMATCH_RHS.
3297 New expansion conditions, "inlist", "inlisti".
3299 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
3301 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
3303 PP/14 fix log_write() format string regression from TF/03.
3304 Bugzilla 1152. Patch from Dmitry Isaikin.
3310 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
3312 PP/02 Harmonised TLS library version reporting; only show if debugging.
3313 Layout now matches that introduced for other libraries in 4.74 PP/03.
3315 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
3317 PP/04 New "dns_use_edns0" global option.
3319 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
3322 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
3323 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
3325 TK/01 Updated PolarSSL code to 0.14.2.
3326 Bugzilla 1097. Patch from Andreas Metzler.
3328 PP/07 Catch divide-by-zero in ${eval:...}.
3329 Fixes bugzilla 1102.
3331 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
3334 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
3335 format-string attack -- SECURITY: remote arbitrary code execution.
3337 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
3338 time unintentionally subject to list matching rules, letting the header
3339 cause arbitrary Exim lookups (of items which can occur in lists, *not*
3340 arbitrary string expansion). This allowed for information disclosure.
3342 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
3343 INT_MIN/-1 -- value coerced to INT_MAX.
3349 NM/01 Workaround for PCRE version dependency in version reporting
3352 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
3353 This fixes portability to compilers other than gcc, notably
3354 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
3356 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
3357 makefiles for portability to HP-UX and POSIX correctness.
3359 PP/01 Permit LOOKUP_foo enabling on the make command-line.
3360 Also via indented variable definition in the Makefile.
3361 (Debugging by Oliver Heesakkers).
3363 PP/02 Restore caching of spamd results with expanded spamd_address.
3364 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
3366 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
3367 Improves build reliability. Fix from: Frank Elsner
3369 NM/02 Fix wide character breakage in the rfc2047 coding
3370 Fixes bug 1064. Patch from Andrey N. Oktyabrski
3372 NM/03 Allow underscore in dnslist lookups
3373 Fixes bug 1026. Patch from Graeme Fowler
3375 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
3376 Code patches from Adam Ciarcinski of NetBSD.
3378 NM/04 Fixed exiqgrep to cope with mailq missing size issue
3381 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
3382 is logged, to avoid truncation. Patch from John Horne.
3384 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
3385 Patch from Jakob Hirsch.
3387 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
3388 SQL string expansion failure details.
3389 Patch from Andrey Oktyabrski.
3391 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
3392 Patch from Simon Arlott.
3394 PP/09 New lookups functionality failed to compile on old gcc which rejects
3395 extern declarations in function scope.
3396 Patch from Oliver Fleischmann
3398 PP/10 Use sig_atomic_t for flags set from signal handlers.
3399 Check getgroups() return and improve debugging.
3400 Fixed developed for diagnosis in bug 927 (which turned out to be
3403 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
3404 Patch from Mark Zealey.
3406 PP/12 Bugzilla 1056: Improved spamd server selection.
3407 Patch from Mark Zealey.
3409 PP/13 Bugzilla 1086: Deal with maildir quota file races.
3410 Based on patch from Heiko Schlittermann.
3412 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
3413 Patch from Uwe Doering, sign-off by Michael Haardt.
3415 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
3416 variable declaration deep within a block. Bug and patch from
3419 PP/15 lookups-Makefile IRIX compatibility coercion.
3421 PP/16 Make DISABLE_DKIM build knob functional.
3423 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
3424 Patch by Simon Arlott
3426 TF/03 Fix valgrind.h portability to C89 compilers that do not support
3427 variable argument macros. Our copy now differs from upstream.
3433 TF/01 Failure to get a lock on a hints database can have serious
3434 consequences so log it to the panic log.
3436 TF/02 Log LMTP confirmation messages in the same way as SMTP,
3437 controlled using the smtp_confirmation log selector.
3439 TF/03 Include the error message when we fail to unlink a spool file.
3441 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
3442 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
3443 for maintaining out-of-tree patches for some time.
3445 PP/01 Bugzilla 139: Documentation and portability issues.
3446 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
3447 Handle per-OS dynamic-module compilation flags.
3449 PP/02 Let /dev/null have normal permissions.
3450 The 4.73 fixes were a little too stringent and complained about the
3451 permissions on /dev/null. Exempt it from some checks.
3452 Reported by Andreas M. Kirchwitz.
3454 PP/03 Report version information for many libraries, including
3455 Exim version information for dynamically loaded libraries. Created
3456 version.h, now support a version extension string for distributors
3457 who patch heavily. Dynamic module ABI change.
3459 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
3460 privilege escalation vulnerability whereby the Exim run-time user
3461 can cause root to append content of the attacker's choosing to
3464 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
3467 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
3468 If dropping privileges for untrusted macros, we disabled normal logging
3469 on the basis that it would fail; for the Exim run-time user, this is not
3470 the case, and it resulted in successful deliveries going unlogged.
3471 Fixed. Reported by Andreas Metzler.
3477 PP/01 Date: & Message-Id: revert to normally being appended to a message,
3478 only prepend for the Resent-* case. Fixes regression introduced in
3479 Exim 4.70 by NM/22 for Bugzilla 607.
3481 PP/02 Include check_rfc2047_length in configure.default because we're seeing
3482 increasing numbers of administrators be bitten by this.
3484 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
3486 PP/03 Bugzilla 994: added openssl_options main configuration option.
3488 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
3490 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
3492 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
3494 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
3495 without a peer certificate, leading to a segfault because of an
3496 assumption that peers always have certificates. Be a little more
3497 paranoid. Problem reported by Martin Tscholak.
3499 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
3500 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
3501 NB: ClamAV planning to remove STREAM in "middle of 2010".
3502 CL also introduces -bmalware, various -d+acl logging additions and
3503 more caution in buffer sizes.
3505 PP/09 Implemented reverse_ip expansion operator.
3507 PP/10 Bugzilla 937: provide a "debug" ACL control.
3509 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
3511 PP/12 Bugzilla 973: Implement --version.
3513 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
3515 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
3517 PP/15 Bugzilla 816: support multiple condition rules on Routers.
3519 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
3520 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
3521 ignore trailing whitespace.
3523 JJ/02 prevent non-panic DKIM error from being sent to paniclog
3525 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
3528 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
3529 Notification from Dr Andrew Aitchison.
3531 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
3532 ExtendedDetectionInfo response format.
3533 Notification from John Horne.
3535 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
3538 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
3539 XSL and documented dependency on system catalogs, with examples of how
3542 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
3545 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
3546 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
3547 configuration file which is writeable by the Exim user or group.
3549 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
3550 of configuration files to cover files specified with the -C option if
3551 they are going to be used with root privileges, not just the default
3554 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
3555 option (effectively making it always true).
3557 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
3558 files to be used while preserving root privileges.
3560 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
3561 that rogue child processes cannot use them.
3563 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
3564 run-time user, instead of root.
3566 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
3567 Exim run-time user without dropping privileges.
3569 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
3570 result string, instead of calling string_vformat() twice with the same
3573 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
3574 for other users. Others should always drop root privileges if they use
3575 -C on the command line, even for a whitelisted configure file.
3577 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
3579 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
3585 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
3586 $data_path, and $header_path variables; fixed documentation bugs and
3589 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
3590 exipick to access non-standard spools, including the "frozen" queue
3593 NM/01 Bugzilla 965: Support mysql stored procedures.
3594 Patch from Alain Williams
3596 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
3598 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
3599 Patch from Andreas Metzler
3601 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
3602 Patch from Kirill Miazine
3604 NM/05 Bugzilla 671: Added umask to procmail example.
3606 JJ/03 installed exipick 20100323.0, fixing doc bug
3608 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
3609 directory. Notification and patch from Dan Rosenberg.
3611 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
3613 TK/02 Improve log output when DKIM signing operation fails.
3615 MH/01 Treat the transport option dkim_domain as a colon separated
3616 list, not as a single string, and sign the message with each element,
3617 omitting multiple occurences of the same signer.
3619 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
3620 Bugzilla 985, 986. Patch by Simon Arlott
3622 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
3623 Patch by Simon Arlott
3625 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
3626 MBX locking. Notification from Dan Rosenberg.
3632 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
3634 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
3636 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
3638 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
3640 NM/04 Bugzilla 915: Flag broken perl installation during build.
3646 TK/01 Added patch by Johannes Berg that expands the main option
3647 "spamd_address" if it starts with a dollar sign.
3649 TK/02 Write list of recipients to X-Envelope-Sender header when building
3650 the mbox-format spool file for content scanning (suggested by Jakob
3653 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
3654 (http://www.dcc-servers.net/) support via dccifd. Activated by
3655 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
3657 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
3658 by Mark Daniel Reidel <mr@df.eu>.
3660 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
3661 When building exim an external PCRE library is now needed -
3662 PCRE is a system library on the majority of modern systems.
3663 See entry on PCRE_LIBS in EDITME file.
3665 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
3666 conversation. Added nologin parameter to request.
3667 Patch contributed by Kirill Miazine.
3669 TF/01 Do not log submission mode rewrites if they do not change the address.
3671 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
3673 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
3674 log files in place. Contributed by Roberto Lima.
3676 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
3678 TF/03 Bugzilla 615: When checking the local_parts router precondition
3679 after a local_part_suffix or local_part_prefix option, Exim now
3680 does not use the address's named list lookup cache, since this
3681 contains cached lookups for the whole local part.
3683 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
3684 Robert Millan. Documentation is in experimental-spec.txt.
3686 TF/04 Bugzilla 668: Fix parallel build (make -j).
3688 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
3690 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
3691 Patch provided by Jan Srzednicki.
3693 TF/05 Leading white space used to be stripped from $spam_report which
3694 wrecked the formatting. Now it is preserved.
3696 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
3697 that they are available at delivery time.
3699 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
3701 TF/08 TLS error reporting now respects the incoming_interface and
3702 incoming_port log selectors.
3704 TF/09 Produce a more useful error message if an SMTP transport's hosts
3705 setting expands to an empty string.
3707 NM/06 Bugzilla 744: EXPN did not work under TLS.
3708 Patch provided by Phil Pennock.
3710 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
3711 Patch provided by Richard Godbee.
3713 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
3714 acl_smtp_notquit, added index entry.
3716 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
3717 Patch provided by Eugene Bujak.
3719 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
3720 accept(). Patch provided by Maxim Dounin.
3722 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
3723 Patch provided by Phil Pennock.
3725 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
3727 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
3728 Patch provided by Brad "anomie" Jorsch.
3730 NM/14 Bugzilla 622: Added timeout setting to transport filter.
3731 Patch provided by Dean Brooks.
3733 TK/05 Add native DKIM support (does not depend on external libraries).
3735 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
3736 Patch provided by Graeme Fowler.
3738 NM/16 Bugzilla 851: Documentation example syntax fix.
3740 NM/17 Changed NOTICE file to remove references to embedded PCRE.
3742 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
3745 NM/19 Bugzilla 745: TLS version reporting.
3746 Patch provided by Phil Pennock.
3748 NM/20 Bugzilla 167: bool: condition support.
3749 Patch provided by Phil Pennock.
3751 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
3752 clients. Patch provided by Phil Pennock.
3754 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
3755 Patch provided by Brad "anomie" Jorsch.
3757 NM/23 Bugzilla 687: Fix misparses in eximstats.
3758 Patch provided by Heiko Schlittermann.
3760 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
3761 Patch provided by Heiko Schlittermann.
3763 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
3764 plus update to original patch.
3766 NM/26 Bugzilla 799: Documentation correction for ratelimit.
3768 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
3769 Patch provided by David Brownlee.
3771 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
3773 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
3775 NM/30 Bugzilla 888: TLS documentation bugfixes.
3777 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
3779 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
3780 Unlike the original bugzilla I have changed all shell scripts in src tree.
3782 NM/33 Bugzilla 898: Transport filter timeout fix.
3783 Patch by Todd Rinaldo.
3785 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
3786 Patch by Serge Demonchaux.
3788 NM/35 Bugzilla 39: Base64 decode bug fixes.
3789 Patch by Jakob Hirsch.
3791 NM/36 Bugzilla 909: Correct connect() call in dcc code.
3793 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
3795 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
3797 NM/39 Bugzilla 911: Fixed MakeLinks build script.
3803 TK/01 Add preliminary DKIM support. Currently requires a forked version of
3804 ALT-N's libdkim that I have put here:
3805 http://duncanthrax.net/exim-experimental/
3807 Note to Michael Haardt: I had to rename some vars in sieve.c. They
3808 were called 'true' and it seems that C99 defines that as a reserved
3809 keyword to be used with 'bool' variable types. That means you could
3810 not include C99-style headers which use bools without triggering
3811 build errors in sieve.c.
3813 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
3814 as mailq or other aliases. Changed the --help handling significantly
3815 to do whats expected. exim_usage() emits usage/help information.
3817 SC/01 Added the -bylocaldomain option to eximstats.
3819 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
3821 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
3823 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
3829 PH/01 Another patch from the Sieve maintainer.
3831 PH/02 When an IPv6 address is converted to a string for single-key lookup
3832 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
3833 dots are used instead of colons so that keys in lsearch files need not
3834 contain colons. This was done some time before quoting was made available
3835 in lsearch files. However, iplsearch files do require colons in IPv6 keys
3836 (notated using the quote facility) so as to distinguish them from IPv4
3837 keys. This meant that lookups for IP addresses in host lists did not work
3838 for iplsearch lookups.
3840 This has been fixed by arranging for IPv6 addresses to be expressed with
3841 colons if the lookup type is iplsearch. This is not incompatible, because
3842 previously such lookups could never work.
3844 The situation is now rather anomalous, since one *can* have colons in
3845 ordinary lsearch keys. However, making the change in all cases is
3846 incompatible and would probably break a number of configurations.
3848 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
3851 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
3852 conversion specification without a maximum field width, thereby enabling
3853 a rogue spamd server to cause a buffer overflow. While nobody in their
3854 right mind would setup Exim to query an untrusted spamd server, an
3855 attacker that gains access to a server running spamd could potentially
3856 exploit this vulnerability to run arbitrary code as the Exim user.
3858 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
3859 $primary_hostname instead of what libspf2 thinks the hosts name is.
3861 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
3862 a directory entry by the name of the lookup key. Previously, if a
3863 symlink pointed to a non-existing file or a file in a directory that
3864 Exim lacked permissions to read, a lookup for a key matching that
3865 symlink would fail. Now it is enough that a matching directory entry
3866 exists, symlink or not. (Bugzilla 503.)
3868 PH/03 The body_linecount and body_zerocount variables are now exported in the
3871 PH/04 Added the $dnslist_matched variable.
3873 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
3874 This means they are set thereafter only if the connection becomes
3877 PH/06 Added the client_condition to authenticators so that some can be skipped
3878 by clients under certain conditions.
3880 PH/07 The error message for a badly-placed control=no_multiline_responses left
3881 "_responses" off the end of the name.
3883 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
3885 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
3886 (without spaces) instead of just copying the configuration text.
3888 PH/10 Added the /noupdate option to the ratelimit ACL condition.
3890 PH/11 Added $max_received_linelength.
3892 PH/12 Added +ignore_defer and +include_defer to host lists.
3894 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
3895 way in which PCRE > 7.0 is built.
3897 PH/14 Implemented queue_only_load_latch.
3899 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
3900 MAIL command. The effect was to mangle the value on 64-bit systems.
3902 PH/16 Another patch from the Sieve maintainer.
3904 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
3906 PH/18 If a system quota error occurred while trying to create the file for
3907 a maildir delivery, the message "Mailbox is full" was not appended to the
3908 bounce if the delivery eventually timed out. Change 4.67/27 below applied
3909 only to a quota excession during the actual writing of the file.
3911 PH/19 It seems that peer DN values may contain newlines (and other non-printing
3912 characters?) which causes problems in log lines. The DN values are now
3913 passed through string_printing() before being added to log lines.
3915 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
3916 and InterBase are left for another time.)
3918 PH/21 Added message_body_newlines option.
3920 PH/22 Guard against possible overflow in moan_check_errorcopy().
3922 PH/23 POSIX allows open() to be a macro; guard against that.
3924 PH/24 If the recipient of an error message contained an @ in the local part
3925 (suitably quoted, of course), incorrect values were put in $domain and
3926 $local_part during the evaluation of errors_copy.
3932 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
3933 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
3936 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
3937 issue a MAIL command.
3939 PH/02 In an ACL statement such as
3941 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
3943 if a client was not listed at all, or was listed with a value other than
3944 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
3945 the condition was not true (as it should be), so access was not denied.
3946 The bug was that the ! inversion was incorrectly passed on to the second
3947 item. This has been fixed.
3949 PH/03 Added additional dnslists conditions == and =& which are different from
3950 = and & when the dns lookup returns more than one IP address.
3952 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
3953 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
3955 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
3956 FSYNC, which compiles an option called disable_fsync that allows for
3957 bypassing fsync(). The documentation is heavily laced with warnings.
3959 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
3961 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
3962 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
3963 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
3964 including adding "make clean"; (3) Added -fPIC when compiling the test
3965 dynamically loaded module, to get rid of a warning.
3967 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
3968 message fails, move_frozen_messages = true and ignore_bounce_errors_after
3969 = 0s. The bug is otherwise harmless.
3971 PH/07 There was a bug in the dovecot authenticator such that the value of
3972 $auth1 could be overwritten, and so not correctly preserved, after a
3973 successful authentication. This usually meant that the value preserved by
3974 the server_setid option was incorrect.
3976 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
3978 PH/09 Installed PCRE release 7.0.
3980 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
3981 run for batched SMTP input. It is now run at the start of every message
3982 in the batch. While fixing this I discovered that the process information
3983 (output by running exiwhat) was not always getting set for -bs and -bS
3984 input. This is fixed, and it now also says "batched" for BSMTP.
3986 PH/11 Added control=no_pipelining.
3988 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
3989 patch, slightly modified), and move the expansion of helo_data till after
3990 the connection is made in the smtp transport (so it can use these
3993 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
3995 PH/14 Added log_selector = +pid.
3997 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
3999 PH/16 Add ${if forany and ${if forall.
4001 PH/17 Added dsn_from option to vary the From: line in DSNs.
4003 PH/18 Flush SMTP output before performing a callout, unless control =
4004 no_callout_flush is set.
4006 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
4007 was true (the default) a successful delivery failed to delete the retry
4008 item, thus causing premature timeout of the address. The bug is now
4011 PH/20 Added hosts_avoid_pipelining to the smtp transport.
4013 PH/21 Long custom messages for fakedefer and fakereject are now split up
4014 into multiline responses in the same way that messages for "deny" and
4015 other ACL rejections are.
4017 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
4018 with slight modification.
4020 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
4021 draft, changing the syntax and factoring some duplicate code".
4023 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
4024 for deliveries of the second and subsequent messages over the same SMTP
4027 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
4028 ${reduce, with only minor "tidies".
4030 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
4032 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
4033 expansion side effects.
4035 PH/27 When a message times out after an over-quota error from an Exim-imposed
4036 quota, the bounce message says "mailbox is full". This message was not
4037 being given when it was a system quota that was exceeded. It now should
4040 MH/03 Made $recipients available in local_scan(). local_scan() already has
4041 better access to the recipient list through recipients_list[], but
4042 $recipients can be useful in postmaster-provided expansion strings.
4044 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
4045 in the case of a MAIL command with additional options following the
4046 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
4047 were accidentally chopped off.
4049 PH/29 SMTP synchronization checks are implemented when a command is read -
4050 there is a check that no more input is waiting when there shouldn't be
4051 any. However, for some commands, a delay in an ACL can mean that it is
4052 some time before the response is written. In this time, more input might
4053 arrive, invalidly. So now there are extra checks after an ACL has run for
4054 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
4055 pipelining has not been advertised.
4057 PH/30 MH's patch to allow iscntrl() characters to be list separators.
4059 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
4060 returned in the SMTP response when smtp_return_error_details was false.
4061 This has been fixed.
4063 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
4064 instead of the C I/O that was originally supplied, because problems were
4065 reported on Solaris.
4067 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
4068 Exim which did not show up earlier: it was assuming that a call to
4069 SSL_CTX_set_info_callback() might give an error value. In fact, there is
4070 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
4071 was a macro that became an assignment, so it seemed to work. This has
4072 changed to a proper function call with a void return, hence the compile
4073 error. Exim's code has been fixed.
4075 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
4078 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
4080 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
4082 PH/37 If a message is not accepted after it has had an id assigned (e.g.
4083 because it turns out to be too big or there is a timeout) there is no
4084 "Completed" line in the log. When some messages of this type were
4085 selected by exigrep, they were listed as "not completed". Others were
4086 picked up by some special patterns. I have improved the selection
4087 criteria to be more general.
4089 PH/38 The host_find_failed option in the manualroute router can now be set
4090 to "ignore", to completely ignore a host whose IP address cannot be
4091 found. If all hosts are ignored, the behaviour is controlled by the new
4092 host_all_ignored option.
4094 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
4095 homing or because of multiple MX records with /mx) generated more than
4096 one IP address, and the following item turned out to be the local host,
4097 all the secondary addresses of the first item were incorrectly removed
4098 from the list, along with the local host and any following hosts (which
4099 is what is supposed to happen).
4101 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
4102 whoever called Exim into the -H file. In the case of the daemon it was
4103 behaving confusingly. When first started, it used values for whoever
4104 started the daemon, but after a SIGHUP it used the Exim user (because it
4105 calls itself on a restart). I have changed the code so that it now always
4108 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
4109 message are rejected with the same error (e.g. no authentication or bad
4110 sender address), and a DATA command is nevertheless sent (as can happen
4111 with PIPELINING or a stupid MUA), the error message that was given to the
4112 RCPT commands is included in the rejection of the DATA command. This is
4113 intended to be helpful for MUAs that show only the final error to their
4116 PH/42 Another patch from the Sieve maintainer.
4118 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
4119 Eximstats - Fixed some broken HTML links and added missing column headers
4121 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
4122 columns for Rejects, Temp Rejects, Ham, and Spam rows.
4124 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
4126 PH/43 Yet another patch from the Sieve maintainer.
4128 PH/44 I found a way to check for a TCP/IP connection going away before sending
4129 the response to the final '.' that terminates a message, but only in the
4130 case where the client has not sent further data following the '.'
4131 (unfortunately, this is allowed). However, in many cases there won't be
4132 any further data because there won't be any more messages to send. A call
4133 to select() can be used: if it shows that the input is "ready", there is
4134 either input waiting, or the socket has been closed. An attempt to read
4135 the next input character can distinguish the two cases. Previously, Exim
4136 would have sent an OK response which the client would never have see.
4137 This could lead to message repetition. This fix should cure that, at
4138 least in a lot of common cases.
4140 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
4141 advertised in response to EHLO.
4147 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
4148 fixed by 4.65/MH/01 (is this a record?) are fixed:
4150 (i) An empty string was always treated as zero by the numeric comparison
4151 operators. This behaviour has been restored.
4153 (ii) It is documented that the numeric comparison operators always treat
4154 their arguments as decimal numbers. This was broken in that numbers
4155 starting with 0 were being interpreted as octal.
4157 While fixing these problems I realized that there was another issue that
4158 hadn't been noticed. Values of message_size_limit (both the global option
4159 and the transport option) were treated as octal if they started with 0.
4160 The documentation was vague. These values are now always treated as
4161 decimal, and I will make that clear in the documentation.
4167 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
4168 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
4171 MH/01 Don't check that the operands of numeric comparison operators are
4172 integers when their expansion is in "skipping" mode (fixes bug
4173 introduced by 4.64-PH/07).
4175 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
4176 child addresses, Exim now panics and dies. Previously, because the count
4177 is held in a short int, deliveries were likely to be lost. As such a
4178 large number of recipients for a single message is ridiculous
4179 (performance will be very, very poor), I have chosen to impose a limit
4180 rather than extend the field.
4186 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
4187 leftover -K file (the existence of which was triggered by #402).
4188 While we were at it, introduced process PID as part of the -K
4189 filename. This should rule out race conditions when creating
4192 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
4193 processing considerably. Previous code took too long for large mails,
4194 triggering a timeout which in turn triggers #401.
4196 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
4197 in the DK code in transports.c. sendfile() is not really portable,
4198 hence the _LINUX specificness.
4200 TF/01 In the add_headers option to the mail command in an Exim filter,
4201 there was a bug that Exim would claim a syntax error in any
4202 header after the first one which had an odd number of characters
4205 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
4206 callout verification, Exim cached a "reject" for the entire domain. This
4207 is correct for most verifications, but it is not correct for a recipient
4208 verification with use_sender or use_postmaster set, because in that case
4209 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
4210 case of MAIL FROM:<> rejection from other early rejections (e.g.
4211 rejection of HELO). When verifying a recipient using a non-null MAIL
4212 address, the cache is ignored if it shows MAIL FROM:<> rejection.
4213 Whatever the result of the callout, the value of the domain cache is
4214 left unchanged (for any other kind of callout, getting as far as trying
4215 RCPT means that the domain itself is ok).
4217 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
4220 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
4221 manifest itself as EPIPE rather than ECONNECT. When tidying away a
4222 session, the daemon ignores ECONNECT errors and logs others; it now
4223 ignores EPIPE as well.
4225 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
4226 (quoted-printable decoding).
4228 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
4229 later the small subsequent patch to fix an introduced bug.
4231 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
4233 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
4235 PH/08 An error is now given if message_size_limit is specified negative.
4237 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
4238 to be given (somewhat) arbitrary names.
4240 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
4243 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
4244 miscellaneous code fixes
4246 PH/10 Added the log_reject_target ACL modifier to specify where to log
4249 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
4250 hostname. This is wrong, because it relates to the incoming message (and
4251 probably the interface on which it is arriving) and not to the outgoing
4252 callout (which could be using a different interface). This has been
4253 changed to use the value of the helo_data option from the smtp transport
4254 instead - this is what is used when a message is actually being sent. If
4255 there is no remote transport (possible with a router that sets up host
4256 addresses), $smtp_active_hostname is used.
4258 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
4259 tweaks were necessary in order to get it to work (see also 21 below):
4260 (a) The code assumed that strncpy() returns a negative number on buffer
4261 overflow, which isn't the case. Replaced with Exim's string_format()
4263 (b) There were several signed/unsigned issues. I just did the minimum
4264 hacking in of casts. There is scope for a larger refactoring.
4265 (c) The code used strcasecmp() which is not a standard C function.
4266 Replaced with Exim's strcmpic() function.
4267 (d) The code set only $1; it now sets $auth1 as well.
4268 (e) A simple test gave the error "authentication client didn't specify
4269 service in request". It would seem that Dovecot has changed its
4270 interface. Fortunately there's a specification; I followed it and
4271 changed what the client sends and it appears to be working now.
4273 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
4276 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
4277 address A is aliased to B and C, where B exists and C does not. Without
4278 -v the output is "A verified" because verification stops after a
4279 successful redirection if more than one address is generated. However,
4280 with -v the child addresses are also verified. Exim was outputting "A
4281 failed to verify" and then showing the successful verification for C,
4282 with its parentage. It now outputs "B failed to verify", showing B's
4283 parentage before showing the successful verification of C.
4285 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
4286 look up a TXT record in a specific list after matching in a combined
4289 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
4290 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
4291 they consult the DNS. I had assumed they would set it the way they
4292 wanted; and indeed my experiments on Linux seem to show that in some
4293 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
4294 To be on the safe side, however, I have now made the interface to
4295 host_find_byname() similar to host_find_bydns(), with an argument
4296 containing the DNS resolver options. The host_find_byname() function now
4297 sets these options at its start, just as host_find_bydns() does. The smtp
4298 transport options dns_qualify_single and dns_search_parents are passed to
4299 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
4300 of host_find_byname() use the default settings of RES_DEFNAMES
4301 (qualify_single) but not RES_DNSRCH (search_parents).
4303 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
4304 spool_read_header() do less string testing, by means of a preliminary
4305 switch on the second character of optional "-foo" lines. (This is
4306 overdue, caused by the large number of possibilities that now exist.
4307 Originally there were few.) While I was there, I also converted the
4308 str(n)cmp tests so they don't re-test the leading "-" and the first
4309 character, in the hope this might squeeze out yet more improvement.
4311 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
4312 flag allowing group syntax was set by the header_syntax check but not
4313 turned off, possible causing trouble later; (2) The flag was not being
4314 set at all for the header_verify test, causing "group"-style headers to
4315 be rejected. I have now set it in this case, and also caused header_
4316 verify to ignore an empty address taken from a group. While doing this, I
4317 came across some other cases where the code for allowing group syntax
4318 while scanning a header line wasn't quite right (mostly, not resetting
4319 the flag correctly in the right place). These bugs could have caused
4320 trouble for malformed header lines. I hope it is now all correct.
4322 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
4323 with the "reply" argument non-NULL. The code, however (which originally
4324 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
4325 but it didn't always do it. This confused somebody who was copying the
4326 code for some other use. I have removed all the tests.
4328 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
4329 feature that was used to support insecure browsers during the U.S. crypto
4330 embargo. It requires special client support, and Exim is probably the
4331 only MTA that supported it -- and would never use it because real RSA is
4332 always available. This code has been removed, because it had the bad
4333 effect of slowing Exim down by computing (never used) parameters for the
4334 RSA_EXPORT functionality.
4336 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
4337 authenticator to fail if there's a tab character in the incoming data
4338 (there should never be unless someone is messing about, as it's supposed
4339 to be base64-encoded). Also added, on Timo's advice, the "secured" option
4340 if the connection is using TLS or if the remote IP is the same as the
4341 local IP, and the "valid-client-cert option" if a client certificate has
4344 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
4345 authenticators. This can be used for authorization after authentication
4346 succeeds. (In the case of plaintext, it servers for both authentication
4349 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
4350 if any retry times were supplied.
4352 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
4353 connection, where there is no client IP address to check. In this
4354 situation, the verify now always succeeds.
4356 PH/25 Applied John Jetmore's -Mset patch.
4358 PH/26 Added -bem to be like -Mset, but loading a message from a file.
4360 PH/27 In a string expansion for a processed (not raw) header when multiple
4361 headers of the same name were present, leading whitespace was being
4362 removed from all of them, but trailing whitespace was being removed only
4363 from the last one. Now trailing whitespace is removed from each header
4364 before concatenation. Completely empty headers in a concatenation (as
4365 before) are ignored.
4367 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
4368 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
4370 PH/29 [Removed. This was a change that I later backed out, and forgot to
4371 correct the ChangeLog entry (that I had efficiently created) before
4372 committing the later change.]
4374 PH/30 Exim was sometimes attempting to deliver messages that had suffered
4375 address errors (4xx response to RCPT) over the same connection as other
4376 messages routed to the same hosts. Such deliveries are always "forced",
4377 so retry times are not inspected. This resulted in far too many retries
4378 for the affected addresses. The effect occurred only when there were more
4379 hosts than the hosts_max_try setting in the smtp transport when it had
4380 the 4xx errors. Those hosts that it had tried were not added to the list
4381 of hosts for which the message was waiting, so if all were tried, there
4382 was no problem. Two fixes have been applied:
4384 (i) If there are any address or message errors in an SMTP delivery, none
4385 of the hosts (tried or untried) are now added to the list of hosts
4386 for which the message is waiting, so the message should not be a
4387 candidate for sending over the same connection that was used for a
4388 successful delivery of some other message. This seems entirely
4389 reasonable: after all the message is NOT "waiting for some host".
4390 This is so "obvious" that I'm not sure why it wasn't done
4391 previously. Hope I haven't missed anything, but it can't do any
4392 harm, as the worst effect is to miss an optimization.
4394 (ii) If, despite (i), such a delivery is accidentally attempted, the
4395 routing retry time is respected, so at least it doesn't keep
4396 hammering the server.
4398 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
4399 in ${readsocket because some servers need this prod.
4401 PH/32 Added some extra debug output when updating a wait-xxx database.
4403 PH/33 The hint "could be header name not terminated by colon", which has been
4404 given for certain expansion errors for a long time, was not being given
4405 for the ${if def:h_colon_omitted{... case.
4407 PH/34 The spec says: "With one important exception, whenever a domain list is
4408 being scanned, $domain contains the subject domain." There was at least
4409 one case where this was not true.
4411 PH/35 The error "getsockname() failed: connection reset by peer" was being
4412 written to the panic log as well as the main log, but it isn't really
4413 panic-worthy as it just means the connection died rather early on. I have
4414 removed the panic log writing for the ECONNRESET error when getsockname()
4417 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
4418 runs only) independently of the message's sender address. This meant
4419 that, if the 4xx error was in fact related to the sender, a different
4420 message to the same recipient with a different sender could confuse
4421 things. In particular, this can happen when sending to a greylisting
4422 server, but other circumstances could also provoke similar problems.
4423 I have changed the default so that the retry time for these errors is now
4424 based a combination of the sender and recipient addresses. This change
4425 can be overridden by setting address_retry_include_sender=false in the
4428 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
4429 remote server are returned as part of bounce messages. This was not
4430 happening for LMTP over a pipe (the lmtp transport), but now it is the
4431 same for both kinds of LMTP.
4433 PH/38 Despite being documented as not happening, Exim was rewriting addresses
4434 in header lines that were in fact CNAMEs. This is no longer the case.
4436 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
4437 and queue runs started by the daemon processed all messages. This has
4438 been fixed so that -R and -S can now usefully be given with -q<time>.
4440 PH/40 Import PCRE release 6.7 (fixes some bugs).
4442 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
4444 PH/42 Give an error if -q is specified more than once.
4446 PH/43 Renamed the variables $interface_address and $interface_port as
4447 $received_ip_address and $received_port, to make it clear that these
4448 values apply to message reception, and not to the outgoing interface when
4449 a message is delivered. (The old names remain recognized, of course.)
4451 PH/44 There was no timeout on the connect() call when using a Unix domain
4452 socket in the ${readsocket expansion. There now is.
4454 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
4455 be meaningful with "accept".
4457 SC/01 Eximstats V1.43
4458 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
4460 SC/02 Eximstats V1.44
4461 Use a glob alias rather than an array ref in the generated
4462 parser. This improves both readability and performance.
4464 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
4465 Collect SpamAssassin and rejection statistics.
4466 Don't display local sender or destination tables unless
4467 there is data to show.
4468 Added average volumes into the top table text output.
4470 SC/04 Eximstats V1.46
4471 Collect data on the number of addresses (recipients)
4472 as well as the number of messages.
4474 SC/05 Eximstats V1.47
4475 Added 'Message too big' to the list of mail rejection
4476 reasons (thanks to Marco Gaiarin).
4478 SC/06 Eximstats V1.48
4479 Mainlog lines which have GMT offsets and are too short to
4480 have a flag are now skipped.
4482 SC/07 Eximstats V1.49 (Alain Williams)
4483 Added the -emptyok flag.
4485 SC/08 Eximstats V1.50
4486 Fixes for obtaining the IP address from reject messages.
4488 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
4489 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
4490 whitespace changes from 4.64-PH/27
4492 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
4495 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
4496 are found, allow negative numbers in numeric criteria)
4498 JJ/06 exipick.20061117.2, added new $message_body_missing variable
4500 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
4501 to match changes made in 4.64-PH/43
4503 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
4505 PH/47 Put in an explicit test for a DNS lookup of an address record where the
4506 "domain" is actually an IP address, and force a failure. This locks out
4507 those revolvers/nameservers that support "A-for-A" lookups, in
4508 contravention of the specifications.
4510 PH/48 When a host name was looked up from an IP address, and the subsequent
4511 forward lookup of the name timed out, the host name was left in
4512 $sender_host_name, contrary to the specification.
4514 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
4515 restricted to single-key lookups, Exim was not diagnosing an error if
4516 * or *@ was used with a query-style lookup.
4518 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
4520 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
4521 long ago, but noone interested enough thought of it. Let's just say that
4522 the "1.1" means that there are some new functions that weren't there at
4523 some point in the past.
4525 PH/51 Error processing for expansion failure of helo_data from an smtp
4526 transport during callout processing was broken.
4528 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
4529 tested/used via the -bh/-bhc/-bs options.
4531 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
4532 bug, fixed in subsequent PCRE releases).
4534 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
4535 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
4537 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
4543 SC/01 Use a glob alias rather than an array ref in eximstats generated
4544 parser. This improves both readability and performance.
4546 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
4547 Don't display local sender or destination tables in eximstats unless
4548 there is data to show.
4549 Added average volumes into the eximstats top table text output.
4551 SC/03 Collect data on the number of addresses (recipients) as well
4552 as the number of messages in eximstats.
4554 TF/01 Correct an error in the documentation for the redirect router. Exim
4555 does (usually) call initgroups() when daemonizing.
4557 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
4558 with consistent privilege compared to when running as a daemon.
4560 TF/03 Note in the spec that $authenticated_id is not set for local
4561 submissions from trusted users.
4563 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
4564 Thanks to Dean Brooks <dean@iglou.com> for the patch.
4566 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
4567 by adding some example configuration directives to the default
4568 configuration file. A little bit of work is required to uncomment the
4569 directives and define how usernames and passwords are checked, but
4570 there is now a framework to start from.
4572 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
4573 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
4574 without this. I don't know how relevant this is to other LDAP libraries.
4576 PH/02 Add the verb name to the "unknown ACL verb" error.
4578 PH/03 Magnus Holmgren's patch for filter_prepend_home.
4580 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
4582 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
4583 directory not expanded when it should be if an expanded home directory
4584 was set for the address (which is overridden by the transport).
4586 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
4589 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
4590 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
4591 because it is too late at that time, and has no effect.
4593 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
4594 security issue with \' (bugzilla #107). I could not use the
4595 PQescapeStringConn() function, because it needs a PGconn value as one of
4598 PH/08 When testing addresses using -bt, indicate those final addresses that
4599 are duplicates that would not cause an additional delivery. At least one
4600 person was confused, thinking that -bt output corresponded to deliveries.
4601 (Suppressing duplicates isn't a good idea as you lose the information
4602 about possibly different redirections that led to the duplicates.)
4604 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
4605 systems where poll() doesn't work, in particular OS X.
4607 PH/10 Added more information to debugging output for retry time not reached.
4609 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
4610 operations in malware.c.
4612 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
4615 PH/13 If write_rejectlog was set false when logging was sent to syslog with
4616 syslog_duplication set false, log lines that would normally be written
4617 both the the main log and to the reject log were not written to syslog at
4620 PH/14 In the default configuration, change the use of "message" in ACL warn
4621 statements to "add_header".
4623 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
4624 not followed by a command (e.g. "seen endif").
4626 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
4627 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
4630 PH/17 Added extra conditions to the default value of delay_warning_condition
4634 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
4635 { match{$h_precedence:}{(?i)bulk|list|junk} } \
4636 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
4639 The Auto-Submitted: and various List- headers are standardised, whereas I
4640 don't think Precedence: ever was.
4642 PH/18 Refactored debugging code in route_finduser() to show more information,
4643 in particular, the error code if getpwnam() issues one.
4645 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
4646 This is apparently needed in addition to the PH/07 change above to avoid
4647 any possible encoding problems.
4649 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
4650 but not after initializing Perl.
4652 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
4653 output them only if debugging. By default they are written stderr,
4654 apparently, which is not desirable.
4656 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
4659 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
4662 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
4664 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
4665 authenticated or an ident call has been made. Suppress the default
4666 values for $authenticated_id and $authenticated_sender (but permit -oMai
4667 and -oMas) when testing with -bh.
4669 PH/24 Re-jigged the order of the tests in the default configuration so that the
4670 tests for valid domains and recipients precede the DNS black list and CSA
4671 tests, on the grounds that those ones are more expensive.
4673 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
4674 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
4675 This bug exists in every version of Exim that I still have, right back to
4678 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
4679 However, an attempt to turn on case-sensitivity in a regex key by
4680 including (?-i) didn't work because the subject string was already
4681 lowercased, and the effects were non-intuitive. It turns out that a
4682 one-line patch can be used to allow (?-i) to work as expected.
4688 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
4689 other effects) broke the use of negated acl sub-conditions.
4691 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
4694 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
4695 "Deny" causes Exim to reject the incoming connection with a 554 error.
4696 Unfortunately, if there is a major crisis, such as a disk failure,
4697 tcp-wrappers gives "deny", whereas what one would like would be some
4698 kind of temporary error. A kludge has been added to help with this.
4699 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
4700 554 error is used if errno is still zero or contains ENOENT (which occurs
4701 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
4704 PH/03 Add -lutil to the default FreeBSD LIBS setting.
4706 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
4707 errors. Otherwise a message that provokes a temporary error (when other
4708 messages do not) can cause a whole host to time out.
4710 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
4711 addresses were routed directly to files or pipes from a redirect router.
4712 File deliveries just didn't batch; pipe deliveries might have suffered
4715 PH/06 A failure to get a lock for a hints database would erroneously always say
4716 "Failed to get write lock", even when it was really a read lock.
4718 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
4719 of 0600. This has been changed to use the value of the lockfile_mode
4720 option (which defaults to 0600).
4722 PH/08 Applied small patch from the Sieve maintainer.
4724 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
4725 folder from quota calculations, a direct delivery into this folder messed
4726 up the contents of the maildirsize file. This was because the regex was
4727 used only to exclude .Trash (or whatever) when the size of the mailbox
4728 was calculated. There was no check that a delivery was happening into an
4729 excluded directory. This bug has been fixed by ignoring all quota
4730 processing for deliveries into excluded directories.
4732 PH/10 Added the maildirfolder_create_regex option to appendfile.
4738 PH/01 The code for finding all the local interface addresses on a FreeBSD
4739 system running IPv6 was broken. This may well have applied to all BSD
4740 systems, as well as to others that have similar system calls. The broken
4741 code found IPv4 interfaces correctly, but gave incorrect values for the
4742 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
4743 that it would not match correctly against @[] and not recognize the IPv6
4746 PH/02 The ipliteral router was not recognizing addresses of the form user@
4747 [ipv6:....] because it didn't know about the "ipv6:" prefix.
4749 PH/03 Added disable_ipv6.
4751 PH/04 Changed $reply_address to use the raw form of the headers instead of the
4752 decoded form, because it is most often used to construct To: headers
4753 lines in autoreplies, and the decoded form may well be syntactically
4754 invalid. However, $reply_address has leading white space removed, and all
4755 newlines turned into spaces so that the autoreply transport does not
4758 PH/05 If group was specified without a user on a router, and no group or user
4759 was specified on a transport, the group from the router was ignored.
4761 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
4762 for visible compile-time settings that can be used to change these
4763 numbers, for those that want even more. Backwards compatibility with old
4764 spool files has been maintained. However, going back to a previous Exim
4765 release will lost any variables that are in spool files.
4767 PH/07 Two small changes when running in the test harness: increase delay when
4768 passing a TCP/IP connection to a new process, in case the original
4769 process has to generate a bounce, and remove special handling of
4770 127.0.0.2 (sic), which is no longer necessary.
4772 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
4773 be the same on different OS.
4775 PH/09 Moved a debug statement in filter processing to avoid a race problem when
4778 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
4779 whether --show-vars was specified or not
4781 JJ/02 exipick: Added support for new ACL variable spool format introduced
4784 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
4785 syntactically invalid From: or Reply-to: line, and a filter used this to
4786 generate an autoreply, and therefore failed to obtain an address for the
4787 autoreply, Exim could try to deliver to a non-existent relative file
4788 name, causing unrelated and misleading errors. What now happens is that
4789 it logs this as a hard delivery error, but does not attempt to create a
4792 PH/11 The exinext utility has a -C option for testing purposes, but although
4793 the given file was scanned by exinext itself; it wasn't being passed on
4794 when Exim was called.
4796 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
4797 an end-of-file indication when reading a command response.
4799 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
4800 compiled. In many other places in Exim, IPv6 addresses are always
4801 recognized, so I have changed this. It also means that IPv4 domain
4802 literals of the form [IPV4:n.n.n.n] are now always recognized.
4804 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
4805 used if the router is not running as root, for example, when verifying at
4806 ACL time, or when using -bh. The debugging output from this situation was
4807 non-existent - all you got was a failure to exec. I have made two
4810 (a) Failures to set uid/gid, the current directory, or a process leader
4811 in a subprocess such as that created by queryprogram now generate
4812 suitable debugging output when -d is set.
4814 (b) The queryprogram router detects when it is not running as root,
4815 outputs suitable debugging information if -d is set, and then runs
4816 the subprocess without attempting to change uid/gid.
4818 PH/15 Minor change to Makefile for building test_host (undocumented testing
4821 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
4822 additional section of a DNS packet that returns MX or SRV records.
4823 Instead, it always explicitly searches for A/AAAA records. This avoids
4824 major problems that occur when a DNS server includes only records of one
4825 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
4826 fixed another bug: if SRV records were looked up and the corresponding
4827 address records were *not* found in the additional section, the port
4828 values from the SRV records were lost.
4830 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
4831 using the correct key (the original address) when searching the retry
4832 rules in order to find which one to use for generating the retry hint.
4834 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
4835 adding the default one. Similarly, if it contains a Reply-To: header, the
4836 errors_reply_to option, if set, is not used.
4838 PH/19 When calculating a retry time, Exim used to measure the "time since
4839 failure" by looking at the "first failed" field in the retry record. Now
4840 it does not use this if it is later than than the arrival time of the
4841 message. Instead it uses the arrival time. This makes for better
4842 behaviour in cases where some deliveries succeed, thus re-setting the
4843 "first failed" field. An example is a quota failure for a huge message
4844 when small messages continue to be delivered. Without this change, the
4845 "time since failure" will always be short, possible causing more frequent
4846 delivery attempts for the huge message than are intended.
4847 [Note: This change was subsequently modified - see PH/04 for 4.62.]
4849 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
4850 $1, $2, $3) because the numerical variables can be reset during some
4851 expansion items (e.g. "match"), thereby losing the authentication data.
4853 PH/21 Make -bV show the size of off_t variables so that the test suite can
4854 decide whether to run tests for quotas > 2G.
4856 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
4857 mailbox_size, and mailbox_filecount in the appendfile transport. If a
4858 filecount value is greater than 2G or if a quota value is greater than 2G
4859 on a system where the size of off_t is not greater than 4, a panic error
4862 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
4863 never match. The debug and -bh output now contains an explicit error
4864 message indicating a malformed IPv4 address or mask.
4866 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
4867 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
4868 PH/23 above applies.
4870 PH/25 Do not write to syslog when running in the test harness. The only
4871 occasion when this arises is a failure to open the main or panic logs
4872 (for which there is an explicit test).
4874 PH/26 Added the /no_tell option to "control=freeze".
4876 PH/27 If a host name lookup failed very early in a connection, for example, if
4877 the IP address matched host_lookup and the reverse lookup yielded a name
4878 that did not have a forward lookup, an error message of the form "no IP
4879 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
4880 could be logged. Now it outputs the IP address instead of "NULL".
4882 PH/28 An enabling patch from MH: add new function child_open_exim2() which
4883 allows the sender and the authenticated sender to be set when
4884 submitting a message from within Exim. Since child_open_exim() is
4885 documented for local_scan(), the new function should be too.
4887 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
4888 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
4889 results in an empty string is now treated as unset.
4891 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
4893 PH/31 Added sender_verify_fail logging option.
4895 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
4896 needed by RFC 822 but not by RFC 2822 was commented out. I have now
4897 tidied the source and removed it altogether.
4899 PH/33 When a queue run was abandoned because the load average was too high, a
4900 log line was always written; now it is written only if the queue_run log
4901 selector is set. In addition, the log line for abandonment now contains
4902 information about the queue run such as the pid. This is always present
4903 in "start" and "stop" lines but was omitted from the "abandon" line.
4905 PH/34 Omit spaces between a header name and the colon in the error message that
4906 is given when verify = headers_syntax fails (if there are lots of them,
4907 the message gets confusing).
4909 PH/35 Change the default for dns_check_names_pattern to allow slashes within
4910 names, as there are now some PTR records that contain slashes. This check
4911 is only to protect against broken name servers that fall over on strange
4912 characters, so the fact that it applies to all lookups doesn't matter.
4914 PH/36 Now that the new test suite is complete, we can remove some of the
4915 special code in Exim that was needed for the old test suite. For example,
4916 sorting DNS records because real resolvers return them in an arbitrary
4917 order. The new test suite's fake resolver always returns records in the
4920 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
4921 bounces) except when queue_only is set, to avoid logging races between
4922 the different processes.
4924 PH/38 Panic-die if .include specifies a non-absolute path.
4926 PH/39 A tweak to the "H" retry rule from its user.
4928 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
4929 a label. They prevented compilation on older perls.
4931 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
4932 a warning to be raised on newish perls.
4934 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
4935 on queue. Changes to match documented behaviour of showing count of
4936 messages matching specified criteria.
4938 PH/40 Changed the default ident timeout from 30s to 5s.
4940 PH/41 Added support for the use of login_cap features, on those BSD systems
4941 that have them, for controlling the resources used by pipe deliveries.
4943 PH/42 The content-scanning code uses fopen() to create files in which to put
4944 message data. Previously it was not paying any attention to the mode of
4945 the files. Exim runs with umask(0) because the rest of the code creates
4946 files with open(), and sets the required mode explicitly. Thus, these
4947 files were ending up world-writeable. This was not a big issue, because,
4948 being within the spool directory, they were not world-accessible. I have
4949 created a function called modefopen, which takes an additional mode
4950 argument. It sets umask(777), creates the file, chmods it to the required
4951 mode, then resets the umask. All the relevant calls to fopen() in the
4952 content scanning code have been changed to use this function.
4954 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
4955 to 24 hours. This avoids potential overflow problems when processing G
4956 and H retry rules. I suspect nobody ever tinkers with this value.
4958 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
4960 PH/45 When the plaintext authenticator is running as a client, the server's
4961 challenges are checked to ensure they are valid base64 strings. By
4962 default, the authentication attempt is cancelled if an invalid string is
4963 received. Setting client_ignore_invalid_base64 true ignores these errors.
4964 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
4965 they are received. Thus, the responses can be made to depend on the
4966 challenges. If an invalid string is ignored, an empty string is placed in
4969 PH/46 Messages that are created by the autoreply transport now contains a
4970 References: header, in accordance with RFCs 2822 and 3834.
4972 PH/47 Added authenticated_sender_force to the smtp transport.
4974 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
4976 PH/49 Installed latest patch from the Sieve maintainer.
4978 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
4979 was also set, the appendfile transport was unnecessarily scanning a
4980 directory of message files (e.g. for maildir delivery) to find the count
4981 of files (along with the size), even though it did not need this
4982 information. It now does the scan only if it needs to find either the
4983 size of the count of files.
4985 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
4987 PH/52 Two bugs concerned with error handling when the smtp transport is
4990 (i) Exim was not creating retry information for temporary errors given
4991 for individual recipients after the DATA command when the smtp transport
4992 was used in LMTP mode. This meant that they could be retried too
4993 frequently, and not timed out correctly.
4995 (ii) Exim was setting the flag that allows error details to be returned
4996 for LMTP errors on RCPT commands, but not for LMTP errors for individual
4997 recipients that were returned after the DATA command.
4999 PH/53 This is related to PH/52, but is more general: for any failing address,
5000 when detailed error information was permitted to be returned to the
5001 sender, but the error was temporary, then after the final timeout, only
5002 "retry timeout exceeded" was returned. Now it returns the full error as
5003 well as "retry timeout exceeded".
5005 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
5006 do this, and (what is worse) MTAs that accept it.
5008 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
5009 will now be deprecated.
5011 PH/56 New os.c-cygwin from the Cygwin maintainer.
5013 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
5014 formats (previously only available in exim formats via -bpr, -bpru,
5015 and -bpra. Now also available in native and exiqgrep formats)
5017 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
5018 with very large, slow to parse queues
5020 JJ/08 exipick: added ! as generic prefix to negate any criteria format
5022 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
5024 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
5025 responses to authentication challenges, though it was showing the
5026 challenges; (ii) I've removed the CR characters from the debug output for
5029 PH/58 Allow for the insertion of a newline as well as a space when a string
5030 is turned into more than one encoded-word during RFC 2047 encoding. The
5031 Sieve code now uses this.
5033 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
5034 data_4xx, lost_connection, tls_required.
5036 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
5037 message was being sent as an SMTP response.
5039 PH/61 Add -l and -k options to exicyclog.
5041 PH/62 When verifying, if an address was redirected to one new address, so that
5042 verification continued, and the new address failed or deferred after
5043 having set something in $address_data, the value of $address_data was not
5044 passed back to the ACL. This was different to the case when no
5045 redirection occurred. The value is now passed back in both cases.
5047 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
5048 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
5049 use login_cap.h, so on its own it isn't the distinguishing feature. The
5050 new name refers directly to the setclassresources() function.
5052 PH/65 Added configuration files for NetBSD3.
5054 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
5056 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
5057 is preferred over IPv4.
5059 PH/68 The bounce_return_message and bounce_return_body options were not being
5060 honoured for bounces generated during the reception of non-SMTP messages.
5061 In particular, this applied to messages rejected by the ACL. This bug has
5062 been fixed. However, if bounce_return_message is true and bounce_return_
5063 body is false, the headers that are returned for a non-SMTP message
5064 include only those that have been read before the error was detected.
5065 (In the case of an ACL rejection, they have all been read.)
5067 PH/69 The HTML version of the specification is now built in a directory called
5068 spec_html instead of spec.html, because the latter looks like a path with
5069 a MIME-type, and this confuses some software.
5071 PH/70 Catch two compiler warnings in sieve.c.
5073 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
5074 function verify_get_ident() calls ip_connect() to connect a socket, but
5075 if the "connect()" function timed out, ip_connect() used to close the
5076 socket. However, verify_get_ident() also closes the socket later, and in
5077 between Exim writes to the log, which may get opened at this point. When
5078 the socket was closed in ip_connect(), the log could get the same file
5079 descriptor number as the socket. This naturally causes chaos. The fix is
5080 not to close the socket in ip_connect(); the socket should be closed by
5081 the function that creates it. There was only one place in the code where
5082 this was missing, in the iplookup router, which I don't think anybody now
5083 uses, but I've fixed it anyway.
5085 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
5086 well as to direct DNS lookups. Otherwise the handling of names in host
5087 lists is inconsistent and therefore confusing.
5093 PH/01 Two changes to the default runtime configuration:
5095 (1) Move the checks for relay_from_hosts and authenticated clients from
5096 after to before the (commented out) DNS black list checks.
5098 (2) Add control=submission to the relay_from_hosts and authenticated
5099 clients checks, on the grounds that messages accepted by these
5100 statements are most likely to be submissions.
5102 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
5104 (1) Generate an error if the third argument for the ${prvs expansion is
5107 (2) Treat a missing third argument of ${prvscheck as if it were an empty
5110 (3) Reset the variables that are obtained from the first argument of
5111 ${prvscheck and used in the second argument before leaving the code,
5112 because their memory is reclaimed, so using them afterwards may do
5115 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
5116 one (it's much easier than Tom thought :-).
5118 (5) Because of (4), we can now allow for the use of $prvscheck_result
5119 inside the third argument.
5121 PH/03 For some reason, the default setting of PATH when running a command from
5122 a pipe transport was just "/usr/bin". I have changed it to
5125 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
5126 anything to be listed in the output from -bV.
5128 PH/05 When a filter generated an autoreply, the entire To: header line was
5129 quoted in the delivery log line, like this:
5131 => >A.N.Other <ano@some.domain> <original@ddress> ...
5133 This has been changed so that it extracts the operative address. There
5134 may be more than one such address. If so, they are comma-separated, like
5137 => >ano@some.domain,ona@other.domain <original@ddress> ...
5139 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
5140 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
5141 not being looked up in the rDNS to get a host name, Exim was showing the
5142 IP address twice in Received: lines, even though the IP addresses were
5143 identical. For example:
5145 Received: from [1.2.3.4] (helo=[1.2.3.4])
5147 However, if the real host name was known, it was omitting the HELO data
5148 if it matched the actual IP address. This has been tidied up so that it
5149 doesn't show the same IP address twice.
5151 PH/07 When both +timestamp and +memory debugging was on, the value given by
5152 $tod_xxx expansions could be wrong, because the tod_stamp() function was
5153 called by the debug printing, thereby overwriting the timestamp buffer.
5154 Debugging no longer uses the tod_stamp() function when +timestamp is set.
5156 PH/08 When the original message was included in an autoreply transport, it
5157 always said "this is a copy of the message, including all the headers",
5158 even if body_only or headers_only was set. It now gives an appropriate
5161 PH/09 Applied a patch from the Sieve maintainer which:
5163 o fixes some comments
5164 o adds the (disabled) notify extension core
5165 o adds some debug output for the result of if/elsif tests
5166 o points to the current vacation draft in the documentation
5167 and documents the missing references header update
5171 o fixes a bug in processing the envelope test (when testing
5172 multiple envelope elements, the last element determined the
5175 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
5176 Electronic Mail") by including:
5178 Auto-submitted: auto-generated
5180 in the messages that it generates (bounce messages and others, such as
5181 warnings). In the case of bounce messages for non-SMTP messages, there was
5182 also a typo: it was using "Auto_submitted" (underscore instead of
5183 hyphen). Since every message generated by Exim is necessarily in response
5184 to another message, thes have all been changed to:
5186 Auto-Submitted: auto-replied
5188 in accordance with these statements in the RFC:
5190 The auto-replied keyword:
5192 - SHOULD be used on messages sent in direct response to another
5193 message by an automatic process,
5195 - MUST NOT be used on manually-generated messages,
5197 - MAY be used on Delivery Status Notifications (DSNs) and Message
5198 Disposition Notifications (MDNs),
5200 - MUST NOT be used on messages generated by automatic or periodic
5201 processes, except for messages which are automatic responses to
5204 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
5205 to the default Received: header definition.
5207 PH/12 Added log selector acl_warn_skipped (default on).
5209 PH/13 After a successful wildlsearch lookup, discard the values of numeric
5210 variables because (a) they are in the wrong storage pool and (b) even if
5211 they were copied, it wouldn't work properly because of the caching.
5213 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
5214 checking when decoding. Apparently there are clients that generate
5215 overlong encoded strings. Why am I not surprised?
5217 PH/15 If the first argument of "${if match_address" was not empty, but did not
5218 contain an "@" character, Exim crashed. Now it writes a panic log message
5219 and treats the condition as false.
5221 PH/16 In autoreply, treat an empty string for "once" the same as unset.
5223 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
5224 extension "envelope-auth". The code is finished and in agreement with
5225 other implementations, but there is no documentation so far and in fact,
5226 nobody wrote the draft yet. This extension is currently #undef'ed, thus
5227 not changing the active code.
5229 Print executed "if" and "elsif" statements when debugging is used. This
5230 helps a great deal to understand what a filter does.
5232 Document more things not specified clearly in RFC3028. I had all this
5233 sorted out, when out of a sudden new issues came to my mind. Oops."
5235 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
5238 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
5239 canonical form (as documented). However, after a host name lookup from
5240 the IP address, check_host() was doing a simple string comparison with
5241 addresses acquired from the DNS when checking that the found name did
5242 have the original IP as one of its addresses. Since any found IPv6
5243 addresses are likely to be in abbreviated form, the comparison could
5244 fail. Luckily, there already exists a function for doing the comparison
5245 by converting both addresses to binary, so now that is used instead of
5246 the text comparison.
5248 PH/20 There was another similar case to PH/19, when a complete host name was
5249 given in a host list; looking up its IP address could give an abbreviated
5250 form, whereas the current host's name might or might not be abbreviated.
5251 The same fix has been applied.
5257 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
5258 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
5261 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
5262 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
5264 PH/03 Typo: missing ".o" in src/pcre/Makefile.
5266 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
5267 header line, restrict the check to what is listed in RFCs 2369 and 2929.
5268 Also, for "Auto-Submitted", treat anything other than "no" as
5269 non-personal, in accordance with RFC 3834. (Previously it treated
5270 anything starting "auto-" as non-personal.)
5272 TF/01 The control=submission/name=... option had a problem with syntax
5273 errors if the name included a slash character. The /name= option
5274 now slurps the rest of the string, so it can include any characters
5275 but it must come last in the list of options (after /sender_retain
5278 PH/05 Some modifications to the interface to the fake nameserver for the new
5286 TK/01 Added the "success_on_redirect" address verification option. See
5287 NewStuff for rationale and an example.
5289 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
5291 PH/02 Patch to exigrep to allow it to work on syslog lines.
5293 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
5294 fread() to skip over the body file's header line, because in Cygwin the
5295 header line is locked and is inaccessible.
5297 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
5298 co-exist for some time) to make it clear that it is the Exim ID that is
5299 referenced, not the Message-ID: header line.
5301 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
5302 string_format() function, because snprintf() does not exist on all
5305 PH/06 The use of forbid_filter_existstest now also locks out the use of the
5306 ${stat: expansion item.
5308 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
5309 protocol synchronization error", to keep the pedants happy.
5311 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
5312 well as for IRIX systems, when gcc is being used. See the host.c source
5315 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
5317 PH/10 Named domain lists were not working if used in a queue_smtp_domains
5320 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
5321 transport and to the smtp transport in LMTP mode.
5323 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
5325 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
5326 run a filter in a subprocess. This could lead to confusion in subsequent
5327 lookups in the parent process. There should also be a search_tidyup() at
5328 the end of the subprocess.
5330 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
5331 only if the host matched helo_try_verify_hosts, which caused the
5332 verification to occur when the EHLO/HELO command was issued. The ACL just
5333 tested the remembered result. Now, if a previous verification attempt has
5334 not happened, "verify = helo" does it there and then.
5336 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
5338 TK/03 Fix log output including CR from clamd.
5340 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
5341 exist provoked a memory error which could cause a segfault.
5343 PH/15 Installed PCRE 6.2
5345 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
5347 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
5348 of the problem. Specifically, suggested +O2 rather than +O1 for the
5351 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
5353 PH/20 If a delivery was routed to a non-standard port by means of an SRV
5354 record, the port was not correctly logged when the outgoing_port log
5355 selector was set (it logged the transort's default port).
5357 PH/21 Added support for host-specific ports to manualroute, queryprogram,
5358 fallback_hosts, and "hosts" in the smtp transport.
5360 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
5361 host errors such as "Connection refused".
5363 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
5364 authentication with radiusclient 0.4.9:
5366 - Error returned from rc_read_config was caught wrongly
5367 - Username/password not passed on to radius server due to wrong length.
5369 The presumption is that some radiusclient API changes for 4.51/PH/17
5370 were not taken care of correctly. The code is still untested by me (my
5371 Linux distribution still has 0.3.2 of radiusclient), but it was
5372 contributed by a Radius user.
5374 PH/24 When doing a callout, the value of $domain wasn't set correctly when
5375 expanding the "port" option of the smtp transport.
5377 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
5378 while reading a MIME header. Thanks to Tom Hughes for a patch.
5380 PH/24 Include config.h inside local_scan.h so that configuration settings are
5383 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
5384 that in an ACL for RCPT (for example), you can examine exactly what was
5387 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
5388 commands, but it was not correctly comparing the address with the actual
5389 client host address. Thus, it would show the EHLO address in Received:
5390 header lines when this was not necessary.
5392 PH/27 Added the % operator to ${eval:}.
5394 PH/28 Exim tries to create and chdir to its spool directory when it starts;
5395 it should be ignoring failures (because with -C, for example, it has lost
5396 privilege). It wasn't ignoring creation failures other than "already
5399 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
5402 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
5403 because an input line was too long, either on its own, or by virtue of
5404 too many continuations, the temporary file was not being removed, and the
5405 return code was incorrect.
5407 PH/31 Missing "BOOL" in function definition in filtertest.c.
5409 PH/32 Applied Sieve patches from the maintainer.
5411 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
5413 PH/33 Added "verify = not_blind".
5415 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
5416 Local/Makefile (with some defaults set). These are used in built scripts
5417 such as exicyclog, but they have never been used in the exim_install
5418 script (though there are many overriding facilities there). I have
5419 arranged that the exim_install script now takes note of these two
5422 PH/35 Installed configuration files for Dragonfly.
5424 PH/36 When a locally submitted message by a trusted user did not contain a
5425 From: header, and the sender address was obtained from -f or from an SMTP
5426 MAIL command, and the trusted user did not use -F to supply a sender
5427 name, $originator_name was incorrectly used when constructing a From:
5428 header. Furthermore, $originator_name was used for submission mode
5429 messages from external hosts without From: headers in a similar way,
5430 which is clearly wrong.
5432 PH/37 Added control=suppress_local_fixups.
5434 PH/38 When log_selector = +received_sender was set, and the addition of the
5435 sender made the log line's construction buffer exactly full, or one byte
5436 less than full, an overflow happened when the terminating "\n" was
5439 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
5440 when the result of a list match is failure because a DNS lookup failed.
5442 PH/40 RM_COMMAND is now used in the building process.
5444 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
5445 the "build-* directories that it finds.
5447 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
5448 address in a domain literal was a prefix of an interface address.
5450 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
5451 when verifying a sender address, unless rewrite_headers is false.
5453 PH/44 Wrote a long comment about why errors_to addresses are verified as
5454 recipients, not senders.
5456 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
5457 the ratelimit ACL was added.
5459 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
5461 PH/46 Added extra information about PostgreSQL errors to the error string.
5463 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
5464 avoiding the need to install special zones in a real server. This is
5465 backwards compatible; if it can't find the fake resolver, it drops back.
5466 Thus, both old and new test suites can be run.
5468 TF/02 Added util/ratelimit.pl
5470 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
5471 clock is set back in time.
5473 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
5474 Candler <B.Candler@pobox.com>.
5476 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
5477 OK for addresses that are the result of redirections.
5479 PH/48 A number of further additions for the benefit of the new test suite,
5480 including a fake gethostbyname() that interfaces to the fake DNS resolver
5483 TF/06 The fix for widen_domains has also been applied to qualify_single and
5484 search_parents which are the other dnslookup options that can cause
5487 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
5490 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
5492 TF/07 Exim produced the error message "an SRV record indicated no SMTP
5493 service" if it encountered an MX record with an empty target hostname.
5494 The message is now "an MX or SRV record indicated no SMTP service".
5496 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
5497 if the DNS of the sending site is misconfigured. This is quite a
5498 common situation. This change restores the behaviour of treating a
5499 helo verification defer as a failure.
5501 PH/51 If self=fail was set on a router, the bounce message did not include the
5502 actual error message.
5508 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
5510 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
5511 command itself ended in error, the underlying message about the transport
5512 filter timeout was being overwritten with the pipe command error. Now the
5513 underlying error message should be appended to the second error message.
5515 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
5517 PH/02 Reduce the amount of output that "make" produces by default. Full output
5518 can still be requested.
5520 PH/03 The warning log line about a condition test deferring for a "warn" verb
5521 was being output only once per connection, rather than after each
5522 occurrence (because it was using the same function as for successful
5523 "warn" verbs). This seems wrong, so I have changed it.
5525 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
5526 it should not have, which might have caused a crash in the right
5527 circumstances, but probably never did.
5529 PH/04 Installed a modified version of Tony Finch's patch to make submission
5530 mode fix the return path as well as the Sender: header line, and to
5531 add a /name= option so that you can make the user's friendly name appear
5534 TF/03 Added the control = fakedefer ACL modifier.
5536 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
5537 Mark Lowes for thorough testing.
5539 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
5541 TK/03 Merged latest SRS patch from Miles Wilton.
5543 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
5544 with the definition in sysexits.h (which is #included earlier).
5545 Fortunately, Exim does not actually use EX_OK. The code used to try to
5546 preserve the sysexits.h value, by assuming that macro definitions were
5547 scanned for macro replacements. I have been disabused of this notion,
5548 so now the code just undefines EX_OK before #including unistd.h.
5550 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
5551 in the smtp transport. When a block could not be written in a single
5552 write() function, the timeout was being re-applied to each part-write.
5553 This seems wrong - if the receiver was accepting one byte at a time it
5554 would take for ever. The timeout is now adjusted when this happens. It
5555 doesn't have to be particularly precise.
5557 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
5558 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
5560 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
5561 without a domain if the check to <postmaster@domain> fails.
5563 SC/01 Eximstats: added -xls and the ability to specify output files
5564 (patch written by Frank Heydlauf).
5566 SC/02 Eximstats: use FileHandles for outputting results.
5568 SC/03 Eximstats: allow any combination of xls, txt, and html output.
5570 SC/04 Eximstats: fixed display of large numbers with -nvr option
5572 SC/05 Eximstats: fixed merging of reports with empty tables.
5574 SC/06 Eximstats: added the -include_original_destination flag
5576 SC/07 Eximstats: removed tabs and trailing whitespace.
5578 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
5580 TK/06 MBOX spool code: Add real "From " MBOX separator line
5581 so the .eml file is really in mbox format (even though
5582 most programs do not really care). Patch from Alex Miller.
5584 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
5585 The latter is generated from $received_to and is only set if the
5586 message has one envelope recipient. SA can use these headers,
5587 obviously out-of-the-box. Patch from Alex Miller.
5589 PH/08 The ${def test on a variable was returning false if the variable's
5590 value was "0", contrary to what the specification has always said!
5591 The result should be true unless the variable is empty.
5593 PH/09 The syntax error of a character other than { following "${if
5594 def:variable_name" (after optional whitespace) was not being diagnosed.
5595 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
5596 accidental colon was present, for example, could give incorrect results.
5598 PH/10 Tidied the code in a number of places where the st_size field of a stat()
5599 result is used (not including appendfile, where other changes are about
5602 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
5603 This involved changing a lot of size variables from int to off_t. It
5604 should work with maildirs and everything.
5606 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
5607 spamd dying while we are connected to it.
5609 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
5612 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
5613 function for BDB 4.3.
5615 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
5617 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
5618 the sources for PCRE have been re-arranged and more files are now
5621 PH/15 The code I had for printing potentially long long variables in PH/11
5622 above was not the best (it lost precision). The length of off_t variables
5623 is now inspected at build time, and an appropriate printing format (%ld
5624 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
5625 to be "long long int" or "long int". This is needed for the internal
5626 formatting function string_vformat().
5628 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
5629 the configuration file to be ":syslog", then the script "guesses" where
5630 the logs files are, rather than using the compiled in default. In our
5631 case the guess is not the same as the compiled default, so the script
5632 suddenly stopped working when I started to use syslog. The patch checks
5633 to see if log_file_path is "". If so, it attempts to read it from exim
5634 with no configuration file to get the compiled in version, before it
5635 falls back to the previous guessing code."
5637 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
5638 implementing BATV in an Exim configuration. See NewStuff for the gory
5641 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
5642 Makefile that are specific to HP-UX.
5644 PH/18 If the "use_postmaster" option was set for a recipient callout together
5645 with the "random" option, the postmaster address was used as the MAIL
5646 FROM address for the random test, but not for the subsequent recipient
5647 test. It is now used for both.
5649 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
5650 patch removes a few documentation additions to RFC 3028, because the
5651 latest draft now contains them. It adds the new en;ascii-case comparator
5652 and a new error check for 8bit text in MIME parts. Comparator and
5653 require names are now matched exactly. I enabled the subaddress
5654 extension, but it is not well tested yet (read: it works for me)."
5656 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
5657 rework some of the code of TK/09 above to avoid the hardwired use of
5658 "%lld" and "long long". Replaced the call to snprintf() with a call to
5661 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
5662 records point to non-existent hosts", "retry timeout exceeded", and
5663 "retry time not reached for any host after a long failure period".
5665 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
5666 experimental DomainKeys support:
5668 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
5669 (2) On an error such as an illegally used "control", the wrong name for
5670 the control was given.
5672 These problems did NOT occur unless DomainKeys support was compiled.
5674 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
5676 PH/24 Added ${if match_ip condition.
5678 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
5679 databases so that it will be absolutely obvious if a crash occurs in the
5680 DB library. This is a regular occurrence (often caused by mis-matched
5683 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
5684 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
5685 picked up on a user's system that detects such things. There doesn't seem
5686 to be a gcc warning option for this - only an attribute that has to be
5687 put on the function's prototype. It seems that in Fedora Core 4 they have
5688 set this on a number of new functions. No doubt there will be more in due
5691 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
5692 specify a transport. However, if an address that was verified by such a
5693 router was the subject of a callout, Exim crashed because it tried to
5694 read the rcpt_include_affixes from the non-existent transport. Now it
5695 just assumes that the setting of that option is false. This bug was
5696 introduced by 4.51/PH/31.
5698 PH/28 Changed -d+all to exclude +memory, because that information is very
5699 rarely of interest, but it makes the output a lot bigger. People tend to
5700 do -d+all out of habit.
5702 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
5703 code in os-type was giving problems when libc.so lives in lib64, like on
5706 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
5707 aren't the modern standard, and it seems that some systems' include files
5708 don't always have them. Exim was already checking for some of the newer
5709 ones like T_AAAA, and defining it itself. I've added checks for all the
5710 record types that Exim uses.
5712 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
5713 not automatically generating a new one, as it is supposed to. This
5714 prevented TLS from working. If the file did exist, but contained invalid
5715 data, a new version was generated, as expected. It was only the case of a
5716 non-existent file that was broken.
5718 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
5719 with a change in libdomainkeys > 0.64.
5721 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
5722 from DNS. If the selector record carries the flag, it now has
5723 precedence over the domain-wide flag.
5725 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
5727 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
5728 the use of a port name that isn't defined in /etc/services) occurred, the
5729 message was deferred as in a normal delivery, and thus remained on the
5730 spool, instead of being failed because of the mua_wrapper setting. This
5731 is now fixed, and I tidied up some of the mua_wrapper messages at the
5734 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
5735 the messages in a hash of arrays rather than using individual hashes.
5736 This is a bit cleaner and results in dramatic memory savings, albeit
5737 at a slight CPU cost.
5739 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
5740 as requested by Marc Sherman.
5742 SC/10 Eximstats: added histograms for user specified patterns as requested
5745 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
5747 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
5748 fopen() in the content-scanning modules that did not already have it.
5754 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
5755 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
5757 TK/02 Fix ACL "control" statement not being available in MIME ACL.
5759 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
5761 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
5762 to test Sieve filters that use "vacation".
5764 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
5765 that changes the way the GnuTLS parameters are stored in the cache file.
5766 The new format can be generated externally. For backward compatibility,
5767 if the data in the cache doesn't make sense, Exim assumes it has read an
5768 old-format file, and it generates new data and writes a new file. This
5769 means that you can't go back to an older release without removing the
5772 PH/03 A redirect router that has both "unseen" and "one_time" set does not
5773 work if there are any delivery delays because "one_time" forces the
5774 parent to be marked "delivered", so its unseen clone is never tried
5775 again. For this reason, Exim now forbids the simultaneous setting of
5778 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
5779 redirected to themselves ("homonym" addresses). Read the long ChangeLog
5780 entry if you want to know the details. The fix, however, neglected to
5781 consider the case when local delivery batching is involved. The test for
5782 "previously delivered" was not happening when checking to see if an
5783 address could be batched with a previous (undelivered) one; under
5784 certain circumstances this could lead to multiple deliveries to the same
5787 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
5788 in its include files, and this causes problems building Exim.
5790 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
5791 header_syntax) but Exim was just ignoring anything given after a slash.
5792 In particular, this caused confusion with an attempt to use "verify =
5793 reverse_host_lookup/defer_ok". An error is now given when options are
5794 supplied for verify items that do not have them. (Maybe reverse_host_
5795 lookup should have a defer_ok option, but that's a different point.)
5797 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
5798 defined by RFC 821) to 2048, because there were problems with some AUTH
5799 commands, and RFC 1869 says the size should be increased for extended
5800 SMTP commands that take arguments.
5802 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
5805 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
5806 "unknown" error; now it says that the functionality isn't in the binary.
5808 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
5809 an address' error message when a string expansion fails (syntax or
5810 whatever). Otherwise the password may appear in the log. Following change
5811 PH/42 below, there is no longer a chance of it appearing in a bounce
5814 PH/11 Installed exipick version 20050225.0 from John Jetmore.
5816 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
5817 of its addresses was ever tried. (Bugzilla bug #2.)
5819 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
5820 the result incorrectly in the debug output. (It correctly added a newline
5821 to what was transported.)
5823 TF/01 Added $received_time.
5825 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
5826 commented out examples of how to interface to a virus scanner and to
5827 SpamAssassin. Also added commented examples of av_scanner and
5828 spamd_address settings.
5830 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
5831 and controls are allowed in which ACLs. There were a couple of minor
5832 errors. Some of the entries in the conditions table (which is a table of
5833 where they are NOT allowed) were getting very unwieldy; rewrote them as a
5834 negation of where the condition IS allowed.
5836 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
5838 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
5839 header file does not have a version number, so I've had to invent a new
5840 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
5841 API. The code is untested by me (my Linux distribution still has 0.3.2 of
5842 radiusclient), but it was contributed by a Radius user.
5844 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
5845 files or directories, for OpenSSL.
5847 PH/19 When an Exim process that is running as root has to create an Exim log
5848 file, it does so in a subprocess that runs as exim:exim so as to get the
5849 ownership right at creation (otherwise, other Exim processes might see
5850 the file with the wrong ownership). There was no test for failure of this
5851 fork() call, which would lead to the process getting stuck as it waited
5852 for a non-existent subprocess. Forks do occasionally fail when resources
5853 run out. I reviewed all the other calls to fork(); they all seem to check
5856 PH/20 When checking for unexpected SMTP input at connect time (before writing
5857 the banner), Exim was not dealing correctly with a non-positive return
5858 from the read() function. If the client had disconnected by this time,
5859 the result was a log entry for a synchronization error with an empty
5860 string after "input=" when read() returned zero. If read() returned -1
5861 (an event I could not check), uninitialized data bytes were printed.
5862 There were reports of junk text (parts of files, etc) appearing after
5865 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
5867 PH/22 Added support for macro redefinition, and (re)definition in between
5868 driver and ACL definitions.
5870 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
5871 forgetting to use the resulting value; it was using the unexpanded value.
5873 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
5874 hadn't been configured. The fix is from Juergen Kreileder, who
5875 understands it better than I do:
5877 "Here's what I see happening with three configured cyrus_sasl
5878 authenticators configured (plain, login, cram-md5):
5880 On startup auth_cyrus_sasl_init() gets called for each of these.
5881 This means three calls to sasl_listmech() without a specified mech_list.
5882 => SASL tests which mechs of all available mechs actually work
5883 => three warnings about OTP not working
5884 => the returned list contains: plain, login, cram-md5, digest-md5, ...
5886 With the patch, sasl_listmech() also gets called three times. But now
5887 SASL's mech_list option is set to the server_mech specified in the the
5888 authenticator. Or in other words, the answer from sasl_listmech()
5889 gets limited to just the mech you're testing for (which is different
5891 => the return list contains just 'plain' or 'login', 'cram-md5' or
5892 nothing depending on the value of ob->server_mech.
5894 I've just tested the patch: Authentication still works fine,
5895 unavailable mechs specified in the exim configuration are still
5896 caught, and the auth.log warnings about OTP are gone."
5898 PH/25 When debugging is enabled, the contents of the command line are added
5899 to the debugging output, even when log_selector=+arguments is not
5902 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
5903 answer is "GNU", and only if the return is "GNU/something" is the answer
5906 PH/27 $acl_verify_message is now set immediately after the failure of a
5907 verification in an ACL, and so is available in subsequent modifiers. In
5908 particular, the message can be preserved by coding like this:
5910 warn !verify = sender
5911 set acl_m0 = $acl_verify_message
5913 Previously, $acl_verify_message was set only while expanding "message"
5914 and "log_message" when a very denied access.
5916 PH/28 Modified OS/os.c-Linux with
5918 -#ifndef OS_LOAD_AVERAGE
5919 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
5921 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
5922 nomenclature these days.)
5924 PH/29 Installed patch from the Sieve maintainer that adds the options
5925 sieve_useraddress and sieve_subaddress to the redirect router.
5927 PH/30 In these circumstances:
5928 . Two addresses routed to the same list of hosts;
5929 . First host does not offer TLS;
5930 . First host accepts first address;
5931 . First host gives temporary error to second address;
5932 . Second host offers TLS and a TLS session is established;
5933 . Second host accepts second address.
5934 Exim incorrectly logged both deliveries with the TLS parameters (cipher
5935 and peerdn, if requested) that were in fact used only for the second
5938 PH/31 When doing a callout as part of verifying an address, Exim was not paying
5939 attention to any local part prefix or suffix that was matched by the
5940 router that accepted the address. It now behaves in the same way as it
5941 does for delivery: the affixes are removed from the local part unless
5942 rcpt_include_affixes is set on the transport.
5944 PH/32 Add the sender address, as F=<...>, to the log line when logging a
5945 timeout during the DATA phase of an incoming message.
5947 PH/33 Sieve envelope tests were broken for match types other than :is. I have
5948 applied a patch sanctioned by the Sieve maintainer.
5950 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
5951 the uid or gid is negative. A case of a negative gid caused this to be
5952 noticed. The fix allows for either to be negative.
5954 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
5955 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
5958 PH/36 The change PH/12 above was broken. Fixed it.
5960 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
5961 the grounds that routing the same address twice would always produce the
5962 same answer. This might have been true once, but it is certainly no
5963 longer true now. Routing a child address may depend on the previous
5964 routing that produced that child. Some complicated redirection strategies
5965 went wrong when messages had multiple recipients, and made Exim's
5966 behaviour dependent on the order in which the addresses were given.
5968 I have moved the duplicate checking until after the routing is complete.
5969 Exim scans the addresses that are assigned to local and remote
5970 transports, and removes any duplicates. This means that more work will be
5971 done, as duplicates will always all be routed, but duplicates are
5972 presumably rare, so I don't expect this is of any significance.
5974 For deliveries to pipes, files, and autoreplies, the duplicate checking
5975 still happens during the routing process, since they are not going to be
5978 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
5979 It corrects a timeout issue with spamd. This is Ian's comment: "The
5980 background is that sometimes spamd either never reads data from a
5981 connection it has accepted, or it never writes response data. The exiscan
5982 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
5983 blindly assumes that writes won't block so it may never time out."
5985 PH/39 Allow G after quota size as well as K and M.
5987 PH/40 The value set for $authenticated_id in an authenticator may not contain
5988 binary zeroes or newlines because the value is written to log lines and
5989 to spool files. There was no check on this. Now the value is run through
5990 the string_printing() function so that such characters are converted to
5991 printable escape sequences.
5993 PH/41 $message_linecount is a new variable that contains the total number of
5994 lines in the message. Compare $body_linecount, which is the count for the
5997 PH/42 Exim no longer gives details of delivery errors for specific addresses in
5998 bounce and delay warning messages, except in certain special cases, which
6001 (a) An SMTP error message from a remote host;
6002 (b) A message specified in a :fail: redirection;
6003 (c) A message specified in a "fail" command in a system filter;
6004 (d) A message specified in a FAIL return from the queryprogram router;
6005 (e) A message specified by the cannot_route_message router option.
6007 In these cases only, Exim does include the error details in bounce and
6008 warning messages. There are also a few cases where bland messages such
6009 as "unrouteable address" or "local delivery error" are given.
6011 PH/43 $value is now also set for the "else" part of a ${run expansion.
6013 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
6014 being worked on, but at least Exim now implements the latest version to
6017 PH/45 In a pipe transport, although a timeout while waiting for the pipe
6018 process to complete was treated as a delivery failure, a timeout while
6019 writing the message to the pipe was logged, but erroneously treated as a
6020 successful delivery. Such timeouts include transport filter timeouts. For
6021 consistency with the overall process timeout, these timeouts are now
6022 treated as errors, giving rise to delivery failures by default. However,
6023 there is now a new Boolean option for the pipe transport called
6024 timeout_defer, which, if set TRUE, converts the failures into defers for
6025 both kinds of timeout. A transport filter timeout is now identified in
6028 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
6029 systems where "make" and "gmake" are different, calling "gmake" at top
6030 level broke things. I've arranged for the value of $(MAKE) to be passed
6031 from the Makefile to this script so that it can call the same version of
6035 A note about Exim versions 4.44 and 4.50
6036 ----------------------------------------
6038 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
6039 changes of various kinds. As a consequence, a big documentation update was
6040 needed. This delayed the release for rather longer than seemed good, especially
6041 in the light of a couple of (minor) security issues. Therefore, the changes
6042 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
6043 release. So 4.44 and 4.50 are in effect two different branches that both start
6046 I have left the 4.50 change log unchanged; it contains all the changes since
6047 4.43. The change log for 4.44 is below; many of its items are identical to
6048 those for 4.50. This seems to be the most sensible way to preserve the
6049 historical information.
6055 1. Minor wording change to the doc/README.SIEVE file.
6057 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
6058 computation of the current number of files was incorrect.
6060 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
6061 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
6064 4. Give more explanation in the error message when the command for a transport
6065 filter fails to execute.
6067 5. There are several places where Exim runs a non-Exim command in a
6068 subprocess. The SIGUSR1 signal should be disabled for these processes. This
6069 was being done only for the command run by the queryprogram router. It is
6070 now done for all such subprocesses. The other cases are: ${run, transport
6071 filters, and the commands run by the lmtp and pipe transports.
6073 6. Added CONFIGURE_GROUP build-time option.
6075 7. Some older OS have a limit of 256 on the maximum number of file
6076 descriptors. Exim was using setrlimit() to set 1000 as a large value
6077 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
6078 systems. I've change it so that if it can't get 1000, it tries for 256.
6080 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
6081 was an oversight, and furthermore, ever since the addition of extra
6082 controls (e.g. 4.43/32), the checks on when to allow different forms of
6083 "control" were broken. There should now be diagnostics for all cases when a
6084 control that does not make sense is encountered.
6086 9. Added the /retain_sender option to "control=submission".
6088 10. $recipients is now available in the predata ACL (oversight).
6090 11. Tidy the search cache before the fork to do a delivery from a message
6091 received from the command line. Otherwise the child will trigger a lookup
6092 failure and thereby defer the delivery if it tries to use (for example) a
6093 cached ldap connection that the parent has called unbind on.
6095 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
6096 of $address_data from the recipient verification was clobbered by the
6097 sender verification.
6099 13. The value of address_data from a sender verification is now available in
6100 $sender_address_data in subsequent conditions in the ACL statement.
6102 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
6104 15. Added a new option "connect=<time>" to callout options, to set a different
6107 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
6108 was its contents. (It was OK if the option was not defined at all.)
6110 17. A "Completed" log line is now written for messages that are removed from
6111 the spool by the -Mrm option.
6113 18. New variables $sender_verify_failure and $recipient_verify_failure contain
6114 information about exactly what failed.
6116 19. Added -dd to debug only the daemon process.
6118 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
6119 handles timeouts, both on the server side and network timeouts. Renamed the
6120 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
6122 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
6123 It is now set to "smtps".
6125 22. $host_address is now set to the target address during the checking of
6126 ignore_target_hosts.
6128 23. When checking ignore_target_hosts for an ipliteral router, no host name was
6129 being passed; this would have caused $sender_host_name to have been used if
6130 matching the list had actually called for a host name (not very likely,
6131 since this list is usually IP addresses). A host name is now passed as
6134 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
6135 code that specifies a non-restarting handler (typically sigaction() in
6136 modern systems) in an attempt to fix a rare and obscure crash bug.
6138 25. Narrowed the window for a race in the daemon that could cause it to ignore
6139 SIGCHLD signals. This is not a major problem, because they are used only to
6140 wake it up if nothing else does.
6142 26. A malformed maildirsize file could cause Exim to calculate negative values
6143 for the mailbox size or file count. Odd effects could occur as a result.
6144 The maildirsize information is now recalculated if the size or filecount
6147 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
6148 support for a long time. Removed HAVE_SYS_VFS_H.
6150 28. Installed the latest version of exipick from John Jetmore.
6152 29. In an address list, if the pattern was not a regular expression, an empty
6153 subject address (from a bounce message) matched only if the pattern was an
6154 empty string. Non-empty patterns were not even tested. This was the wrong
6155 because it is perfectly reasonable to use an empty address as part of a
6156 database query. An empty address is now tested by patterns that are
6157 lookups. However, all the other forms of pattern expect the subject to
6158 contain a local part and a domain, and therefore, for them, an empty
6159 address still always fails if the pattern is not itself empty.
6161 30. Exim went into a mad DNS loop when attempting to do a callout where the
6162 host was specified on an smtp transport, and looking it up yielded more
6163 than one IP address.
6165 31. Re-factored the code for checking spool and log partition space into a
6166 function that finds that data and another that does the check. The former
6167 is then used to implement four new variables: $spool_space, $log_space,
6168 $spool_inodes, and $log_inodes.
6170 32. The RFC2047 encoding function was originally intended for short strings
6171 such as real names; it was not keeping to the 75-character limit for
6172 encoded words that the RFC imposes. It now respects the limit, and
6173 generates multiple encoded words if necessary. To be on the safe side, I
6174 have increased the buffer size for the ${rfc2047: expansion operator from
6177 33. It is now permitted to omit both strings after an "if" condition; if the
6178 condition is true, the result is "true". As before, when the second string
6179 is omitted, a false condition yields an empty string. This makes it less
6180 cumbersome to write custom ACL and router conditions.
6182 34. Failure to deliver a bounce message always caused it to be frozen, even if
6183 there was an errors_to setting on the router. The errors_to setting is now
6186 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
6187 canonical form (fully expanded) before being placed in
6188 $sender_host_address.
6190 36. The table in the code that translates DNS record types into text (T_A to
6191 "A" for instance) was missing entries for NS and CNAME. It is just possible
6192 that this could have caused confusion if both these types were looked up
6193 for the same domain, because the text type is used as part of Exim's
6194 per-process caching. But the chance of anyone hitting this buglet seems
6197 37. The dnsdb lookup has been extended in a number of ways.
6199 (1) There is a new type, "zns", which walks up the domain tree until it
6200 finds some nameserver records. It should be used with care.
6202 (2) There is a new type, "mxh", which is like "mx" except that it returns
6203 just the host names, not the priorities.
6205 (3) It is now possible to give a list of domains (or IP addresses) to be
6206 looked up. The behaviour when one of the lookups defers can be
6207 controlled by a keyword.
6209 (4) It is now possible to specify the separator character for use when
6210 multiple records are returned.
6212 38. The dnslists ACL condition has been extended: it is now possible to supply
6213 a list of IP addresses and/or domains to be looked up in a particular DNS
6216 39. Added log_selector=+queue_time_overall.
6218 40. When running the queue in the test harness, wait just a tad after forking a
6219 delivery process, to get repeatability of debugging output.
6221 41. Include certificate and key file names in error message when GnuTLS fails
6222 to set them up, because the GnuTLS error message doesn't include the name
6223 of the failing file when there is a problem reading it.
6225 42. Allow both -bf and -bF in the same test run.
6227 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
6229 44. The "Exiscan patch" is now merged into the mainline Exim source.
6231 45. Sometimes the final signoff response after QUIT could fail to get
6232 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
6233 < 0 before doing a fflush(). This bug looks as though it goes back to the
6234 introduction of TLS in release 3.20, but "sometimes" must have been rare
6235 because the tests only now provoked it.
6237 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
6238 (this can affect the format of dates).
6240 47. exim_tidydb, when checking for the continued existence of a message for
6241 which it has found a message-specific retry record, was not finding
6242 messages that were in split spool directories. Consequently, it was
6243 deleting retry records that should have stayed in existence.
6245 48. Steve fixed some bugs in eximstats.
6247 49. The SPA authentication driver was not abandoning authentication and moving
6248 on to the next authenticator when an expansion was forced to fail,
6249 contradicting the general specification for all authenticators. Instead it
6250 was generating a temporary error. It now behaves as specified.
6252 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
6253 (the order specifies the preference for clients). The order is now AES256,
6254 AES128, 3DES, ARCFOUR128.
6256 51. Small patch to Sieve code - explicitly set From: when generating an
6259 52. Exim crashed if a remote delivery caused a very long error message to be
6260 recorded - for instance if somebody sent an entire SpamAssassin report back
6261 as a large number of 550 error lines. This bug was coincidentally fixed by
6262 increasing the size of one of Exim's internal buffers (big_buffer) that
6263 happened as part of the Exiscan merge. However, to be on the safe side, I
6264 have made the code more robust (and fixed the comments that describe what
6267 53. Now that there can be additional text after "Completed" in log lines (if
6268 the queue_time_overall log selector is set), a one-byte patch to exigrep
6269 was needed to allow it to recognize "Completed" as not the last thing in
6272 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
6273 patch that reportedly fixes this has been added. I am not expert enough to
6274 create a test for it. This is what the patch creator wrote:
6276 "I found a little strange behaviour of ldap code when working with
6277 Windows 2003 AD Domain, where users was placed in more than one
6278 Organization Units. When I tried to give exim partial DN, the exit code
6279 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
6280 But simultaneously result of request was absolutely normal ldap result,
6281 so I produce this patch..."
6283 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
6284 REFERENCE, so I have modified the code to exclude the patch when that macro
6287 55. Some experimental protocols are using DNS PTR records for new purposes. The
6288 keys for these records are domain names, not reversed IP addresses. The
6289 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
6290 leaves it alone. Component reversal etc. now happens only for IP addresses.
6293 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
6295 57. Double the size of the debug message buffer (to 2048) so that more of very
6296 long debug lines gets shown.
6298 58. The exicyclog utility now does better if the number of log files to keep
6299 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
6301 59. Two changes related to the smtp_active_hostname option:
6303 (1) $smtp_active_hostname is now available as a variable.
6304 (2) The default for smtp_banner uses $smtp_active_hostname instead
6305 of $primary_hostname.
6307 60. The host_aton() function is supposed to be passed a string that is known
6308 to be a valid IP address. However, in the case of IPv6 addresses, it was
6309 not checking this. This is a hostage to fortune. Exim now panics and dies
6310 if the condition is not met. A case was found where this could be provoked
6311 from a dnsdb PTR lookup with an IPv6 address that had more than 8
6312 components; fortuitously, this particular loophole had already been fixed
6313 by change 4.50/55 above.
6315 If there are any other similar loopholes, the new check in host_aton()
6316 itself should stop them being exploited. The report I received stated that
6317 data on the command line could provoke the exploit when Exim was running as
6318 exim, but did not say which command line option was involved. All I could
6319 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
6320 running as the user.
6323 61. There was a buffer overflow vulnerability in the SPA authentication code
6324 (which came originally from the Samba project). I have added a test to the
6325 spa_base64_to_bits() function which I hope fixes it.
6328 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
6329 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
6331 63. The daemon start-up calls getloadavg() while still root for those OS that
6332 need the first call to be done as root, but it missed one case: when
6333 deliver_queue_load_max is set with deliver_drop_privilege. This is
6334 necessary for the benefit of the queue runner, because there is no re-exec
6335 when deliver_drop_privilege is set.
6337 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
6338 This has been fixed.
6340 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
6341 was in use, was not putting the data itself into the right store pool;
6342 consequently, it could be overwritten for a subsequent message in the same
6343 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
6346 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
6348 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
6349 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
6350 the function were treating the return as a boolean value, which happened to
6351 work because 0=false and not-0=true, but is not correct code.
6353 68. The host_aton() function was not handling scoped IPv6 addresses (those
6354 with, for example, "%eth0" on the end) correctly.
6356 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
6357 negated items (that is, ~something) in unsigned ints. Some compilers
6358 apparently mutter when there is no cast.
6360 70. If an address verification called from an ACL failed, and did not produce a
6361 user-specific message (i.e. there was only a "system" message), nothing was
6362 put in $acl_verify_message. In this situation, it now puts the system
6365 71. Change 4.23/11 added synchronization checking at the start of an SMTP
6366 session; change 4.31/43 added the unwanted input to the log line - except
6367 that it did not do this in the start of session case. It now does.
6369 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
6370 This is wrong and can cause the other end to generate a synchronization
6371 error if it is another Exim or anything else that does the synchronization
6372 check. A QUIT command is no longer sent after a timeout.
6374 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
6375 during host lookups.
6377 74. The defer_ok option of callout verification was not working if it was used
6378 when verifying addresses in header lines, that is, for this case:
6380 verify = header_sender/callout=defer_ok
6382 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
6383 those file descriptors could be used for SMTP connections. If anything
6384 wrote to stderr (the example that came up was "warn" in embedded Perl), it
6385 could be sent to the SMTP client, causing chaos. The daemon now opens
6386 stdin, stdout, and stderr to /dev/null when it puts itself into the
6389 76. Arrange for output from Perl's "warn" command to be written to Exim's main
6390 log by default. The user can override this with suitable Perl magic.
6392 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
6393 the log message when discard triggers, was not working for the DATA ACL or
6394 for the non-SMTP ACL.
6396 78. Error message wording change in sieve.c.
6398 79. If smtp_accept_max_per_host was set, the number of connections could be
6399 restricted to fewer than expected, because the daemon was trying to set up
6400 a new connection before checking whether the processes handling previous
6401 connections had finished. The check for completed processes is now done
6402 earlier. On busy systems, this bug wouldn't be noticed because something
6403 else would have woken the daemon, and it would have reaped the completed
6406 80. If a message was submitted locally by a user whose login name contained one
6407 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
6408 It caused a spool format error. I have fixed the spool reading code. A
6409 related problem was that the "from" clause in the Received: line became
6410 illegal because of the space(s). It is now covered by ${quote_local_part.
6412 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
6415 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
6417 83. Patch from Sieve maintainer to fix unterminated string problem in
6418 "vacation" handling.
6420 84. Some minor changes to the Linux configuration files to help with other
6421 OS variants using glibc.
6423 85. One more patch for Sieve to update vacation handling to latest spec.
6426 ----------------------------------------------------
6427 See the note above about the 4.44 and 4.50 releases.
6428 ----------------------------------------------------
6434 1. Change 4.43/35 introduced a bug that caused file counts to be
6435 incorrectly computed when quota_filecount was set in an appendfile
6438 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
6439 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
6442 3. Give more explanation in the error message when the command for a transport
6443 filter fails to execute.
6445 4. There are several places where Exim runs a non-Exim command in a
6446 subprocess. The SIGUSR1 signal should be disabled for these processes. This
6447 was being done only for the command run by the queryprogram router. It is
6448 now done for all such subprocesses. The other cases are: ${run, transport
6449 filters, and the commands run by the lmtp and pipe transports.
6451 5. Some older OS have a limit of 256 on the maximum number of file
6452 descriptors. Exim was using setrlimit() to set 1000 as a large value
6453 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
6454 systems. I've change it so that if it can't get 1000, it tries for 256.
6456 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
6457 was an oversight, and furthermore, ever since the addition of extra
6458 controls (e.g. 4.43/32), the checks on when to allow different forms of
6459 "control" were broken. There should now be diagnostics for all cases when a
6460 control that does not make sense is encountered.
6462 7. $recipients is now available in the predata ACL (oversight).
6464 8. Tidy the search cache before the fork to do a delivery from a message
6465 received from the command line. Otherwise the child will trigger a lookup
6466 failure and thereby defer the delivery if it tries to use (for example) a
6467 cached ldap connection that the parent has called unbind on.
6469 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
6470 of $address_data from the recipient verification was clobbered by the
6471 sender verification.
6473 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
6474 was its contents. (It was OK if the option was not defined at all.)
6476 11. A "Completed" log line is now written for messages that are removed from
6477 the spool by the -Mrm option.
6479 12. $host_address is now set to the target address during the checking of
6480 ignore_target_hosts.
6482 13. When checking ignore_target_hosts for an ipliteral router, no host name was
6483 being passed; this would have caused $sender_host_name to have been used if
6484 matching the list had actually called for a host name (not very likely,
6485 since this list is usually IP addresses). A host name is now passed as
6488 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
6489 code that specifies a non-restarting handler (typically sigaction() in
6490 modern systems) in an attempt to fix a rare and obscure crash bug.
6492 15. Narrowed the window for a race in the daemon that could cause it to ignore
6493 SIGCHLD signals. This is not a major problem, because they are used only to
6494 wake it up if nothing else does.
6496 16. A malformed maildirsize file could cause Exim to calculate negative values
6497 for the mailbox size or file count. Odd effects could occur as a result.
6498 The maildirsize information is now recalculated if the size or filecount
6501 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
6502 support for a long time. Removed HAVE_SYS_VFS_H.
6504 18. Updated exipick to current release from John Jetmore.
6506 19. Allow an empty sender to be matched against a lookup in an address list.
6507 Previously the only cases considered were a regular expression, or an
6510 20. Exim went into a mad DNS lookup loop when doing a callout where the
6511 host was specified on the transport, if the DNS lookup yielded more than
6514 21. The RFC2047 encoding function was originally intended for short strings
6515 such as real names; it was not keeping to the 75-character limit for
6516 encoded words that the RFC imposes. It now respects the limit, and
6517 generates multiple encoded words if necessary. To be on the safe side, I
6518 have increased the buffer size for the ${rfc2047: expansion operator from
6521 22. Failure to deliver a bounce message always caused it to be frozen, even if
6522 there was an errors_to setting on the router. The errors_to setting is now
6525 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
6526 canonical form (fully expanded) before being placed in
6527 $sender_host_address.
6529 24. Updated eximstats to version 1.33
6531 25. Include certificate and key file names in error message when GnuTLS fails
6532 to set them up, because the GnuTLS error message doesn't include the name
6533 of the failing file when there is a problem reading it.
6535 26. Expand error message when OpenSSL has problems setting up cert/key files.
6538 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
6539 (this can affect the format of dates).
6541 28. exim_tidydb, when checking for the continued existence of a message for
6542 which it has found a message-specific retry record, was not finding
6543 messages that were in split spool directories. Consequently, it was
6544 deleting retry records that should have stayed in existence.
6546 29. eximstats updated to version 1.35
6547 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
6548 1.35 - bugfix such that pie charts by volume are generated correctly
6550 30. The SPA authentication driver was not abandoning authentication and moving
6551 on to the next authenticator when an expansion was forced to fail,
6552 contradicting the general specification for all authenticators. Instead it
6553 was generating a temporary error. It now behaves as specified.
6555 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
6556 (the order specifies the preference for clients). The order is now AES256,
6557 AES128, 3DES, ARCFOUR128.
6559 31. Small patch to Sieve code - explicitly set From: when generating an
6562 32. Exim crashed if a remote delivery caused a very long error message to be
6563 recorded - for instance if somebody sent an entire SpamAssassin report back
6564 as a large number of 550 error lines. This bug was coincidentally fixed by
6565 increasing the size of one of Exim's internal buffers (big_buffer) that
6566 happened as part of the Exiscan merge. However, to be on the safe side, I
6567 have made the code more robust (and fixed the comments that describe what
6570 33. Some experimental protocols are using DNS PTR records for new purposes. The
6571 keys for these records are domain names, not reversed IP addresses. The
6572 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
6573 leaves it alone. Component reversal etc. now happens only for IP addresses.
6576 34. The host_aton() function is supposed to be passed a string that is known
6577 to be a valid IP address. However, in the case of IPv6 addresses, it was
6578 not checking this. This is a hostage to fortune. Exim now panics and dies
6579 if the condition is not met. A case was found where this could be provoked
6580 from a dnsdb PTR lookup with an IPv6 address that had more than 8
6581 components; fortuitously, this particular loophole had already been fixed
6582 by change 4.50/55 or 4.44/33 above.
6584 If there are any other similar loopholes, the new check in host_aton()
6585 itself should stop them being exploited. The report I received stated that
6586 data on the command line could provoke the exploit when Exim was running as
6587 exim, but did not say which command line option was involved. All I could
6588 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
6589 running as the user.
6592 35. There was a buffer overflow vulnerability in the SPA authentication code
6593 (which came originally from the Samba project). I have added a test to the
6594 spa_base64_to_bits() function which I hope fixes it.
6597 36. The daemon start-up calls getloadavg() while still root for those OS that
6598 need the first call to be done as root, but it missed one case: when
6599 deliver_queue_load_max is set with deliver_drop_privilege. This is
6600 necessary for the benefit of the queue runner, because there is no re-exec
6601 when deliver_drop_privilege is set.
6603 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
6604 was in use, was not putting the data itself into the right store pool;
6605 consequently, it could be overwritten for a subsequent message in the same
6606 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
6609 38. Sometimes the final signoff response after QUIT could fail to get
6610 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
6611 < 0 before doing a fflush(). This bug looks as though it goes back to the
6612 introduction of TLS in release 3.20, but "sometimes" must have been rare
6613 because the tests only now provoked it.
6619 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
6620 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
6621 Now it returns an integer. A number of calls were still expecting a T/F
6622 return. Fortuitously, in all cases, the tests worked in OK situations,
6623 which is the norm. However, things would have gone wrong on any write
6624 failures on the smtp file descriptor. This function is used when sending
6625 messages over SMTP and also when doing verify callouts.
6627 2. When Exim is called to do synchronous delivery of a locally submitted
6628 message (the -odf or -odi options), it no longer closes stderr before doing
6631 3. Implemented the mua_wrapper option.
6633 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
6635 5. Implemented the functions header_remove(), header_testname(),
6636 header_add_at_position(), and receive_remove_recipient(), and exported them
6639 6. If an ACL "warn" statement specified the addition of headers, Exim already
6640 inserted X-ACL-Warn: at the start if there was no header name. However, it
6641 was not making this test for the second and subsequent header lines if
6642 there were newlines in the string. This meant that an invalid header could
6643 be inserted if Exim was badly configured.
6645 7. Allow an ACL "warn" statement to add header lines at the start or after all
6646 the Received: headers, as well as at the end.
6648 8. Added the rcpt_4xx retry error code.
6650 9. Added postmaster_mailfrom=xxx to callout verification option.
6652 10. Added mailfrom=xxxx to the callout verification option, for verify=
6655 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
6656 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
6658 12. Inserted some casts to stop certain compilers warning when using pointer
6659 differences as field lengths or precisions in printf-type calls (mostly
6660 affecting debugging statements).
6662 13. Added optional readline() support for -be (dynamically loaded).
6664 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
6665 same clock tick as a message's arrival, so that its received time was the
6666 same as the "first fail" time on the retry record, and that message
6667 remained on the queue past the ultimate address timeout, every queue runner
6668 would try a delivery (because it was past the ultimate address timeout) but
6669 after another failure, the ultimate address timeout, which should have then
6670 bounced the address, did not kick in. This was a "< instead of <=" error;
6671 in most cases the first failure would have been in the next clock tick
6672 after the received time, and all would be well.
6674 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
6675 being recognized when the domain list was tested by the match_domain
6676 condition in an expansion string.
6678 16. Added the ${str2b64: operator.
6680 17. Exim was always calling setrlimit() to set a large limit for the number of
6681 processes, without checking whether the existing limit was already
6682 adequate. (It did check for the limit on file descriptors.) Furthermore,
6683 errors from getrlimit() and setrlimit() were being ignored. Now they are
6684 logged to the main and panic logs, but Exim does carry on, to try to do its
6685 job under whatever limits there are.
6687 18. Imported PCRE 5.0.
6689 19. Trivial typo in log message " temporarily refused connection" (the leading
6692 20. If the log selector return_path_on_delivery was set and an address was
6693 redirected to /dev/null, the delivery process crashed because it assumed
6694 that a return path would always be set for a "successful" delivery. In this
6695 case, the whole delivery is bypassed as an optimization, and therefore no
6698 21. Internal re-arrangement: the function for sending a challenge and reading
6699 a response while authentication was assuming a zero-terminated challenge
6700 string. It's now changed to take a pointer and a length, to allow for
6701 binary data in such strings.
6703 22. Added the cyrus_sasl authenticator (code supplied by MBM).
6705 23. Exim was not respecting finduser_retries when seeking the login of the
6706 uid under which it was called; it was always trying 10 times. (The default
6707 setting of finduser_retries is zero.) Also, it was sleeping after the final
6708 failure, which is pointless.
6710 24. Implemented tls_on_connect_ports.
6712 25. Implemented acl_smtp_predata.
6714 26. If the domain in control=submission is set empty, Exim assumes that the
6715 authenticated id is a complete email address when it generates From: or
6716 Sender: header lines.
6718 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
6719 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
6720 chown and chgrp in /bin and hostname in /usr/bin.
6722 28. Exim was keeping the "process log" file open after each use, just as it
6723 does for the main log. This opens the possibility of it remaining open for
6724 long periods when the USR1 signal hits a daemon. Occasional processlog
6725 errors were reported, that could have been caused by this. Anyway, it seems
6726 much more sensible not to leave this file open at all, so that is what now
6729 29. The long-running daemon process does not normally write to the log once it
6730 has entered its main loop, and it closes the log before doing so. This is
6731 so that log files can straightforwardly be renamed and moved. However,
6732 there are a couple of unusual error situations where the daemon does write
6733 log entries, and I had neglected to close the log afterwards.
6735 30. The text of an SMTP error response that was received during a remote
6736 delivery was being truncated at 512 bytes. This is too short for some of
6737 the long messages that one sometimes sees. I've increased the limit to
6740 31. It is now possible to make retry rules that apply only when a message has a
6741 specific sender, in particular, an empty sender.
6743 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
6744 it possible to be selective about when SMTP synchronization is enforced.
6746 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
6748 32. Implemented hosts_connection_nolog.
6750 33. Added an ACL for QUIT.
6752 34. Setting "delay_warning=" to disable warnings was not working; it gave a
6755 35. Added mailbox_size and mailbox_filecount to appendfile.
6757 36. Added control = no_multiline_responses to ACLs.
6759 37. There was a bug in the logic of the code that waits for the clock to tick
6760 in the case where the clock went backwards by a substantial amount such
6761 that the microsecond fraction of "now" was more than the microsecond
6762 fraction of "then" (but the whole seconds number was less).
6764 38. Added support for the libradius Radius client library this is found on
6765 FreeBSD (previously only the radiusclient library was supported).
6771 1. When certain lookups returned multiple values in the form name=value, the
6772 quoting of the values was not always being done properly. Specifically:
6773 (a) If the value started with a double quote, but contained no whitespace,
6775 (b) If the value contained whitespace other than a space character (i.e.
6776 tabs or newlines or carriage returns) it was not quoted.
6777 This fix has been applied to the mysql and pgsql lookups by writing a
6778 separate quoting function and calling it from the lookup code. The fix
6779 should probably also be applied to nisplus, ibase and oracle lookups, but
6780 since I cannot test any of those, I have not disturbed their existing code.
6782 2. A hit in the callout cache for a specific address caused a log line with no
6783 reason for rejecting RCPT. Now it says "Previous (cached) callout
6784 verification failure".
6786 3. There was an off-by-one bug in the queryprogram router. An over-long
6787 return line was truncated at 256 instead of 255 characters, thereby
6788 overflowing its buffer with the terminating zero. As well as fixing this, I
6789 have increased the buffer size to 1024 (and made a note to document this).
6791 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
6792 when Exim is waiting for an SMTP response from a remote server, Exim
6793 restarts its select() call on the socket, thereby resetting its timeout.
6794 This is not a problem when such interrupts are rare. Somebody set up a cron
6795 job to run exiwhat every 2 minutes, which is less than the normal select()
6796 timeout (5 or 10 minutes). This meant that the select() timeout never
6797 kicked in because it was always reset. I have fixed this by comparing the
6798 time when an interrupt arrives with the time at the start of the first call
6799 to select(). If more time than the timeout has elapsed, the interrupt is
6800 treated as a timeout.
6802 5. Some internal re-factoring in preparation for the addition of Sieve
6803 extensions (by MH). In particular, the "personal" test is moved to a
6804 separate function, and given an option for scanning Cc: and Bcc: (which is
6805 not set for Exim filters).
6807 6. When Exim created an email address using the login of the caller as the
6808 local part (e.g. when creating a From: or Sender: header line), it was not
6809 quoting the local part when it contained special characters such as @.
6811 7. Installed new OpenBSD configuration files.
6813 8. Reworded some messages for syntax errors in "and" and "or" conditions to
6814 try to make them clearer.
6816 9. Callout options, other than the timeout value, were being ignored when
6817 verifying sender addresses in header lines. For example, when using
6819 verify = header_sender/callout=no_cache
6821 the cache was (incorrectly) being used.
6823 10. Added a missing instance of ${EXE} to the exim_install script; this affects
6824 only the Cygwin environment.
6826 11. When return_path_on_delivery was set as a log selector, if different remote
6827 addresses in the same message used different return paths and parallel
6828 remote delivery occurred, the wrong values would sometimes be logged.
6829 (Whenever a remote delivery process finished, the return path value from
6830 the most recently started remote delivery process was logged.)
6832 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
6833 lines when AUTH and/or TLS are in use. This is the "received protocol"
6834 field. Exim used to use "asmtp" for authenticated SMTP, without any
6835 indication (in the protocol name) for TLS use. Now it follows the RFC and
6836 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
6837 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
6838 names appear in log lines as well as in Received: header lines.
6840 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
6841 extensions, and comparison tests, and to fix some bugs.
6843 14. Changes to the "personal" filter test:
6845 (1) The test was buggy in that it was just doing the equivalent of
6846 "contains" tests on header lines. For example, if a user's address was
6847 anne@some.where, the "personal" test would incorrectly be true for
6849 To: susanne@some.where
6851 This test is now done by extracting each address from the header in turn,
6852 and checking the entire address. Other tests that are part of "personal"
6853 are now done using regular expressions (for example, to check local parts
6854 of addresses in From: header lines).
6856 (2) The list of non-personal local parts in From: addresses has been
6857 extended to include "listserv", "majordomo", "*-request", and "owner-*",
6858 taken from the Sieve specification recommendations.
6860 (3) If the message contains any header line starting with "List-" it is
6861 treated as non-personal.
6863 (4) The test for "circular" in the Subject: header line has been removed
6864 because it now seems ill-conceived.
6866 15. Minor typos in src/EDITME comments corrected.
6868 16. Installed latest exipick from John Jetmore.
6870 17. If headers_add on a router specified a text string that was too long for
6871 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
6872 of string_sprintf() is now avoided.
6874 18. $message_body_size was not set (it was always zero) when running the DATA
6875 ACL and the local_scan() function.
6877 19. For the "mail" command in an Exim filter, no default was being set for
6878 the once_repeat time, causing a random time value to be used if "once" was
6879 specified. (If the value happened to be <= 0, no repeat happened.) The
6880 default is now 0s, meaning "never repeat". The "vacation" command was OK
6881 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
6882 (I found it when inspecting the code).
6884 20. There is now an overall timeout for performing a callout verification. It
6885 defaults to 4 times the callout timeout, which applies to individual SMTP
6886 commands during the callout. The overall timeout applies when there is more
6887 than one host that can be tried. The timeout is checked before trying the
6888 next host. This prevents very long delays if there are a large number of
6889 hosts and all are timing out (e.g. when the network connections are timing
6890 out). The value of the overall timeout can be changed by specifying an
6891 additional sub-option for "callout", called "maxwait". For example:
6893 verify = sender/callout=5s,maxwait=20s
6895 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
6896 to the end before writing, but this should make it even safer).
6898 22. Exim was forgetting that it had advertised PIPELINING for the second and
6899 subsequent messages on an SMTP connection. It was also not resetting its
6900 memory on STARTTLS and an internal HELO.
6902 23. When Exim logs an SMTP synchronization error within a session, it now
6903 records whether PIPELINING has been advertised or not.
6905 24. Added 3 instances of "(long int)" casts to time_t variables that were being
6906 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
6907 rather than long int.
6909 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
6911 26. Added the never_mail option to autoreply.
6917 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
6918 crash if the getsockname() call failed; this can happen if a connection is
6919 closed very soon after it is established. The problem was simply in the
6920 order in which certain operations were done, causing Exim to try to write
6921 to the SMTP stream before it had set up the file descriptor. The bug has
6922 been fixed by making things happen in the correct order.
6928 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
6929 before the connection was closed, thus losing the rejection response.
6931 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
6932 some early Solaris releases, but causes trouble in current releases where
6933 socklen_t is defined.
6935 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
6938 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
6941 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
6942 that releasing the top of it at the end releases what was used for sub-
6943 expansions (unless the block got too big). However, discard this block if
6944 the first thing is a variable or header, so that we can use its block when
6945 it is dynamic (useful for very large $message_headers, for example).
6947 6. Lookups now cache *every* query, not just the most recent. A new, separate
6948 store pool is used for this. It can be recovered when all lookup caches are
6949 flushed. Lookups now release memory at the end of their result strings.
6950 This has involved some general refactoring of the lookup sources.
6952 7. Some code has been added to the store_xxx() functions to reduce the amount
6953 of flapping under certain conditions.
6955 8. log_incoming_interface used to affect only the <= reception log lines. Now
6956 it causes the local interface and port to be added to several more SMTP log
6957 lines, for example "SMTP connection from", and rejection lines.
6959 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
6961 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
6963 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
6964 could be overwritten at the end of the current message (or the start of a
6965 new message if it was set in a HELO ACL). The value is now preserved for
6966 the duration of the SMTP connection.
6968 12. If a transport had a headers_rewrite setting, and a matching header line
6969 contained an unqualified address, that address was qualified, even if it
6970 did not match any rewriting rules. The underlying bug was that the values
6971 of the flags that permit the existence of unqualified sender and recipient
6972 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
6973 non-local messages, and by -bnq for local messages) were not being
6974 preserved with the message after it was received.
6976 13. When Exim was logging an SMTP synchronization error, it could sometimes log
6977 "next input=" as part of the text comprising the host identity instead of
6978 the correct text. The code was using the same buffer for two different
6979 strings. However, depending on which order the printing function evaluated
6980 its arguments, the bug did not always show up. Under Linux, for example, my
6981 test suite worked just fine.
6983 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
6984 doesn't work with some older versions of Perl. It has been changed to "my",
6985 which in any case is probably the better facility to use.
6987 15. A really picky compiler found some instances of statements for creating
6988 error messages that either had too many or two few arguments for the format
6991 16. The size of the buffer for calls to the DNS resolver has been increased
6992 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
6993 for addresses that have a lot of PTR records. This alleviates a problem; it
6994 does not fully solve it.
6996 17. A dnsdb lookup for PTR records that receives more data than will fit in the
6997 buffer now truncates the list and logs the incident, which is the same
6998 action as happens when Exim is looking up a host name and its aliases.
6999 Previously in this situation something unpredictable would happen;
7000 sometimes it was "internal error: store_reset failed".
7002 18. If a server dropped the connection unexpectedly when an Exim client was
7003 using GnuTLS and trying to read a response, the client delivery process
7004 crashed while trying to generate an error log message.
7006 19. If a "warn" verb in an ACL added multiple headers to a message in a single
7007 string, for example:
7009 warn message = H1: something\nH2: something
7011 the text was added as a single header line from Exim's point of view
7012 though it ended up OK in the delivered message. However, searching for the
7013 second and subsequent header lines using $h_h2: did not work. This has been
7014 fixed. Similarly, if a system filter added multiple headers in this way,
7015 the routers could not see them.
7017 20. Expanded the error message when iplsearch is called with an invalid key to
7018 suggest using net-iplsearch in a host list.
7020 21. When running tests using -bh, any delays imposed by "delay" modifiers in
7021 ACLs are no longer actually imposed (and a message to that effect is
7024 22. If a "gecos" field in a passwd entry contained escaped characters, in
7025 particular, if it contained a \" sequence, Exim got it wrong when building
7026 a From: or a Sender: header from that name. A second bug also caused
7027 incorrect handling when an unquoted " was present following a character
7028 that needed quoting.
7030 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
7031 was not being matched caselessly.
7033 24. Arranged for all hyphens in the exim.8 source to be escaped with
7036 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
7037 properly. Recipient callout cache records were still being keyed to include
7038 the sender, even when use_sender was set false. This led to far more
7039 callouts that were necessary. The sender is no longer included in the key
7040 when use_sender is false.
7042 26. Added "control = submission" modifier to ACLs.
7044 27. Added the ${base62d: operator to decode base 62 numbers.
7046 28. dnsdb lookups can now access SRV records.
7048 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
7049 the configuration file.
7051 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
7052 (-v) mode. This makes the output for a verbose queue run more intelligible.
7054 31. Added a use_postmaster feature to recipient callouts.
7056 32. Added the $body_zerocount variable, containing the number of binary zero
7057 bytes in the message body.
7059 33. The time of last modification of the "new" subdirectory is now used as the
7060 "mailbox time last read" when there is a quota error for a maildir
7063 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
7065 35. Added +ignore_unknown as a special item in host lists.
7067 36. Code for decoding IPv6 addresses in host lists is now included, even if
7068 IPv6 support is not being compiled. This fixes a bug in which an IPv6
7069 address was recognized as an IP address, but was then not correctly decoded
7070 into binary, causing unexpected and incorrect effects when compared with
7077 1. Very minor rewording of debugging text in manualroute to say "list of
7078 hosts" instead of "hostlist".
7080 2. If verify=header_syntax was set, and a header line with an unqualified
7081 address (no domain) and a large number of spaces between the end of the
7082 name and the colon was received, the reception process suffered a buffer
7083 overflow, and (when I tested it) crashed. This was caused by some obsolete
7084 code that should have been removed. The fix is to remove it!
7086 3. When running in the test harness, delay a bit after writing a bounce
7087 message to get a bit more predictability in the log output.
7089 4. Added a call to search_tidyup() just before forking a reception process. In
7090 theory, someone could use a lookup in the expansion of smtp_accept_max_
7091 per_host which, without the tidyup, could leave open a database connection.
7093 5. Added the variables $recipient_data and $sender_data which get set from a
7094 lookup success in an ACL "recipients" or "senders" condition, or a router
7095 "senders" option, similar to $domain_data and $local_part_data.
7097 6. Moved the writing of debug_print from before to after the "senders" test
7100 7. Change 4.31/66 (moving the time when the Received: is generated) caused
7101 problems for message scanning, either using a data ACL, or using
7102 local_scan() because the Received: header was not generated till after they
7103 were called (in order to set the time as the time of reception completion).
7104 I have revised the way this works. The header is now generated after the
7105 body is received, but before the ACL or local_scan() are called. After they
7106 are run, the timestamp in the header is updated.
7112 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
7113 before starting a queue runner without re-exec. This happened only when
7114 deliver_drop_privilege was set or when the Exim user was set to root. The
7115 effect of the bug was that timeouts during subsequent deliveries caused
7116 crashes instead of being properly handled. The handler is now left at its
7117 default (and expected) setting.
7119 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
7120 message, again when deliver_drop_privilege is set or Exim is run as root.
7121 The bug described in (1) was not present in this case, but the tidying up
7122 of the other signals was missing. I have made the two cases consistent.
7124 3. The ignore_target_hosts setting on a manualroute router was being ignored
7125 for hosts that were looked up using the /MX notation.
7127 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
7130 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
7131 operated on the sender address. After changing the $sender_address to <>
7132 for the sender address verify, Exim was re-instated it as the original
7133 (before rewriting) address, but remembering that it had rewritten it, so it
7134 wasn't rewriting it again. This bug also had the effect of breaking the
7135 sender address verification caching when the sender address was rewritten.
7137 6. The ignore_target_hosts option was being ignored by the ipliteral router.
7138 This has been changed so that if the ip literal address matches
7139 ignore_target_hosts, the router declines.
7141 7. Added expansion conditions match_domain, match_address, and match_local_
7142 part (NOT match_host).
7144 8. The placeholder for the Received: header didn't have a length field set.
7146 9. Added code to Exim itself and to exim_lock to test for a specific race
7147 condition that could lead to file corruption when using MBX delivery. The
7148 issue is with the lockfile that is created in /tmp. If this file is removed
7149 after a process has opened it but before that process has acquired a lock,
7150 there is the potential for a second process to recreate the file and also
7151 acquire a lock. This could lead to two Exim processes writing to the file
7152 at the same time. The added code performs the same test as UW imapd; it
7153 checks after acquiring the lock that its file descriptor still refers to
7154 the same named file.
7156 10. The buffer for building added header lines was of fixed size, 8192 bytes.
7157 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
7160 11. Added the smtp_active_hostname option. If used, this will typically be made
7161 to depend on the incoming interface address. Because $interface_address is
7162 not set up until the daemon has forked a reception process, error responses
7163 that can happen earlier (such as "too many connections") no longer contain
7166 12. If an expansion in a condition on a "warn" statement fails because a lookup
7167 defers, the "warn" statement is abandoned, and the next ACL statement is
7168 processed. Previously this caused the whole ACL to be aborted.
7170 13. Added the iplsearch lookup type.
7172 14. Added ident_timeout as a log selector.
7174 15. Added tls_certificate_verified as a log selector.
7176 16. Added a global option tls_require_ciphers (compare the smtp transport
7177 option of the same name). This controls incoming TLS connections.
7179 17. I finally figured out how to make tls_require_ciphers do a similar thing
7180 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
7181 before starting the TLS session.
7183 18. Tabs are now shown as \t in -bP output.
7185 19. If the log selector return_path_on_delivery was set, Exim crashed when
7186 bouncing a message because it had too many Received: header lines.
7188 20. If two routers both had headers_remove settings, and the first one included
7189 a superfluous trailing colon, the final name in the first list and the
7190 first name in the second list were incorrectly joined into one item (with a
7191 colon in the middle).
7197 1. Added -C and -D options to the exinext utility, mainly to make it easier
7198 to include in the automated testing, but these could be helpful when
7199 multiple configurations are in use.
7201 2. The exinext utility was not formatting the output nicely when there was
7202 an alternate port involved in the retry record key, nor when there was a
7203 message id as well (for retries that were specific to a specific message
7204 and a specific host). It was also confused by IPv6 addresses, because of
7205 the additional colons they contain. I have fixed the IPv4 problem, and
7206 patched it up to do a reasonable job for IPv6.
7208 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
7209 delivery, the log line now contains "pipelined" if PIPELINING was used.
7211 4. An SMTP transport process used to panic and die if the bind() call to set
7212 an explicit outgoing interface failed. This has been changed; it is now
7213 treated in the same way as a connect() failure.
7215 5. A reference to $sender_host_name in the part of a conditional expansion
7216 that was being skipped was still causing a DNS lookup. This no longer
7219 6. The def: expansion condition was not recognizing references to header lines
7220 that used bh_ and bheader_.
7222 7. Added the _cache feature to named lists.
7224 8. The code for checking quota_filecount in the appendfile transport was
7225 allowing one more file than it should have been.
7227 9. For compatibility with Sendmail, the command line option
7235 and sets the incoming protocol and host name (for trusted callers). The
7236 host name and its colon can be omitted when only the protocol is to be set.
7237 Note the Exim already has two private options, -pd and -ps, that refer to
7238 embedded Perl. It is therefore impossible to set a protocol value of "d" or
7239 "s", but I don't think that's a major issue.
7241 10. A number of refactoring changes to the code, none of which should affect
7244 (a) The number of logging options was getting close to filling up the
7245 32-bit word that was used as a bit map. I have split them into two classes:
7246 those that are passed in the argument to log_write(), and those that are
7247 only ever tested independently outside of that function. These are now in
7248 separate 32-bit words, so there is plenty of room for expansion again.
7249 There is no change in the user interface or the logging behaviour.
7251 (b) When building, for example, log lines, the code previously used a
7252 macro that called string_cat() twice, in order to add two strings. This is
7253 not really sufficiently general. Furthermore, there was one instance where
7254 it was actually wrong because one of the argument was used twice, and in
7255 one call a function was used. (As it happened, calling the function twice
7256 did not affect the overall behaviour.) The macro has been replaced by a
7257 function that can join an arbitrary number of extra strings onto a growing
7260 (c) The code for expansion conditions now uses a table and a binary chop
7261 instead of a serial search (which was left over from when there were very
7262 few conditions). Also, it now recognizes conditions like "pam" even when
7263 the relevant support is not compiled in: a suitably worded error message is
7264 given if an attempt is made to use such a condition.
7266 11. Added ${time_interval:xxxxx}.
7268 12. A bug was causing one of the ddress fields not to be passed back correctly
7269 from remote delivery subprocesses. The field in question was not being
7270 subsequently used, so this caused to problems in practice.
7272 13. Added new log selectors queue_time and deliver_time.
7274 14. Might have fixed a bug in maildirsizefile handling that threw up
7275 "unexpected character" debug warnings, and recalculated the data
7276 unnecessarily. In any case, I expanded the warning message to give more
7279 15. Added the message "Restricted characters in address" to the statements in
7280 the default ACL that block characters like @ and % in local parts.
7282 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
7283 Three changes have been made:
7285 (a) There was a serious bug; a negative response to MAIL caused the whole
7286 recipient domain to be cached as invalid, thereby blocking all messages
7287 to all local parts at the same domain, from all senders. This bug has
7288 been fixed. The domain is no longer cached after a negative response to
7289 MAIL if the sender used is not empty.
7291 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
7294 (c) A new callout option, "use_sender" has been added for people who want
7295 the modified behaviour.
7301 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
7304 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
7305 indeed breaks things for older releases.
7307 3. Added additional logging to the case where there is a problem reading data
7308 from a filter that is running in a subprocess using a pipe, in order to
7309 try to track down a specific problem.
7311 4. Testing facility fudge: when running in the test harness and attempting
7312 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
7313 getting "No route to host". Convert this to a timeout.
7315 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
7318 6. Some OS don't have socklen_t but use size_t instead. This affects the
7319 fifth argument of getsockopt() amongst other things. This is now
7320 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
7321 can be set for individual OS. I have set it for SunOS5, OSF1, and
7322 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
7323 some earlier ones do not.
7325 7. Change 4.30/15 was not doing the test caselessly.
7327 8. The standard form for an IPv6 address literal was being rejected by address
7328 parsing in, for example, MAIL and RCPT commands. An example of this kind of
7329 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
7330 this, as well as the form without the "IPv6" on the front (but only when
7331 address literals are enabled, of course).
7333 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
7335 10. Exim crashed if a message with an empty sender address specified by -f
7336 encountered a router with an errors_to setting. This could be provoked only
7337 by a command such as
7341 where an empty string was supplied; "<>" did not hit this bug.
7343 11. Installed PCRE release 4.5.
7345 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
7346 remained set. It is now erased.
7348 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
7349 times from message ids (which are base 36 rather than the normal 62).
7351 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
7352 were being counted as actual protocol errors, and logged if the log
7353 selector +smtp_protocol_error was set. One cannot be perfect in this test,
7354 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
7355 and DATA following a set of rejected RCPTs do not count as protocol errors.
7356 In other words, Exim assumes they were pipelined, though this may not
7357 actually be the case. Of course, in all cases the client gets an
7358 appropriate error code.
7360 15. If a lookup fails in an ACL condition, a message about the failure may
7361 be available; it is used if testing the ACL cannot continue, because most
7362 such messages specify what the cause of the deferral is. However, some
7363 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
7364 that caused an old message to be retained and used if a later statement
7365 caused a defer, replacing the real cause of the deferral.
7367 16. If an IP address had so many PTR records that the DNS lookup buffer
7368 was not large enough to hold them, Exim could crash while trying to process
7369 the truncated data. It now detects and logs this case.
7371 17. Further to 4.21/58, another change has been made: if (and only if) the
7372 first line of a message (the first header line) ends with CRLF, a bare LF
7373 in a subsequent header line has a space inserted after it, so as not to
7374 terminate the header.
7376 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
7377 unnecessarily, used atoi() instead of strtol(), and didn't check the
7378 termination when getting file sizes from file names by regex.
7380 19. Completely re-implemented the support for maildirsize files, in the light
7381 of a number of problems with the previous contributed implementation
7382 (4.30/29). In particular:
7384 . If the quota is zero, the maildirsize file is maintained, but no quota is
7387 . If the maildir directory does not exist, it is created before any attempt
7388 to write a maildirsize file.
7390 . The quota value in the file is just a cache; if the quota is changed in
7391 the transport, the new value overrides.
7393 . A regular expression is available for excluding directories from the
7396 20. The autoreply transport checks the characters in options that define the
7397 message's headers; it allows continued headers, but it was checking with
7398 isspace() after an embedded newline instead of explicitly looking for a
7401 21. If all the "regular" hosts to which an address was routed had passed their
7402 expiry times, and had not reached their retry times, the address was
7403 bounced, even if fallback hosts were defined. Now Exim should go on to try
7406 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
7407 equivalent code in the SMTP transport. Some hosts send humungous responses
7408 to HELO/EHLO, more than 1024 it seems.
7410 23. Refactoring: code in filter.c used (void *) for "any old type" but this
7411 gives compiler warnings in some environments. I've now done it "properly",
7414 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
7415 (because of problems with the built-in one) was declared to return uschar *
7416 instead of char *, causing compiler failure.
7418 25. Fixed a file descriptor leak when processing alias/forward files.
7420 26. Fixed a minor format string issue in dbfn.c.
7422 27. Typo in exim.c: ("dmbnz" for "dbmnz").
7424 28. If a filter file refered to $h_xxx or $message_headers, and the headers
7425 contained RFC 2047 "words", Exim's memory could, under certain conditions,
7428 29. When a sender address is verified, it is cached, to save repeating the test
7429 when there is more than one recipient in a message. However, when the
7430 verification involves a callout, it is possible for different callout
7431 options to be set for different recipients. It is too complicated to keep
7432 track of this in the cache, so now Exim always runs a verification when a
7433 callout is required, relying on the callout cache for the optimization.
7434 The overhead is duplication of the address routing, but this should not be
7437 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
7438 to be verified with callout=postmaster, and the main callout worked but the
7439 postmaster check failed, the verification correctly failed. However, if a
7440 subsequent RCPT command asked for sender verification *without* the
7441 postmaster check, incorrect caching caused this verification also to fail,
7444 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
7445 it was not caching the DNS options (qualify_single, search_parents) that
7446 were used when the lookup failed. A subsequent lookup with different
7447 options therefore always gave the same answer, though there were cases
7448 where it should not have. (Example: a "domains = !$mx_any" option on a
7449 dnslookup router: the "domains" option is always processed without any
7450 widening, but the router might have qualify_single set.) Now Exim uses the
7451 cached value only when the same options are set.
7453 32. Added John Jetmore's "exipick" utility to the distribution.
7455 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
7456 than a timeout (e.g. a certificate is required, and is not provided), an
7457 Exim server now closes the connection immediately. Previously it waited for
7458 the client to close - but if the client is SSL, it seems that they each
7459 wait for each other, leading to a delay before one of them times out.
7461 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
7462 maintained 0.8.x compatibility because I don't think many are using it, and
7463 it is clearly obsolete.
7465 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
7468 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
7469 client certificate was expired. A simple patch fixes this, though I don't
7470 understand the full logic of why the verify callback is called multiple
7473 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
7474 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
7475 which causes problems with some clients (such as the Certicom SSL Plus
7476 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
7477 disables the coutermeasure allowing Eudora to connect."
7479 38. Exim was not checking that a write() to a log file succeeded. This could
7480 lead to Bad Things if a log got too big, in particular if it hit a file
7481 size limit. Exim now panics and dies if it cannot write to a log file, just
7482 as it does if it cannot open a log file.
7484 39. Modified OS/Makefile-Linux so that it now contains
7486 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
7488 The two -D definitions ensure that Exim is compiled with large file
7489 support, which makes it possible to handle log files that are bigger than
7492 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
7493 instance) a domain was checked against a named list that involved a lookup,
7494 causing $domain_data to be set, then another domain was checked against the
7495 same list, then the first domain was re-checked, the value of $domain_data
7496 after the final check could be wrong. In particular, if the second check
7497 failed, it could be set empty. This bug probably also applied to
7500 41. The strip_trailing_dot option was not being applied to the address given
7501 with the -f command-line option.
7503 42. The code for reading a message's header from the spool was incrementing
7504 $received_count, but never initializing it. This meant that the value was
7505 incorrect (doubled) while delivering a message in the same process in which
7506 it was received. In the most common configuration of Exim, this never
7507 happens - a fresh exec is done - but it can happen when
7508 deliver_drop_privilege is set.
7510 43. When Exim logs an SMTP synchronization error - client data sent too soon -
7511 it now includes up to 150 characters of the unexpected data in the log
7514 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
7515 and building data strings. The size of both of these buffers was 10 000
7516 bytes - far larger than anybody would *ever* want, thought I. Needless to
7517 say, somebody hit the limit. I have increased the maximum line length to
7518 20 000 and the maximum data length of concatenated lines to 100 000. I have
7519 also fixed two bugs, because there was no checking on these buffers. Tsk,
7520 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
7521 buffer is too small.
7523 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
7524 lsearch lookups. Now it does.
7526 46. When parsing a route_list item in a manualroute router, a fixed-length
7527 buffer was used for the list of hosts. I made this 1024 bytes long,
7528 thinking that nobody would ever have a list of hosts that long. Wrong.
7529 Somebody had a whole pile of complicated expansion conditions, and the
7530 string was silently truncated, leading to an expansion error. It turns out
7531 that it is easier to change to an unlimited length (owing to other changes
7532 that have happened since this code was originally written) than to build
7533 structure for giving a limitation error. The length of the item that
7534 expands into the list of hosts is now unlimited.
7536 47. The lsearch lookup could not handle data where the length of text line was
7537 more than 4095 characters. Such lines were truncated, leading to shortened
7538 data being returned. It should now handle lines of any length.
7540 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
7541 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
7544 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
7545 replacing strings such as BIN_DIRECTORY in a source file: the replacement
7546 no longer happens in comment lines. A list of replacements is now placed
7547 at the head of all of the source files, except those whose only change is
7548 to replace PERL_COMMAND in the very first #! line.
7550 50. Replaced the slow insertion sort in queue.c, for sorting the list of
7551 messages on the queue, with a bottom-up merge sort, using code contributed
7552 by Michael Haardt. This should make operations like -bp somewhat faster on
7553 large queues. It won't affect queue runners, except when queue_run_in_order
7556 51. Installed eximstats 1.31 in the distribution.
7558 52. Added support for SRV lookups to the dnslookup router.
7560 53. If an ACL referred to $message_body or $message_body_end, the value was not
7561 reset for any messages that followed in the same SMTP session.
7563 54. The store-handling optimization for building very long strings was not
7564 differentiating between the different store pools. I don't think this
7565 actually made any difference in practice, but I've tidied it.
7567 55. While running the routers to verify a sender address, $sender_address
7568 was still set to the sender address. This is wrong, because when routing to
7569 send a bounce to the sender, it would be empty. Therefore, I have changed
7570 it so that, while verifying a sender address, $sender_address is set to <>.
7571 (There is no change to what happens when verifying a recipient address.)
7573 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
7574 target A or AAAA records (if not already returned) without resetting the
7575 qualify_single or search_parents options of the DNS resolver. These are
7576 inappropriate in this case because the targets of MX and SRV records must
7577 be FQDNs. A broken DNS record could cause trouble if it happened to have a
7578 target that, when qualified, matched something in the local domain. These
7579 two options are now turned off when doing these lookups.
7581 57. It seems that at least some releases of Reiserfs (which does not have the
7582 concept of a fixed number of inodes) returns zero and not -1 for the
7583 number of available inodes. This interacted badly with check_spool_inodes,
7584 which assumed that -1 was the "no such thing" setting. What I have done is
7585 to check that the total number of inodes is greater than zero before doing
7586 the test of how many are available.
7588 58. When a "warn" ACL statement has a log_message modifier, the message is
7589 remembered, and not repeated. This is to avoid a lot of repetition when a
7590 message has many recipients that cause the same warning to be written.
7591 However, Exim was preserving the list of already written lines for an
7592 entire SMTP session, which doesn't seem right. The memory is now reset if a
7593 new message is started.
7595 59. The "rewrite" debugging flag was not showing the result of rewriting in the
7596 debugging output unless log_rewrite was also set.
7598 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
7599 of (int)(handle) when we know that handle contains (void *)(-1).
7601 61. The Exim daemon panic-logs an error return when it closes the incoming
7602 connection. However "connection reset by peer" seems to be common, and
7603 isn't really an error worthy of noting specially, so that particular error
7606 62. When Exim is trying to find all the local interfaces, it used to panic and
7607 die if the ioctl to get the interface flags failed. However, it seems that
7608 on at least one OS (Solaris 9) it is possible to have an interface that is
7609 included in the list of interfaces, but for which you get a failure error
7610 for this call. This happens when the interface is not "plumbed" into a
7611 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
7612 failure of the "get flags" call assumes that the interface is down.
7614 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
7615 makes life easier for people who are doing arithmetic on fields extracted
7616 from dates, where you often get leading zeros that should not be
7617 interpreted as octal.
7619 64. Added qualify_domain to the redirect router, to override the global
7622 65. If a pathologically long header line contained very many addresses (the
7623 report of this problem mentioned 10 000) and each of them was rewritten,
7624 Exim could use up a very large amount of memory. (It kept on making new
7625 copies of the header line as it rewrote, and never released the old ones.)
7626 At the expense of a bit more processing, the header rewriting function has
7627 been changed so that it no longer eats memory in this way.
7629 66. The generation of the Received: header has been moved from the time that a
7630 message starts to be received, to the time that it finishes. The timestamp
7631 in the Received: header should now be very close to that of the <= log
7632 line. There are two side-effects of this change:
7634 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
7635 logged header lines no longer include the local Received: line, because
7636 it has not yet been created. The same applies to a copy of the message
7637 that is returned to a non-SMTP sender when a message is rejected.
7639 (b) When a filter file is tested using -bf, no additional Received: header
7640 is added to the test message. After some thought, I decided that this
7643 This change does not affect the value of $received_for. It is still set
7644 after address rewriting, but before local_scan() is called.
7646 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
7648 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
7649 gave an unhelpful panic error message, and a defer error. I have managed to
7650 change this behaviour so that it now rejects any supplied certificate,
7651 which seems right, as the list of acceptable certificates is empty.
7653 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
7654 gave an unhelpful defer error. I have not managed to make this reject any
7655 supplied certificates, but the error message it gives is "no certificate
7656 supplied", which is not helpful.
7658 70. exigrep's output now also includes lines that are not associated with any
7659 message, but which match the given pattern. Implemented by a patch from
7660 Martin Sluka, which also tidied up the Perl a bit.
7662 71. Recipient callout verification, like sender verification, was using <> in
7663 the MAIL FROM command. This isn't really the right thing, since the actual
7664 sender may affect whether the remote host accepts the recipient or not. I
7665 have changed it to use the actual sender in the callout; this means that
7666 the cache record is now keyed on a recipient/sender pair, not just the
7667 recipient address. There doesn't seem to be a real danger of callout loops,
7668 since a callout by the remote host to check the sender would use <>.
7669 [SEE ABOVE: changed after hitting problems.]
7671 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
7672 temporary errors. However, in the case of such a code being given after
7673 the end of a data transmission (i.e. after ".") Exim was failing to write
7674 a retry record for the message. (Yes, there was some broken host that was
7675 actually sending 8xx at this point.)
7677 73. An unknown lookup type in a host list could cause Exim to panic-die when
7678 the list was checked. (An example that provoked this was putting <; in the
7679 middle of a list instead of at the start.) If this happened during a DATA
7680 ACL check, a -D file could be left lying around. This kind of configuration
7681 error no longer causes Exim to die; instead it causes a defer error. The
7682 incident is still logged to the main and panic logs.
7684 74. Buglet left over from Exim 3 conversion. The message "too many messages
7685 in one connection" was written to the rejectlog but not the mainlog, except
7686 when address rewriting (yes!) was being logged.
7688 75. Added write_rejectlog option.
7690 76. When a system filter was run not as root (that is, when system_filter_user
7691 was set), the values of the $n variables were not being returned to the
7692 main process; thus, they were not subsequently available in the $sn
7695 77. Added +return_path_on_delivery log selector.
7697 78. A connection timeout was being treated differently from recipients deferred
7698 when testing hosts_max_try with a message that was older than the host's
7699 retry timeout. (The host should not be counted, thus allowing all hosts to
7700 be tried at least once before bouncing.) This may have been the cause of an
7701 occasionally reported bug whereby a message would remain on the queue
7702 longer than the retry timeout, but would be bounced if a delivery was
7703 forced. I say "may" because I never totally pinned down the problem;
7704 setting up timeout/retry tests is difficult. See also the next item.
7706 79. The ultimate address timeout was not being applied to errors that involved
7707 a combination of host plus message (for example, a timeout on a MAIL
7708 command). When an address resolved to a number of possible hosts, and they
7709 were not all tried for each delivery (e.g. because of hosts_max_try), a
7710 message could remain on the queue longer than the retry timeout.
7712 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
7715 81. Fixed an obscure SMTP outgoing bug which required at least the following
7716 conditions: (a) there was another message waiting for the same server;
7717 (b) the server returned 5xx to all RCPT commands in the first message so
7718 that the message was not completed; (c) the server dropped the connection
7719 or gave a negative response to the RSET that Exim sends to abort the
7720 transaction. The observed case was a dropped connection after DATA that had
7721 been sent in pipelining mode. That is, the server had advertised PIPELINING
7722 but was not implementing it correctly. The effect of the bug was incorrect
7723 behaviour, such as trying another host, and this could lead to a crash.
7729 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
7730 and daemon.c were passed as pointers to ints; they should have been
7731 pointers to socklen_t variables (which are typically unsigned ints).
7733 2. Some signed/unsigned type warnings in the os.c file for Linux have been
7736 3. Fixed a really odd bug that affected only the testing scheme; patching a
7737 certain fixed string in the binary changed the value of another string that
7738 happened to be identical to the end of the original first string.
7740 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
7741 name", it returns that address as the IP address. On some operating
7742 systems (e.g. Solaris), it also passes back the IP address string as the
7743 "host name". However, on others (e.g. Linux), it passes back an empty
7744 string. Exim wasn't checking for this, and was changing the host name to an
7745 empty string, assuming it had been canonicalized.
7747 5. Although rare, it is permitted to have more than one PTR record for a given
7748 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
7749 all the names associated with an address, because they do in Solaris.
7750 However, it seems that they do not in Linux for data that comes from the
7751 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
7752 I found this out when I moved to a new Linux workstation and tried to run
7753 the Exim test suite.
7755 To get round this problem I have changed the code so that it now does its
7756 own call to the DNS to look up PTR records when searching for a host name.
7757 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
7758 addresses that are only in /etc/hosts are still found.
7760 This behaviour is, however, controlled by an option called host_lookup_
7761 order, which defaults to "bydns:byaddr". If people want to use the other
7762 order, or indeed, just use one or the other means of lookup, they can
7763 specify it in this variable.
7765 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
7766 some operating systems, this comes back from gethostbyaddr() as an empty
7767 string, and this is what Exim used to test for. However, it seems that in
7768 other systems, "." is yielded. Exim now tests for this case too.
7770 7. The values of check_spool_space and check_log_space are now held internally
7771 as a number of kilobytes instead of an absolute number of bytes. If a
7772 numbers is specified without 'K' or 'M', it is rounded up to the nearest
7773 kilobyte. This means that much larger values can be stored.
7775 8. Exim monitor: an attempt to get the action menu when not actually pointing
7776 at a message produces an empty menu entitled "No message selected". This
7777 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
7778 no entries in it ("Shell widget menu has zero width and/or height"). So I
7779 have added a single, blank menu entry in this case.
7781 9. Added ${quote_local_part.
7783 10. MIME decoding is now applied to the contents of Subject: header lines when
7786 11. Now that a reference to $sender_host_address automatically causes a reverse
7787 lookup to occur if necessary (4.13/18), there is no need to arrange for a
7788 host lookup before query-style lookups in lists that might use this
7789 variable. This has therefore been abolished, and the "net-" prefix is no
7790 longer necessary for query-style lookups.
7792 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
7793 have been a typo for LFLAGS, so it has been changed.
7795 13. The install script calls Exim with "-C /dev/null" in order to find the
7796 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
7797 to be output. However, since Exim outputs its version number before the
7798 error, it didn't break the script. It just looked ugly. I fixed this by
7799 always allowing "-C /dev/null" if the caller is root.
7801 14. Ignore overlarge ACL variable number when reading spool file - insurance
7802 against a later release with more variables having written the file.
7804 15. The standard form for an IPv6 address literal was being rejected by EHLO.
7805 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
7806 this, as well as the form without the "IPv6" on the front.
7808 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
7809 OS/Makefile-Darwin file.
7811 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
7812 only to LDAP libraries that do not have LDAP_OPT_DEREF.
7814 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
7815 variable. However, some OS use ints for inodes. I've added cast to long int
7816 to get rid of the compiler warning.
7818 19. I had forgotten to lock out "/../" in configuration file names when
7819 ALT_CONFIG_PREFIX was set.
7821 20. Routers used for verification do not need to specify transports. However,
7822 if such a router generated a host list, and callout was configured, Exim
7823 crashed, because it could not find a port number from the (non-existent)
7824 transport. It now assumes port 25 in this circumstance.
7826 21. Added the -t option to exigrep.
7828 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
7829 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
7830 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
7831 them is set, LOOKUP_LSEARCH is forced.
7833 23. "exim -bV" now outputs a list of lookups that are included in the binary.
7835 24. Added sender and host information to the "rejected by local_scan()" log
7836 line; previously there was no indication of these.
7838 25. Added .include_if_exists.
7840 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
7841 Linux. It turns out that not all file systems support this. Apparently some
7842 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
7843 do it.) To cope with this, the error EINVAL, which means that sync-ing is
7844 not supported on the file descriptor, is now ignored when Exim is trying to
7845 sync a directory. This applies only to Linux.
7847 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
7849 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
7850 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
7853 29. Added support for maildirsize files from supplied patch (modified a bit).
7855 30. The use of :fail: followed by an empty string could lead Exim to respond to
7856 sender verification failures with (e.g.):
7858 550 Verification failed for <xxx>
7859 550 Sender verify failed
7861 where the first response line was missing the '-' that indicates it is not
7862 the final line of the response.
7864 31. The loop for finding the name of the user that called Exim had a hardwired
7865 limit of 10; it now uses the value of finduser_retries, which is used for
7866 all other user lookups.
7868 32. Added $received_count variable, available in data and not_smtp ACLs, and at
7871 33. Exim was neglecting to zero errno before one call of strtol() when
7872 expanding a string and expecting an integer value. On some systems this
7873 resulted in spurious "integer overflow" errors. Also, it was casting the
7874 result into an int without checking.
7876 34. Testing for a connection timeout using "timeout_connect" in the retry rules
7877 did not work. The code looks as if it has *never* worked, though it appears
7878 to have been documented since at least release 1.62. I have made it work.
7880 35. The "timeout_DNS" error in retry rules, also documented since at least
7881 1.62, also never worked. As it isn't clear exactly what this means, and
7882 clearly it isn't a major issue, I have abolished the feature by treating it
7883 as "timeout", and writing a warning to the main and panic logs.
7885 36. The display of retry rules for -brt wasn't always showing the error code
7888 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
7889 timeout_connect_A, timeout_connect_MX.
7891 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
7892 to the empty sender.
7894 39. The daemon was not analysing the content of -oX till after it had closed
7895 stderr and disconnected from the controlling terminal. This meant that any
7896 syntax errors were only noted on the panic log, and the return code from
7897 the command was 0. By re-arranging the code a little, I've made the
7898 decoding happen first, so such errors now appear on stderr, and the return
7899 code is 1. However, the actual setting up of the sockets still happens in
7900 the disconnected process, so errors there are still only recorded on the
7903 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
7904 connections (as happens on some IP stacks) was logged at start up time as
7905 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
7906 it from "IPv6 and IPv4", which means that two separate sockets are being
7909 41. The debug output for gethostbyname2() or getipnodebyname() failures now
7910 says whether AF_INET or AF_INET6 was passed as an argument.
7912 42. Exiwhat output was messed up when time zones were included in log
7915 43. Exiwhat now gives more information about the daemon's listening ports,
7916 and whether -tls-on-connect was used.
7918 44. The "port" option of the smtp transport is now expanded.
7920 45. A "message" modifier in a "warn" statement in a non-message ACL was being
7921 silently ignored. Now an error message is written to the main and panic
7924 46. There's a new ACL modifier called "logwrite" which writes to a log file
7925 as soon as it is encountered.
7927 47. Added $local_user_uid and $local_user_gid at routing time.
7929 48. Exim crashed when trying to verify a sender address that was being
7932 49. Exim was recognizing only a space character after ".include". It now also
7933 recognizes a tab character.
7935 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
7936 extracting the relevant information from the specification. The man page no
7937 longer contains scrambled data for the -d option, and I've added a section
7938 at the front about calling Exim under different names.
7940 51. Added "extra_headers" argument to the "mail" command in filter files.
7942 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
7945 53. Installed eximstats 1.29.
7947 54. Added transport_filter_timeout as a generic transport option.
7949 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
7950 Cc: header lines. This was required by RFC 822, but it not required by RFC
7953 56. Exim used to add From:, Date:, and Message-Id: header lines to any
7954 incoming messages that did not have them. Now it does so only if the
7955 message originates locally, that is, if there is no associated remote host
7956 address. When Resent- header lines are present, this applies to the Resent-
7957 lines rather than the non-Resent- lines.
7959 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
7960 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
7962 58. Messages for configuration errors now include the name of the main
7963 configuration file - useful now that there may be more than one file in a
7964 list (.included file names were always shown).
7966 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
7967 for those rare installations that do not start the daemon as root or run it
7968 setuid root. I've cut out the call to initgroups() if the daemon is not
7971 60. The Exim user and group can now be bound into the binary as text strings
7972 that are looked up at the start of Exim's processing.
7974 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
7976 62. Added $mailstore_basename variable.
7978 63. Installed patch to sieve.c from Michael Haardt.
7980 64. When Exim failed to open the panic log after failing to open the main log,
7981 the original message it was trying to log was written to stderr and debug
7982 output, but if they were not available (the usual case in production), it
7983 was lost. Now it is written to syslog before the two lines that record the
7984 failures to open the logs.
7986 65. Users' Exim filters run in subprocesses under the user's uid. It is
7987 possible for a "deliver" command or an alias in a "personal" command to
7988 provoke an address rewrite. If logging of address rewriting is configured,
7989 this fails because the process is not running as root or exim. There may be
7990 a better way of dealing with this, but for the moment (because 4.30 needs
7991 to be released), I have disabled address rewrite logging when running a
7992 filter in a non-root, non-exim process.
7998 1. The buildconfig auxiliary program wasn't quoting the value set for
7999 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
8000 not defined. This bug was masked in 4.22 by the effect that was fixed in
8003 2. Some messages that were rejected after a message id was allocated were
8004 shown as "incomplete" by exigrep. It no longer does this for messages that
8005 are rejected by local_scan() or the DATA or non-SMTP ACLs.
8007 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
8008 have allow_domain_literals set, the ID did not get logged in the <= line.
8009 Domain literals are now always recognized in Message-ID: header lines.
8011 4. The first argument for a ${extract expansion item is the key name or field
8012 number. Leading and trailing spaces in this item were not being ignored,
8013 causing some misleading effects.
8015 5. When deliver_drop_privilege was set, single queue runner processes started
8016 manually (i.e. by the command "exim -q") or by the daemon (which uses the
8017 same command in the process it spins off) were not dropping privilege.
8019 6. When the daemon running as "exim" started a queue runner, it always
8020 re-executed Exim in the spun-off process. This is a waste of effort when
8021 deliver_drop_privilege is set. The new process now just calls the
8022 queue-runner function directly.
8028 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
8031 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
8032 ignored. Though the use of -oP was forcing the writing of a pid file, it
8033 was always written to the default place.
8035 3. If the message "no IP address found for host xxxx" is generated during
8036 incoming verification, it is now followed by identification of the incoming
8037 connection (so you can more easily find what provoked it).
8039 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
8041 5. Added some features to "harden" Exim a bit more against certain attacks:
8043 (a) There is now a build-time option called FIXED_NEVER_USERS that can
8044 be put in Local/Makefile. This is like the never_users runtime option,
8045 but it cannot be overridden. The default setting is "root".
8047 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
8048 prefix string with which any file named in a -C command line option
8051 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
8052 is retained for -C and -D only if the caller of Exim is root. Without
8053 it, the exim user may also use -C and -D and retain privilege.
8055 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
8056 command line option is disabled.
8058 6. Macro names set by the -D option must start with an upper case letter, just
8059 like macro names defined in the configuration file.
8061 7. Added "dereference=" facility to LDAP.
8063 8. Two instances of the typo "uknown" in the source files are fixed.
8065 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
8066 the Configure-Makefile script screwed up while processing it.
8068 10. Incorporated PCRE 4.4.
8070 11. The SMTP synchronization check was not operating right at the start of an
8071 SMTP session. For example, it could not catch a HELO sent before the client
8072 waited for the greeting. There is now a check for outstanding input at the
8073 point when the greeting is written. Because of the duplex, asynchronous
8074 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
8075 way, but not yet received, when the check is performed.
8077 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
8078 on TCP/IP sockets, because this apparently causes some broken clients to
8081 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
8082 unchanged) from the Cygwin maintainer.
8084 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
8085 was supported instead of testing for mbx. Effectively a typo.
8087 15. The spa authenticator server code was not checking that the input it
8088 received was valid base64.
8090 16. The debug output line for the "set" modifier in ACLs was not showing the
8091 name of the variable that was being set.
8093 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
8095 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
8096 lookup on its own. Something else was needed to trigger the lookup. For
8097 example, a match in host_lookup or the need for a host name in a host list.
8098 Now, if $sender_host_name is referenced and the host name has not yet been
8099 looked up, a lookup is performed. If the lookup fails, the variable remains
8100 empty, and $host_lookup_failed is set to "1".
8102 19. Added "eqi" as a case-independent comparison operator.
8104 20. The saslauthd authentication condition could segfault if neither service
8105 nor realm was specified.
8107 21. If an overflowing value such as "2048M" was set for message_size_limit, the
8108 error message that was logged was misleading, and incoming SMTP
8109 connections were dropped. The message is now more accurate, and temporary
8110 errors are given to SMTP connections.
8112 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
8113 (except RSET) with a 421 error, until QUIT is received. However, it was
8114 failing to send a response to QUIT.
8116 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
8117 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
8118 ACL is now run after the helo_try_verify_hosts code.
8120 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
8121 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
8122 case-independent, so other case variants are also recognized). Apparently
8123 some systems use these upper case variants.
8125 25. If more than two messages were waiting for the same host, and a transport
8126 filter was specified for the transport, Exim sent two messages over the
8127 same TCP/IP connection, and then failed with "socket operation on non-
8128 socket" when it tried to send the third.
8130 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
8132 27. The extern definition of crypt16() in expand.c was not being excluded when
8133 the OS had its own crypt16() function.
8135 28. Added bounce_return_body as a new option, and bounce_return_size_limit
8136 as a preferred synonym for return_size_limit, both as an option and as an
8139 29. Added LIBS=-liconv to OS/Makefile-OSF1.
8141 30. Changed the default configuration ACL to relax the local part checking rule
8142 for addresses that are not in any local domains. For these addresses,
8143 slashes and pipe symbols are allowed within local parts, but the sequence
8144 /../ is explicitly forbidden.
8146 31. SPA server authentication was not clearing the challenge buffer before
8149 32. log_message in a "warn" ACL statement was writing to the reject log as
8150 well as to the main log, which contradicts the documentation and doesn't
8151 seem right (because no rejection is happening). So I have stopped it.
8153 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
8154 However, I am unable to do any testing of this.
8156 34. Fixed an infelicity in the appendfile transport. When checking directories
8157 for a mailbox, to see if any needed to be created, it was accidentally
8158 using path names with one or more superfluous leading slashes; tracing
8159 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
8161 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
8162 contents are added to the log line that is written for every discarded
8163 recipient. (Previously a log_message setting was ignored.)
8165 36. The ${quote: operator now quotes the string if it is empty.
8167 37. The install script runs exim in order to find its version number. If for
8168 some reason other than non-existence or emptiness, which it checks, it
8169 could not run './exim', it was installing it with an empty version number,
8170 i.e. as "exim-". This error state is now caught, and the installation is
8173 38. An argument was missing from the function that creates an error message
8174 when Exim fails to connect to the socket for saslauthd authentication.
8175 This could cause Exim to crash, or give a corrupted message.
8177 39. Added isip, isip4, and isip6 to ${if conditions.
8179 40. The ACL variables $acl_xx are now saved with the message, and can be
8180 accessed later in routers, transports, and filters.
8182 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
8183 strings in the file are not string-expanded.
8185 42. If a MAIL command specified a SIZE value that was too large to fit into an
8186 int variable, the check against message_size_limit failed. Such values are
8187 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
8188 day this will have to be increased, but I don't think I want to be around
8189 when emails are that large.
8196 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
8197 iconv() is not standard in FreeBSD.
8199 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
8200 IPv6 enabled. The observed symptom was a segmentation fault on return from
8201 the function os_common_find_running_interfaces() in src/os.c.
8203 3. In the check_special_case() function in daemon.c I had used "errno" as an
8204 argument name, which causes warnings on some systems. This was basically a
8205 typo, since it was named "eno" in the comments!
8207 4. The code that waits for the clock to tick (at a resolution of some fraction
8208 of a second) so as to ensure message-id uniqueness was always waiting for
8209 at least one whole tick, when it could have waited for less. [This is
8210 almost certainly not relevant at current processor speeds, where it is
8211 unlikely to ever wait at all. But we try to future-proof.]
8213 5. The function that sleeps for a time interval that includes fractions of a
8214 second contained a race. It did not block SIGALRM between setting the
8215 timer, and suspending (a couple of lines later). If the interval was short
8216 and the sigsuspend() was delayed until after it had expired, the suspension
8217 never ended. On busy systems this could lead to processes getting stuck for
8220 6. Some uncommon configurations may cause a lookup to happen in a queue runner
8221 process, before it forks any delivery processes. The open lookup caching
8222 mechanism meant that the open file or database connection was passed into
8223 the delivery process. The problem was that delivery processes always tidy
8224 up cached lookup data. This could cause a problem for the next delivery
8225 process started by the queue runner, because the external queue runner
8226 process does not know about the closure. So the next delivery process
8227 still has data in the lookup cache. In the case of a file lookup, there was
8228 no problem because closing a file descriptor in a subprocess doesn't affect
8229 the parent. However, if the lookup was caching a connection to a database,
8230 the connection was closed, and the second delivery process was likely to
8231 see errors such as "PGSQL: query failed: server closed the connection
8232 unexpectedly". The problem has been fixed by closing all cached lookups
8233 in a queue runner before running a delivery process.
8235 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
8236 seem to have the "const" qualifier which it has on other OS. I've
8239 8. Change 4.21/2 was too strict. It is only if there are two authenticators
8240 *of the same type* (client or server) with the same public name that an
8241 error should be diagnosed.
8243 9. When Exim looked up a host name for an IP address, but failed to find the
8244 original IP address when looking up the host name (a safety check), it
8245 output the message "<ip address> does not match any IP for NULL", which was
8246 confusing, to say the least. The bug was that the host name should have
8247 appeared instead of "NULL".
8249 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
8250 user that is built into the binary, and the -C or -D options is used, root
8251 privilege is dropped before the configuration file is read. In addition,
8252 logging is switched to stderr instead of the normal log files. If the
8253 configuration then re-defines the Exim user, the unprivileged environment
8254 is probably not what is expected, so Exim logs a panic warning message (but
8257 However, if deliver_drop_privilege is set, the unprivileged state may well
8258 be exactly what is intended, so the warning has been cut out in that case,
8259 and Exim is allowed to try to write to its normal log files.
8265 1. smtp_return_error_details was not giving details for temporary sender
8266 or receiver verification errors.
8268 2. Diagnose a configuration error if two authenticators have the same public
8271 3. Exim used not to create the message log file for a message until the first
8272 delivery attempt. This could be confusing when incoming messages were held
8273 for policy or load reasons. The message log file is now created at the time
8274 the message is received, and an initial "Received" line is written to it.
8276 4. The automatically generated man page for command line options had a minor
8277 bug that caused no ill effects; however, a more serious problem was that
8278 the procedure for building the man page automatically didn't always
8279 operate. Consequently, release 4.20 contains an out-of-date version. This
8280 shouldn't happen again.
8282 5. When building Exim with embedded Perl support, the script that builds the
8283 Makefile was calling 'perl' to find its compile-time parameters, ignoring
8284 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
8286 6. The freeze_tell option was not being used for messages that were frozen on
8287 arrival, either by an ACL or by local_scan().
8289 7. Added the smtp_incomplete_transaction log selector.
8291 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
8292 was accepting AUTH without a new EHLO.
8294 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
8295 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
8298 10. Logging of TCP/IP connections (when configured) now happens in the main
8299 daemon process instead of the child process, so that the TCP/IP connection
8300 count is more accurate (but it can never be perfect).
8302 11. The use of "drop" in a nested ACL was not being handled correctly in the
8303 outer ACL. Now, if condition failure induced by the nested "drop" causes
8304 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
8305 or "require"), the connection is dropped.
8307 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
8308 that yield "discard" can now be used with an "accept" or a "discard" verb,
8309 but an error is generated for any others (because I can't see a useful way
8310 to define what should happen).
8312 13. When an ACL is read dynamically from a file (or anywhere else), the lines
8313 are now processed in the same way as lines in the Exim configuration file.
8314 In particular, continuation lines are supported.
8316 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
8318 15. Added -ti meaning -t -i.
8320 16. Check for letters, digits, hyphens, and dots in the names of dnslist
8321 domains, and warn by logging if others are found.
8323 17. At least on BSD, alignment is not guaranteed for the array of ifreq's
8324 returned from GIFCONF when Exim is trying to find the list of interfaces on
8325 a host. The code in os.c has been modified to copy each ifreq to an aligned
8326 structure in all cases.
8328 Also, in some cases, the returned ifreq's were being copied to a 'struct
8329 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
8330 means the last couple of bytes of an IPv6 address could be chopped if the
8331 ifreq contained only a normal sockaddr (14 bytes storage).
8333 18. Named domain lists were not supported in the hosts_treat_as_local option.
8334 An entry such as +xxxx was not recognized, and was treated as a literal
8337 19. Ensure that header lines added by a DATA ACL are included in the reject log
8338 if the ACL subsequently rejects the message.
8340 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
8341 MD5 (which is deprecated).
8343 21. When testing a filter file using -bf, Exim was writing a message when it
8344 took the sender from a "From " line in the message, but it was not doing so
8345 when it took $return_path from a Return-Path: header line. It now does.
8347 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
8348 with a valid header line field name (a series of printing characters
8349 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
8351 23. Changed "disc" in the source to "disk" to conform to the documentation and
8352 the book and for uniformity.
8354 24. Ignore Sendmail's -Ooption=value command line item.
8356 25. When execve() failed while trying to run a command in a pipe transport,
8357 Exim was returning EX_UNAVAILABLE (69) from the subprocess. However, this
8358 could be confused with a return value of 69 from the command itself. This
8359 has been changed to 127, the value the shell returns if it is asked to run
8360 a non-existent command. The wording for the related log line suggests a
8361 non-existent command as the problem.
8363 26. If received_header_text expands to an empty string, do not add a Received:
8364 header line to the message. (Well, it adds a token one on the spool, but
8365 marks it "old" so that it doesn't get used or transmitted.)
8367 27. Installed eximstats 1.28 (addition of -nt option).
8369 28. There was no check for failure on the call to getsockname() in the daemon
8370 code. This can fail if there is a shortage of resources on the system, with
8371 ENOMEM, for example. A temporary error is now given on failure.
8373 29. Contrary to the C standard, it seems that in some environments, the
8374 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
8375 program. Exim now does this explicitly; it affects the formatting of
8376 timestamps using strftime().
8378 30. If exiqsumm was given junk data, it threw up some uninitialized variable
8379 complaints. I've now initialized all the variables, to avoid this.
8381 32. Header lines added by a system filter were not being "seen" during
8382 transport-time rewrites.
8384 33. The info_callback() function passed to OpenSSL is set up with type void
8385 (*)(SSL *, int, int), as described somewhere. However, when calling the
8386 function (actually a macro) that sets it up, the type void(*)() is
8387 expected. I've put in a cast to prevent warnings from picky compilers.
8389 34. If a DNS black list lookup found a CNAME record, but there were no A
8390 records associated with the domain it pointed at, Exim crashed.
8392 35. If a DNS black list lookup returned more than one A record, Exim ignored
8393 all but the first. It now scans all returned addresses if a particular IP
8394 value is being sought. In this situation, the contents of the
8395 $dnslist_value variable are a list of all the addresses, separated by a
8398 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
8399 to accept a host name and all its aliases if the forward lookup for any of
8400 them yielded the IP address of the incoming connection. Now it accepts only
8401 those names whose forward lookup yields the correct IP address. Any other
8402 names are discarded. This closes a loophole whereby a rogue DNS
8403 administrator could create reverse DNS records to break through a
8404 wildcarded host restriction in an ACL.
8406 37. If a user filter or a system filter that ran in a subprocess used any of
8407 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
8408 the wrong values were passed to the pipe command ($thisaddress had the
8409 value of $0, $0 had the value of $1, etc). This bug was introduced by
8410 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
8412 38. Improved the line breaking for long SMTP error messages from ACLs.
8413 Previously, if there was no break point between 40 and 75 characters, Exim
8414 left the rest of the message alone. Two changes have been made: (a) I've
8415 reduced the minimum length to 35 characters; (b) if it can't find a break
8416 point between 35 and 75 characters, it looks ahead and uses the first one
8417 that it finds. This may give the occasional overlong line, but at least the
8418 remaining text gets split now.
8420 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
8421 file descriptors might be low, and that setting 1000 would always raise it.
8422 It turns out that in some environments, the limit is already over 1000 and
8423 that lowering it causes trouble. So now Exim takes care not to decrease it.
8425 40. When delivering a message, the value of $return_path is set to $sender_
8426 address at the start of routing (routers may change the value). By an
8427 oversight, this default was not being set up when an address was tested by
8428 -bt or -bv, which affected the outcome if any router or filter referred to
8431 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
8432 log only when "message" or "log_message" are set. However, if one of the
8433 conditions was an address verification, or a call to a nested ACL, the
8434 messages generated by the underlying test were being passed through. This
8435 no longer happens. The underlying message is available in $acl_verify_
8436 message for both "message" and "log_message" expansions, so it can be
8437 passed through if needed.
8439 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
8440 new expansion $bh_ to give the encoded byte string without charset
8441 translation. Translation happens only if iconv() is available; HAVE_ICONV
8442 indicates this at build time. HEADERS_CHARSET gives the charset to
8443 translate to; headers_charset can change it in the configuration, and
8444 "headers charset" can change it in an individual filter file.
8446 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
8447 that creates RFC 2047 encoded "words" labels them as that charset instead
8448 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
8449 expansion operator; (ii) when Exim creates a From: line for a local
8450 message; (iii) when a header line is rewritten to include a "phrase" part.
8452 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
8453 buggy, causing it to skip the first lines of messages whose message ID
8454 ended in 'D'. This would not have bitten before Exim release 4.14, because
8455 message IDs were unlikely to end in 'D' before then. The effect was to have
8456 incorrect size information for certain domains.
8458 45. #include "config.h" was missing at the start of the crypt16.c module. This
8459 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
8462 46. If there was a timeout during a "random" callout check, Exim treated it as
8463 a failure of the random address, and carried on sending RSET and the real
8464 address. If the delay was just some slowness somewhere, the response to the
8465 original RCPT would be taken as a response to RSET and so on, causing
8466 mayhem of various kinds.
8468 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
8469 when I implemented it. It didn't allow for the fact that some option values
8470 may legitimately be negative (e.g. size_addition), and it didn't even do
8471 the right test for positive values.
8473 48. Domain names in DNS records are case-independent. Exim always looks them up
8474 in lower case. Some resolvers return domain names in exactly the case they
8475 appear in the zone file, that is, they may contain uppercase letters. Not
8476 all resolvers do this - some return always lower case. Exim was treating a
8477 change of case by a resolver as a change of domain, similar to a widening
8478 of a domain abbreviation. This triggered its re-routing code and so it was
8479 trying to route what was effectively the same domain again. This normally
8480 caused routing to fail (because the router wouldn't handle the domain
8481 twice). Now Exim checks for this case specially, and just changes the
8482 casing of the domain that it ultimately uses when it transmits the message
8485 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
8488 50. If a filter generated a file delivery with a non-absolute name (possible if
8489 no home directory exists for the router), the forbid_file option was not
8492 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
8493 the existing equality matching.
8495 52. Exim was using ints instead of ino_t variables in some places where it was
8496 dealing with inode numbers.
8498 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
8499 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
8500 called TMPDIR, and if it finds it is different, it changes its value.
8502 54. The smtp_printf() function is now made available to local_scan() so
8503 additional output lines can be written before returning. There is also an
8504 smtp_fflush() function to enable the detection of a dropped connection.
8505 The variables smtp_input and smtp_batched_input are exported to
8508 55. Changed the default runtime configuration: the message "Unknown user"
8509 has been removed from the ACL, and instead placed on the localuser router,
8510 using the cannot_route_message feature. This means that any verification
8511 failures that generate their own messages won't get overridden. Similarly,
8512 the "Unrouteable address" message that was in the ACL for unverifiable
8513 relay addresses has also been removed.
8515 56. Added hosts_avoid_esmtp to the smtp transport.
8517 57. The exicyclog script was not checking for the esoteric option
8518 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
8519 will work only if exicyclog is run under the appropriate euid.
8521 58. Following a discussion on the list, the rules by which Exim recognises line
8522 endings on incoming messages have been changed. The -dropcr and drop_cr
8523 options are now no-ops, retained only for backwards compatibility. The
8524 following line terminators are recognized: LF CRLF CR. However, special
8525 processing applies to CR:
8527 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
8528 nor a local message in the state where . is a terminator.
8530 (ii) If a bare CR is encountered in a header line, an extra space is added
8531 after the line terminator so as not to end the header. The reasoning
8532 behind this is that bare CRs in header lines are most likely either
8533 to be mistakes, or people trying to play silly games.
8535 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
8536 was being incorrectly given as 18 bytes larger than it should have been.
8537 This is a VOB (very old bug).
8539 60. This may never have affected anything current, but just in case it has:
8540 When the local host is found other than at the start of a list of hosts,
8541 the local host, those with the same MX, and any that follow, are discarded.
8542 When the list in question was part of a longer list of hosts, the following
8543 hosts (not currently being processed) were also being discarded. This no
8544 longer happens. I'm not sure if this situation could ever has previously
8547 61. Added the "/MX" feature to lists of hosts in the manualroute and query
8550 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
8551 header. This is something that is recommended in a new Internet Draft, and
8552 is something that is documented as being done by Sendmail. There are two
8553 possible values. For messages generated by the autoreply transport, Exim
8556 Auto-Submitted: auto-replied
8558 whereas for all other generated messages (e.g. bounces) it adds
8560 Auto-Submitted: auto-generated
8562 63. The "personal" condition in filters now includes a test for the
8563 Auto-Submitted: header. If it contains the string "auto-" the message it
8564 not considered personal.
8566 64. Added rcpt_include_affixes as a generic transport option.
8568 65. Added queue_only_override (default true).
8570 66. Added the syslog_duplication option.
8572 67. If what should have been the first header line of a message consisted of
8573 a space followed by a colon, Exim was mis-interpreting it as a header line.
8574 It isn't of course - it is syntactically invalid and should therefore be
8575 treated as the start of the message body. The misbehaviour could have
8576 caused a number of strange effects, including loss of data in subsequent
8577 header lines, and spool format errors.
8579 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
8580 client host had authenticated. This control can now be exercised by an ACL
8581 for more flexibility.
8583 69. By default, callouts do not happen when testing with -bh. There is now a
8584 variant, -bhc, which does actually run the callout code, including
8585 consulting and updating the callout cache.
8587 70. Added support for saslauthd authentication, courtesy of Alexander
8590 71. If statvfs() failed on the spool or log directories while checking their
8591 size for availability, Exim confusingly gave the error "space shortage".
8592 Furthermore, in debugging mode it crashed with a floating point exception.
8593 These checks are done if check_{spool,log}_{space,inodes} are set, and when
8594 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
8595 serious problem, Exim now writes to the main and panic logs when this
8596 happens, with details of the failure. It then refuses to accept the
8597 incoming message, giving the message "spool directory problem" or "log
8598 directory problem" with a 421 code for SMTP messages.
8600 72. When Exim is about to re-exec itself, it ensures that the file descriptors
8601 0, 1, and 2 exist, because some OS complain for execs without them (see
8602 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
8603 descriptors. However, the code omitted to check that the open succeeded,
8604 causing mysterious errors if for some reason the permissions on /dev/null
8605 got screwed. Now Exim writes a message to the main and panic logs, and
8606 bombs out if it can't open /dev/null.
8608 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
8609 interact so that it is all more flexible. It is supposed to remain
8610 backwards compatible. Also added extra_local_interfaces.
8612 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
8613 to bomb out with an assertion failure - to the client this appears as a
8614 connection drop. This problem occurs in the part of the code that was taken
8615 from the Samba project. Fortunately, the assertion is in a very simple
8616 function, so I have fixed this by reproducing the function inline in the
8617 one place where it is called, and arranging for authentication to fail
8618 instead of killing the process with assert().
8620 75. The SPA client code was not working when the server requested OEM rather
8621 than Unicode encoding.
8623 76. Added code to make require_files with a specific uid setting more usable in
8624 the case where statting the file as root fails - usually a non-root-mounted
8625 NFS file system. When this happens and the failure is EACCES, Exim now
8626 forks a subprocess and does the per-uid checking as the relevant uid.
8628 77. Added process_log_path.
8630 78. If log_file_path was not explicitly set, a setting of check_log_space or
8631 check_log_inodes was ignored.
8633 79. If a space check for the spool or log partitions fails, the incident is now
8634 logged. Of course, in the latter case the data may get lost...
8636 80. Added the %p formatting code to string_format() so that it can be used to
8637 print addresses in debug_print(). Adjusted all the address printing in the
8638 debugging in store.c to use %p rather than %d.
8640 81. There was a concern that a line of code in smtp_in.c could overflow a
8641 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
8642 initially expressed, the concern was not well-founded, because trailing
8643 spaces are removed early. However, if the trailing spaces were followed by
8644 a NULL, they did not get removed, so the overflow was possible. Two fixes
8647 (a) I re-wrote the offending code in a cleaner fashion.
8648 (b) If an incoming SMTP command contains a NULL character, it is rejected
8651 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
8652 runs initgroups(), so that if the Exim user is in any additional groups,
8653 they will be used during message reception.
8659 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / exim.git / blob