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 The hosts_connection_nolog main option now also controls "no MAIL in
9 SMTP connection" log lines.
11 JH/02 Option default value updates:
12 - queue_fast_ramp (main) true (was false)
13 - remote_max_parallel (main) 4 (was 2)
15 JH/03 Cache static regex pattern compilations, for use by ACLs.
17 JH/04 Bug 2903: avoid exit on an attempt to rewrite a malformed address.
18 Make the rewrite never match and keep the logging. Trust the
19 admin to be using verify=header-syntax (to actually reject the message).
21 JH/05 Follow symlinks for placing a watch on TLS creds files. This means
22 (under Linux) we watch the dir containing the final file; previously
23 it would be the dir with the first symlink. We still do not monitor
26 JH/06 Check for bad chars in rDNS for sender_host_name. The OpenBSD (at least)
27 dn_expand() is happy to pass them through.
29 JH/07 OpenSSL Fix auto-reload of changed server OCSP proof. Previously, if
30 the file with the proof had an unchanged name, the new proof(s) were
31 loaded on top of the old ones (and nover used; the old ones were stapled).
33 JH/08 Bug 2915: Fix use-after-free for $regex<n> variables. Previously when
34 more than one message arrived in a single connection a reference from
35 the earlier message could be re-used. Often a sigsegv resulted.
36 These variables were introduced in Exim 4.87.
37 Debug help from Graeme Fowler.
39 JH/09 Fix ${filter } for conditions that modify $value. Previously the
40 modified version would be used in construction the result, and a memory
43 JH/10 GnuTLS: fix for (IOT?) clients offering no TLS extensions at all.
44 Find and fix by Jasen Betts.
46 JH/11 OpenSSL: fix for ancient clients needing TLS support for versions earlier
47 than TLSv1,2, Previously, more-recent versions of OpenSSL were permitting
48 the systemwide configuration to override the Exim config.
50 HS/01 Bug 2728: Introduce EDITME option "DMARC_API" to work around incompatible
51 API changes in libopendmarc.
53 JH/12 Bug 2930: Fix daemon startup. When started from any process apart from
54 pid 1, in the normal "background daemon" mode, having to drop process-
55 group leadership also lost track of needing to create listener sockets.
57 JH/13 Bug 2929: Fix using $recipients after ${run...}. A change made for 4.96
58 resulted in the variable appearing empty. Find and fix by Ruben Jenster.
60 JH/14 Bug 2933: Fix regex substring match variables for null matches. Since 4.96
61 a capture group which obtained no text (eg. "(abc)*" matching zero
62 occurrences) could cause a segfault if the corresponding $<n> was
65 JH/15 Fix argument parsing for ${run } expansion. Previously, when an argument
66 included a close-brace character (eg. it itself used an expansion) an
69 JH/16 Move running the smtp connect ACL to before, for TLS-on-connect ports,
70 starting TLS. Previously it was after, meaning that attackers on such
71 ports had to be screened using the host_reject_connection main config
72 option. The new sequence aligns better with the STARTTLS behaviour, and
73 permits defences against crypto-processing load attacks, even though it
74 is strictly an incompatible change.
75 Also, avoid sending any SMTP fail response for either the connect ACL
76 or host_reject_connection, for TLS-on-connect ports.
78 JH/17 Permit the ACL "encrypted" condition to be used in a HELO/EHLO ACL,
79 Previously this was not permitted, but it makes reasonable sense.
80 While there, restore a restriction on using it from a connect ACL; given
81 the change JH/16 it could only return false (and before 4.91 was not
84 JH/18 Fix a fencepost error in logging. Previously (since 4.92) when a log line
85 was exactly sized compared to the log buffer, a crash occurred with the
86 misleading message "bad memory reference; pool not found".
87 Found and traced by Jasen Betts.
89 JH/19 Bug 2911: Fix a recursion in DNS lookups. Previously, if the main option
90 dns_again_means_nonexist included an element causing a DNS lookup which
91 iteslf returned DNS_AGAIN, unbounded recursion occurred. Possible results
92 included (though probably not limited to) a process crash from stack
93 memory limit, or from excessive open files. Replace this with a paniclog
94 whine (as this is likely a configuration error), and returning
97 JH/20 Bug 2954: (OpenSSL) Fix setting of explicit EC curve/group. Previously
98 this always failed, probably leading to the usual downgrade to in-clear
101 JH/21 Fix TLSA lookups. Previously dns_again_means_nonexist would affect
102 SERVFAIL results, which breaks the downgrade resistance of DANE. Change
103 to not checking that list for these lookups.
105 JH/22 Bug 2434: Add connection-elapsed "D=" element to more connection
108 JH/23 Fix crash in string expansions. Previously, if an empty variable was
109 immediately followed by an expansion operator, a null-indirection read
110 was done, killing the process.
112 JH/24 Bug 2997: When built with EXPERIMENTAL_DSN_INFO, bounce messages can
113 include an SMTP response string which is longer than that supported
114 by the delivering transport. Alleviate by wrapping such lines before
117 JH/25 Bug 2827: Restrict size of References: header in bounce messages to 998
118 chars (RFC limit). Previously a limit of 12 items was made, which with
119 a not-impossible References: in the message being bounced could still
120 be over-large and get stopped in the transport.
122 JH/26 For a ${readsocket } in TLS mode, send a TLS Close Alert before the TCP
123 close. Previously a bare socket close was done.
125 JH/27 Fix ${srs_encode ..}. Previously it would give a bad result for one day
128 JH/28 Bug 2996: Fix a crash in the smtp transport. When finding that the
129 message being considered for delivery was already being handled by
130 another process, and having an SMTP connection already open, the function
131 to close it tried to use an uninitialized variable. This would afftect
132 high-volume sites more, especially when running mailing-list-style loads.
133 Pollution of logs was the major effect, as the other process delivered
134 the message. Found and partly investigated by Graeme Fowler.
136 JH/29 Change format of the internal ID used for message identification. The old
137 version only supported 31 bits for a PID element; the new 64 (on systems
138 which can use Base-62 encoding, which is all currently supported ones
139 but not Darwin (MacOS) or Cygwin, which have case-insensitive filesystems
140 and must use Base-36). The new ID is 23 characters rather than 16, and is
141 visible in various places - notably logs, message headers, and spool file
142 names. Various of the ancillary utilities also have to know the format.
143 As well as the expanded PID portion, the sub-second part of the time
144 recorded in the ID is expanded to support finer precision. Theoretically
145 this permits a receive rate from a single comms channel of better than the
147 The major timestamp part of the ID is not changed; at 6 characters it is
148 usable until about year 3700.
149 Updating from previously releases is fully supported: old-format spool
150 files are still usable, and the utilities support both formats. New
151 message will use the new format. The one hints-DB file type which uses
152 message-IDs (the transport wait- DB) will be discarded if an old-format ID
153 is seen; new ones will be built with only new-format IDs.
154 Optionally, a utility can be used to convert spool files from old to new,
155 but this is only an efficiency measure not a requirement for operation
156 Downgrading from new to old requires running a provided utility, having
157 first stopped all operations. This will convert any spool files from new
158 back to old (losing time-precision and PID information) and remove any
159 wait- hints databases.
161 JH/30 Bug 3006: Fix handling of JSON strings having embedded commas. Previously
162 we treated them as item separators when parsing for a list item, but they
163 need to be protected by the doublequotes. While there, add handling for
166 JH/31 Bug 2998: Fix ${utf8clean:...} to disallow UTF-16 surrogate codepoints.
167 Found and fixed by Jasen Betts. No testcase for this as my usual text
168 editor insists on emitting only valid UTF-8.
170 JH/32 Fix "tls_dhparam = none" under GnuTLS. At least with 3.7.9 this gave
171 a null-indirection SIGSEGV for the receive process.
173 JH/33 Fix free for live variable $value created by a ${run ...} expansion.
174 Although not seen, this could have resulted in a SIGSEGV.
176 JH/34 Bug 3013: Fix use of $recipients within arguments for ${run...}.
177 In 4.96 this would expand to empty.
183 JH/01 Move the wait-for-next-tick (needed for unique message IDs) from
184 after reception to before a subsequent reception. This should
185 mean slightly faster delivery, and also confirmation of reception
188 JH/02 Move from using the pcre library to pcre2. The former is no longer
189 being developed or supported (by the original developer).
191 JH/03 Constification work in the filters module required a major version
192 bump for the local-scan API. Specifically, the "headers_charset"
193 global which is visible via the API is now const and may therefore
194 not be modified by local-scan code.
196 JH/04 Fix ClamAV TCP use under FreeBSD. Previously the OS-specific shim for
197 sendfile() didi not account for the way the ClamAV driver code called it.
199 JH/05 Bug 2819: speed up command-line messages being read in. Previously a
200 time check was being done for every character; replace that with one
203 JH/06 Bug 2815: Fix ALPN sent by server under OpenSSL. Previously the string
204 sent was prefixed with a length byte.
206 JH/07 Change the SMTP feature name for pipelining connect to be compliant with
207 RFC 5321. Previously Dovecot (at least) would log errors during
210 JH/08 Remove stripping of the binaries from the FreeBSD build. This was added
211 in 4.61 without a reason logged. Binaries will be bigger, which might
212 matter on diskspace-constrained systems, but debug is easier.
214 JH/09 Fix macro-definition during "-be" expansion testing. The move to
215 write-protected store for macros had not accounted for these runtime
216 additions; fix by removing this protection for "-be" mode.
218 JH/10 Convert all uses of select() to poll(). FreeBSD 12.2 was found to be
219 handing out large-numbered file descriptors, violating the usual Unix
220 assumption (and required by Posix) that the lowest possible number will be
221 allocated by the kernel when a new one is needed. In the daemon, and any
222 child procesees, values higher than 1024 (being bigger than FD_SETSIZE)
223 are not useable for FD_SET() [and hence select()] and overwrite the stack.
224 Assorted crashes happen.
226 JH/11 Fix use of $sender_host_name in daemon process. When used in certain
227 main-section options or in a connect ACL, the value from the first ever
228 connection was never replaced for subsequent connections. Found by
231 JH/12 Bug 2838: Fix for i32lp64 hard-align platforms. Found for SPARC Linux,
232 though only once PCRE2 was introduced: the memory accounting used under
233 debug offset allocations by an int, giving a hard trap in early startup.
234 Change to using a size_t. Debug and fix by John Paul Adrian Glaubitz.
236 JH/13 Bug 2845: Fix handling of tls_require_ciphers for OpenSSL when a value
237 with underbars is given. The write-protection of configuration introduced
238 in 4.95 trapped when normalisation was applied to an option not needing
241 JH/14 Bug 1895: TLS: Deprecate RFC 5114 Diffie-Hellman parameters.
243 JH/15 Fix a resource leak in *BSD. An off-by-one error resulted in the daemon
244 failing to close the certificates directory, every hour or any time it
247 JH/16 Debugging initiated by an ACL control now continues through into routing
248 and transport processes. Previously debugging stopped any time Exim
249 re-execs, or for processing a queued message.
251 JH/17 The "expand" debug selector now gives more detail, specifically on the
252 result of expansion operators and items.
254 JH/18 Bug 2751: Fix include_directory in redirect routers. Previously a
255 bad comparison between the option value and the name of the file to
256 be included was done, and a mismatch was wrongly identified.
257 4.88 to 4.95 are affected.
259 JH/19 Support for Berkeley DB versions 1 and 2 is withdrawn.
261 JH/20 When built with NDBM for hints DB's check for nonexistence of a name
262 supplied as the db file-pair basename. Previously, if a directory
263 path was given, for example via the autoreply "once" option, the DB
264 file.pag and file.dir files would be created in that directory's
267 JH/21 Remove the "allow_insecure_tainted_data" main config option and the
268 "taint" log_selector. These were previously deprecated.
270 JH/22 Fix static address-list lookups to properly return the matched item.
271 Previously only the domain part was returned.
273 JH/23 Bug 2864: FreeBSD: fix transport hang after 4xx/5xx response. Previously
274 the call into OpenSSL to send a TLS Close was being repeated; this
275 resulted in the library waiting for the peer's Close. If that was never
276 sent we waited forever. Fix by tracking send calls.
278 JH/24 The ${run} expansion item now expands its command string elements after
279 splitting. Previously it was before; the new ordering makes handling
280 zero-length arguments simpler. The old ordering can be obtained by
281 appending a new option "preexpand", after a comma, to the "run".
283 JH/25 Taint-check exec arguments for transport-initiated external processes.
284 Previously, tainted values could be used. This affects "pipe", "lmtp" and
285 "queryprogram" transport, transport-filter, and ETRN commands.
286 The ${run} expansion is also affected: in "preexpand" mode no part of
287 the command line may be tainted, in default mode the executable name
290 JH/26 Fix CHUNKING on a continued-transport. Previously the usabliility of
291 the the facility was not passed across execs, and only the first message
292 passed over a connection could use BDAT; any further ones using DATA.
294 JH/27 Support the PIPECONNECT facility in the smtp transport when the helo_data
295 uses $sending_ip_address and an interface is specified.
296 Previously any use of the local address in the EHLO name disabled
297 PIPECONNECT, the common case being to use the rDNS of it.
299 JH/28 OpenSSL: fix transport-required OCSP stapling verification under session
300 resumption. Previously verify failed because no certificate status is
301 passed on the wire for the restarted session. Fix by using the recorded
302 ocsp status of the stored session for the new connection.
304 JH/29 TLS resumption: the key for session lookup in the client now includes
305 more info that a server could potentially use in configuring a TLS
306 session, avoiding oferring mismatching sessions to such a server.
307 Previously only the server IP was used.
309 JH/30 Fix string_copyn() for limit greater than actual string length.
310 Previously the copied amount was the limit, which could result in a
311 overlapping memcpy for newly allocated destination soon after a
312 source string shorter than the limit. Found/investigated by KM.
314 JH/31 Bug 2886: GnuTLS: Do not free the cached creds on transport connection
315 close; it may be needed for a subsequent connection. This caused a
316 SEGV on primary-MX defer. Found/investigated by Gedalya & Andreas.
318 JH/32 Fix CHUNKING for a second message on a connection when the first was
319 rejected. Previously we did not reset the chunking-offered state, and
320 erroneously rejected the BDAT command. Investigation help from
323 JH/33 Fis ${srs_encode ...} to handle an empty sender address, now returning
324 an empty address. Previously the expansion returned an error.
326 HS/01 Bug 2855: Handle a v4mapped sender address given us by a frontending
327 proxy. Previously these were misparsed, leading to paniclog entries.
333 JH/01 Bug 1329: Fix format of Maildir-format filenames to match other mail-
334 related applications. Previously an "H" was used where available info
335 says that "M" should be, so change to match.
337 JH/02 Bug 2587: Fix pam expansion condition. Tainted values are commonly used
338 as arguments, so an implementation trying to copy these into a local
339 buffer was taking a taint-enforcement trap. Fix by using dynamically
340 created buffers. Similar fix for radius expansion condition.
342 JH/03 Bug 2586: Fix listcount expansion operator. Using tainted arguments is
343 reasonable, eg. to count headers. Fix by using dynamically created
344 buffers rather than a local. Do similar fixes for ACL actions "dcc",
345 "log_reject_target", "malware" and "spam"; the arguments are expanded
346 so could be handling tainted values.
348 JH/04 Bug 2590: Fix -bi (newaliases). A previous code rearrangement had
349 broken the (no-op) support for this sendmail command. Restore it
350 to doing nothing, silently, and returning good status.
352 JH/05 Bug 2593: Fix "vacation" in Exim filter. Previously, when a "once"
353 record path was given (or the default used) without a leading directory
354 path, an error occurred on trying to open it. Use the transport's working
357 JH/06 Bug 2594: Change the name used for certificate name checks in the smtp
358 transport. Previously it was the name on the DNS A-record; use instead
359 the head of the CNAME chain leading there (if there is one). This seems
360 to align better with RFC 6125.
362 JH/07 Bug 2597: Fix a resource leak. Using a lookup in obtaining a value for
363 smtp_accept_max_per_host allocated resources which were not released
364 when the limit was exceeded. This eventually crashed the daemon. Fix
365 by adding a release action in that path.
367 JH/08 Bug 2598: Fix verify ACL condition. The options for the condition are
368 expanded; previously using tainted values was rejected. Fix by using
369 dynamically-created buffers.
371 JH/09 Relax restrictions on ACL verify condition needing access to message
372 headers. Previously they were only permitted in data and non-smtp ACLs;
373 permit also mime, dkim, prdr quit and notquit. Applies to header-syntax,
374 not_blind, header_sender and header_names_ascii verification.
376 JH/10 Bug 2603: Fix coding of string copying to only evaluate arguments once.
377 Previously a macro used one argument twice; when called with the
378 argument as an expression having side-effects, incorrect operation
379 resulted. Use an inlineable function.
381 JH/11 Bug 2604: Fix request to cutthrough-deliver when a connection is already
382 held open for a verify callout. Previously this wan not accounted for
383 and a corrupt onward SMTP conversation resulted.
385 JH/12 Bug 2607: Fix the ${srs_encode } expansion to handle quoted local_parts.
386 Previously they were embedded naively in the constructed address; when
387 needed, strip the quoting and quote the entire local_part.
388 Also make the inbound_srs expansion condition handle quoting.
390 JH/13 Fix dsearch "subdir" filter to ignore ".". Previously only ".." was
391 excluded, not matching the documentation.
393 JH/14 Bug 2606: Fix a segfault in sqlite lookups. When no, or a bad, filename
394 was given for the sqlite_dbfile a trap resulted.
396 JH/15 Bug 2620: Fix "spam" ACL condition. Previously, tainted values for the
397 "name" argument resulted in a trap. There is no reason to disallow such;
398 this was a coding error.
400 JH/16 Bug 2615: Fix pause during message reception, on systems that have been
401 suspended/resumed. The Linux CLOCK_MONOTONIC does not account for time
402 spent suspended, ignoring the POSIX definition. Previously we assumed
403 it did and a constant offset from real time could be used as a correction.
404 Change to using the same clock source for the start-of-message and the
405 post-message next-tick-wait. Also change to using CLOCK_BOOTTIME if it
406 exists, just to get a clock slightly more aligned to reality.
408 JH/17 Bug 2295: Fix DKIM signing to always semicolon-terminate. Although the
409 RFC says it is optional some validators care. The missing char was not
410 intended but triggered by a line-wrap alignment. Discovery and fix by
411 Guillaume Outters, hacked on by JH.
413 JH/18 Bug 2617: Fix a taint trap in parse_fix_phrase(). Previously when the
414 name being quoted was tainted a trap would be taken. Fix by using
415 dynamically created buffers. The routine could have been called by a
416 rewrite with the "h" flag, by using the "-F" command-line option, or
417 by using a "name=" option on a control=submission ACL modifier.
419 JH/19 SPF: change the Authentication-Results expansion component to give
420 smtp.helo when the sender domain is empty. Previously it gave
423 JH/20 Bug 2631: ACL dnslist conditions now ignore and log any lookups returns
424 not in 127.0.0.0/8 to help in spotting list domains taken over by a
425 domain-parking registrar.
427 JH/21 Bug 2630: Fix eol-replacement string for the ${readsocket } expansion.
428 Previously when a whitespace character was specified it was not inserted
429 after removing the newline.
431 JH/22 Bug 2265: Force SNI usage for smtp transport DANE'd connections, to be
432 the domain part of the recipient address. This overrides any tls_sni
433 option set, which was previously used.
435 JH/23 Logging: with the +tls_sni log_selector, do not wrap the received SNI
438 JH/24 Bug 2634: Fix a taint trap seen on NetBSD: the testing coded for
439 is_tainted() had an off-by-one error in the overenthusiastic direction.
440 Find and fix by Gavan. Although NetBSD is not a supported platform for
441 4.94 this bug could affect other platforms.
443 PP/01 Fix default prime selection to be consistent.
444 One path used ike23 still, instead of exim.dev.20160529.3; now both
445 execution flows will use the same DH primes (currently
446 exim.dev.20160529.3).
448 JH/25 OpenSSL: Fix back-compatibility behaviour surrounding tls_certificates
449 option in smtp transport, to match the documentation. Previously
450 verification was not being done in some cases where it should have been.
452 JH/26 Bug 2646: fix a memory usage issue in ldap lookups. Previously, when more
453 than one server was defined and depending on the platform memory layout
454 details, an internal consistency trap could be hit while walking the list
457 JH/27 Bug 2648: fix the passing of an authenticator public-name through spool
458 files. The value is used by the authresults expansion item. Previously
459 if this was used in a router or transport, a crash could result.
461 JH/28 Fix spurious logging of select error. Some platforms, notably FreeBSD,
462 have a sufficient incidence of EINTR returns from select that an
463 interaction with other operations done by the main daemon loop exposed
464 a bug in the error-handling. This was benign apart from the log
467 JH/29 Bug 2675: add outgoing-interface I= element to deferred "==" log lines,
468 for consistency with delivered "=>" and failed "**" lines. While we're
469 there, handle PRX and TFO.
471 JH/30 Bug 2677: fix matching of long addresses. Since 4.93 a limit of 256 was
472 applied. This resulted, if any header-line rewrite rules were configured,
473 in a panic-log triggerable by sending a message with a long address in
474 a header. Fix by increasing the arbitrary limit to larger than a single
475 (dewrapped) 5322 header line maximum size.
477 JH/31 The ESMTP option name advertised for the SUPPORT_EARLY_PIPE build option
478 is changed from X_PIPE_CONNECT to PIPE_CONNECT. This is in line with
479 RFC 6648 which deprecates X- options in protocols as a general practice.
480 Changeover between the implementations is handled by the mechanisms
483 JH/32 Bug 2599: fix delay of delivery to a local address where there is also
484 a remote which uses callout/hold. Previously the local was queued.
486 JH/33 Fix a taint trap in the ${listextract } expansion when the source data
489 JH/34 Fix the placement of a multiple-message delivery marker in the delivery
490 log line. The asterisk is now consistently appended to the remote IP
491 (and port, if given), and will also be provided on defer and fail log
492 lines. Previously it could be placed on the local IP if that was being
493 logged, and was only provided on delivery lines.
495 JH/35 Bug 2343: Harden exim_tidydb against corrupt wait- files.
497 JH/36 Bug 2687: Fix interpretation of multiple ^ chars in a plaintext
498 authenticator client_send option. Previously the next char, after a pair
499 was collapsed, was taken verbatim (so ^^^foo became ^^foo; ^^^^foo became
500 ^^\x00foo). Fixed to get ^\x00foo and ^^foo respectively to match the
501 documentation. There is still no way to get a leading ^ immediately
502 after a NUL (ie. for the password of a PLAIN method authenticator.
504 JH/37 Enforce the expected size, for fixed-size records read from hints-DB
505 files. For bad sizes read, delete the record and whine to paniclog.
507 JH/38 When logging an AUTH failure, as server, do not include sensitive
508 information. Previously, the credentials would be included if given
509 as part of the AUTH command line and an ACL denied authentication.
511 JH/39 Bug 2691: fix $local_part_data. When the matching list element
512 referred to a file, bad data was returned. This likely also affected
515 JH/40 The gsasl authenticator now supports caching of the salted password
516 generated by the client-side implementation. This required the addition
517 of a new variable: $auth4.
519 JH/41 Fix daemon SIGHUP on FreeBSD. Previously, a named socket for IPC was
520 left undeleted; the attempt to re-create it then failed - resulting in
521 the usual "SIGHUP tp have daemon reload configuration" to not work.
522 This affected any platform not supporting "abstract" Unix-domain
523 sockets (i.e. not Linux).
525 JH/42 Bug 2693: Harden against a peer which reneges on a 452 "too many
526 recipients" response to RCPT in a later response, with a 250. The
527 previous coding assumed this would not happen, and under PIPELINING
528 would result in both lost and duplicate recipients for a message.
530 JH/43 Bug 2694: Fix weighted distribution of work to multiple spamd servers.
531 Previously the weighting was incorrectly applied. Similar fix for socks
532 proxies. Found and fixed by Heiko Schlichting.
534 JH/44 Bug 2701: Fix list-expansion of dns_ipv4_lookup. Previously, it did
535 not handle sub-lists included using the +namedlist syntax. While
536 investigating, the same found for dns_trust_aa, dns_again_means_nonexist,
537 dnssec_require_domains, dnssec_request_domains, srv_fail_domains,
540 JH/45 Use a (new) separate store pool-pair for DKIM verify working data.
541 Previously the permanent pool was used, so the sore could not be freed.
542 This meant a connection with many messages would use continually-growing
545 JH/46 Use an exponentially-increasing block size when malloc'ing store. Do it
546 per-pool so as not to waste too much space. Previously a constant size
547 was used which resulted in O(n^2) behaviour; now we get O(n log n) making
548 DOS attacks harder. The cost is wasted memory use in the larger blocks.
550 JH/47 Use explicit alloc/free for DNS lookup workspace. This permits using the
551 same space repeatedly, and a smaller process footprint.
553 JH/48 Use a less bogus-looking filename for a temporary used for DH-parameters
554 for GnuTLS. Previously the name started "%s" which, while not a bug,
555 looked as if if might be one.
557 JH/49 Bug 2710: when using SOCKS for additional messages after the first (a
558 "continued connection") make the $proxy_* variables available. Previously
559 the information was not passed across the exec() call for subsequent
560 transport executions. This also mean that the log lines for the
561 messages can show the proxy information.
563 JH/50 Bug 2672: QT elements in log lines, unless disabled, now exclude the
564 receive time. With modern systems the difference is significant.
565 The historical behaviour can be restored by disabling (a new) log_selector
566 "queue_time_exclusive".
568 JH/51 Taint-check ACL line. Previously, only filenames (for out-of-line ACL
569 content) were specifically tested for. Now, also cover expansions
570 resulting in ACL names and inline ACL content.
572 JH/52 Fix ${ip6norm:} operator. Previously, any trailing line text was dropped,
573 making it unusable in complex expressions.
575 JH/53 Bug 2743: fix immediate-delivery via named queue. Previously this would
576 fail with a taint-check on the spoolfile name, and leave the message
579 HS/01 Enforce absolute PID file path name.
581 HS/02 Handle SIGINT as we handle SIGTERM: terminate the Exim process.
583 PP/01 Add a too-many-bad-recipients guard to the default config's RCPT ACL.
585 PP/02 Bug 2643: Correct TLS DH constants.
586 A missing NUL termination in our code-generation tool had led to some
587 incorrect Diffie-Hellman constants in the Exim source.
588 Reported by kylon94, code-gen tool fix by Simon Arlott.
590 PP/03 Impose security length checks on various command-line options.
591 Fixes CVE-2020-SPRSS reported by Qualys.
593 PP/04 Fix Linux security issue CVE-2020-SLCWD and guard against PATH_MAX
594 better. Reported by Qualys.
596 PP/05 Fix security issue CVE-2020-PFPSN and guard against cmdline invoker
597 providing a particularly obnoxious sender full name.
600 PP/06 Fix CVE-2020-28016 (PFPZA): Heap out-of-bounds write in parse_fix_phrase()
602 PP/07 Refuse to allocate too little memory, block negative/zero allocations.
605 PP/08 Change default for recipients_max from unlimited to 50,000.
607 PP/09 Fix security issue with too many recipients on a message (to remove a
608 known security problem if someone does set recipients_max to unlimited,
609 or if local additions add to the recipient list).
610 Fixes CVE-2020-RCPTL reported by Qualys.
612 PP/10 Fix security issue in SMTP verb option parsing
613 Fixes CVE-2020-EXOPT reported by Qualys.
615 PP/11 Fix security issue in BDAT state confusion.
616 Ensure we reset known-good where we know we need to not be reading BDAT
617 data, as a general case fix, and move the places where we switch to BDAT
618 mode until after various protocol state checks.
619 Fixes CVE-2020-BDATA reported by Qualys.
621 HS/03 Die on "/../" in msglog file names
623 QS/01 Creation of (database) files in $spool_dir: only uid=0 or the uid of
624 the Exim runtime user are allowed to create files.
626 QS/02 PID file creation/deletion: only possible if uid=0 or uid is the Exim
629 QS/03 When reading the output from interpreted forward files we do not
630 pass the pipe between the parent and the interpreting process to
631 executed child processes (if any).
633 QS/04 Always die if requested from internal logging, even is logging is
636 JH/54 DMARC: recent versions of the OpenDMARC library appear to have broken
637 the API; compilation noo longer completes with DMARC support included.
638 This affects 1.4.1-1 on Fedora 33 (1.3.2-3 is functional); and has
639 been reported on other platforms.
641 JH/55 TLS: as server, reject connections with ALPN indicating non-smtp use.
643 JH/56 Make the majority of info read from config files readonly, for defence-in-
644 depth against exploits. Suggestion by Qualys.
645 Not supported on Solaris 10.
647 JH/57 Fix control=fakreject for a custom message containing tainted data.
648 Previously this resulted in a log complaint, due to a re-expansion present
649 since fakereject was originally introduced.
651 JH/58 GnuTLS: Fix certextract expansion. If a second modifier after a tag
652 modifier was given, a loop resulted.
654 JH/59 DKIM: Fix small-message verification under TLS with chunking. If a
655 pipelined SMTP command followed the BDAT LAST then it would be
656 incorrectly treated as part of the message body, causing a verification
659 JH/60 Bug 2805: Fix logging of domain-literals in Message_ID: headers. They
660 require looser validation rules than those for 821-level addresses,
661 which only permit IP addresses.
667 JH/01 Avoid costly startup code when not strictly needed. This reduces time
668 for some exim process initialisations. It does mean that the logging
669 of TLS configuration problems is only done for the daemon startup.
671 JH/02 Early-pipelining support code is now included unless disabled in Makefile.
673 JH/03 DKIM verification defaults no long accept sha1 hashes, to conform to
674 RFC 8301. They can still be enabled, using the dkim_verify_hashes main
677 JH/04 Support CHUNKING from an smtp transport using a transport_filter, when
678 DKIM signing is being done. Previously a transport_filter would always
679 disable CHUNKING, falling back to traditional DATA.
681 JH/05 Regard command-line recipients as tainted.
683 JH/06 Bug 340: Remove the daemon pid file on exit, when due to SIGTERM.
685 JH/07 Bug 2489: Fix crash in the "pam" expansion condition. It seems that the
686 PAM library frees one of the arguments given to it, despite the
687 documentation. Therefore a plain malloc must be used.
689 JH/08 Bug 2491: Use tainted buffers for the transport smtp context. Previously
690 on-stack buffers were used, resulting in a taint trap when DSN information
691 copied from a received message was written into the buffer.
693 JH/09 Bug 2493: Harden ARC verify against Outlook, whick has been seen to mix
694 the ordering of its ARC headers. This caused a crash.
696 JH/10 Bug 2492: Use tainted memory for retry record when needed. Previously when
697 a new record was being constructed with information from the peer, a trap
700 JH/11 Bug 2494: Unset the default for dmarc_tld_file. Previously a naiive
701 installation would get error messages from DMARC verify, when it hit the
702 nonexistent file indicated by the default. Distros wanting DMARC enabled
703 should both provide the file and set the option.
704 Also enforce no DMARC verification for command-line sourced messages.
706 JH/12 Fix an uninitialised flag in early-pipelining. Previously connections
707 could, depending on the platform, hang at the STARTTLS response.
709 JH/13 Bug 2498: Reset a counter used for ARC verify before handling another
710 message on a connection. Previously if one message had ARC headers and
711 the following one did not, a crash could result when adding an
712 Authentication-Results: header.
714 JH/14 Bug 2500: Rewind some of the common-coding in string handling between the
715 Exim main code and Exim-related utities. The introduction of taint
716 tracking also did many adjustments to string handling. Since then, eximon
717 frequently terminated with an assert failure.
719 JH/15 When PIPELINING, synch after every hundred or so RCPT commands sent and
720 check for 452 responses. This slightly helps the inefficieny of doing
721 a large alias-expansion into a recipient-limited target. The max_rcpt
722 transport option still applies (and at the current default, will override
723 the new feature). The check is done for either cause of synch, and forces
724 a fast-retry of all 452'd recipients using a new MAIL FROM on the same
725 connection. The new facility is not tunable at this time.
727 JH/16 Fix the variables set by the gsasl authenticator. Previously a pointer to
728 library live data was being used, so the results became garbage. Make
729 copies while it is still usable.
731 JH/17 Logging: when the deliver_time selector ise set, include the DT= field
732 on delivery deferred (==) and failed (**) lines (if a delivery was
733 attemtped). Previously it was only on completion (=>) lines.
735 JH/18 Authentication: the gsasl driver not provides the $authN variables in time
736 for the expansion of the server_scram_iter and server_scram_salt options.
738 WB/01 SPF: DNS lookups for the obsolete SPF RR type done by the libspf2 library
739 are now specifically given a NO_DATA response without hitting the system
740 resolver. The library goes on to do the now-standard TXT lookup.
741 Use of dnsdb lookups is not affected.
743 JH/19 Bug 2507: Modules: on handling a dynamic-module (lookups) open failure,
744 only retrieve the errormessage once. Previously two calls to dlerror()
745 were used, and the second one (for mainlog/paniclog) retrieved null
748 JH/20 Taint checking: disallow use of tainted data for
749 - the appendfile transport file and directory options
750 - the pipe transport command
751 - the autoreply transport file, log and once options
752 - file names used by the redirect router (including filter files)
754 - paths used by single-key lookups
755 Previously this was permitted.
757 JH/21 Bug 2501: Fix init call in the heimdal authenticator. Previously it
758 adjusted the size of a major service buffer; this failed because the
759 buffer was in use at the time. Change to a compile-time increase in the
760 buffer size, when this authenticator is compiled into exim.
762 JH/22 Taint-checking: move to safe-mode taint checking on all platforms. The
763 previous fast-mode was untenable in the face of glibs using mmap to
764 support larger malloc requests.
766 PP/01 Update the openssl_options possible values through OpenSSL 1.1.1c.
767 New values supported, if defined on system where compiled:
768 allow_no_dhe_kex, cryptopro_tlsext_bug, enable_middlebox_compat,
769 no_anti_replay, no_encrypt_then_mac, prioritize_chacha, tlsext_padding
771 JH/23 Performance improvement in the initial phase of a two-pass queue run. By
772 running a limited number of proceses in parallel, a benefit is gained. The
773 amount varies with the platform hardware and load. The use of the option
774 queue_run_in_order means we cannot do this, as ordering becomes
777 JH/24 Bug 2524: fix the cyrus_sasl auth driver gssapi usage. A previous fix
778 had introduced a string-copy (for ensuring NUL-termination) which was not
779 appropriate for that case, which can include embedded NUL bytes in the
780 block of data. Investigation showed the copy to actually be needless, the
781 data being length-specified.
783 JH/25 Fix use of concurrent TLS connections under GnuTLS. When a callout was
784 done during a receiving connection, and both used TLS, global info was
785 used rather than per-connection info for tracking the state of data
786 queued for transmission. This could result in a connection hang.
788 JH/26 Fix use of the SIZE parameter on MAIL commands, on continued connections.
789 Previously, when delivering serveral messages down a single connection
790 only the first would provide a SIZE. This was due to the size information
791 not being properly tracked.
793 JH/27 Bug 2530: When operating in a timezone with sub-minute offset, such as
794 TAI (at 37 seconds currently), pretend to be in UTC for time-related
795 expansion and logging. Previously, spurious values such as a future
796 minute could be seen.
798 JH/28 Bug 2533: Fix expansion of ${tr } item. When called in some situations
799 it could crash from a null-deref. This could also affect the
800 ${addresses: } operator and ${readsock } item.
802 JH/29 Bug 2537: Fix $mime_part_count. When a single connection had a non-mime
803 message following a mime one, the variable was not reset.
805 JH/30 When an pipelined-connect fails at the first response, assume incorrect
806 cached capability (perhaps the peer reneged?) and immediately retry in
809 JH/31 Fix spurious detection of timeout while writing to transport filter.
811 JH/32 Bug 2541: Fix segfault on bad cmdline -f (sender) argument. Previously
812 an attempt to copy the string was made before checking it.
814 JH/33 Fix the dsearch lookup to return an untainted result. Previously the
815 taint of the lookup key was maintained; we now regard the presence in the
816 filesystem as sufficient validation.
818 JH/34 Fix the readsocket expansion to not segfault when an empty "options"
819 argument is supplied.
821 JH/35 The dsearch lookup now requires that the directory is an absolute path.
822 Previously this was not checked, and nonempty relative paths made an
823 access under Exim's current working directory.
825 JH/36 Bug 2554: Fix msg:defer event for the hosts_max_try_hardlimit case.
826 Previously no event was raised.
828 JH/37 Bug 2552: Fix the check on spool space during reception to use the SIZE
829 parameter supplied by the sender MAIL FROM command. Previously it was
830 ignored, and only the check_spool_space option value for the required
833 JH/38 Fix $dkim_key_length. This should, after a DKIM verification, present
834 the size of the signing public-key. Previously it was instead giving
835 the size of the signature hash.
837 JH/39 DKIM verification: the RFC 8301 restriction on sizes of RSA keys is now
838 the default. See the (new) dkim_verify_min_keysizes option.
840 JH/40 Fix a memory-handling bug: when a connection carried multiple messages
841 and an ACL use a lookup for checking either the local_part or domain,
842 stale data could be accessed. Ensure that variable references are
843 dropped between messages.
845 JH/41 Bug 2571: Fix SPA authenticator. Running as a server, an offset supplied
846 by the client was not checked as pointing within response data before
847 being used. A malicious client could thus cause an out-of-bounds read and
848 possibly gain authentication. Fix by adding the check.
850 JH/42 Internationalisation: change the default for downconversion in the smtp
851 transport to be "if needed". Previously it was "as previously set" for
852 the message, which usually meant "if needed" for message-submission but
853 "no" for everything else. However, MTAs have been seen using SMTPUTF8
854 even when the envelope addresses did not need it, resulting in forwarding
855 failures to non-supporting MTAs. A downconvert in such cases will be
856 a no-op on the addresses, merely dropping the use of SMTPUTF8 by the
857 transport. The change does mean that addresses needing conversion will
858 be converted when previously a delivery failure would occur.
860 JH/43 Fix possible long line in DSN. Previously when a very long SMTP error
861 response was received it would be used unchecked in a fail-DSN, violating
862 standards on line-length limits. Truncate if needed.
864 HS/01 Remove parameters of the link to www.open-spf.org. The linked form
865 doesn't work. (Additionally add a new main config option to configure the
872 JH/01 OpenSSL: With debug enabled output keying information sufficient, server
873 side, to decode a TLS 1.3 packet capture.
875 JH/02 OpenSSL: Suppress the sending of (stateful) TLS1.3 session tickets.
876 Previously the default library behaviour applied, sending two, each in
879 JH/03 Debug output for ACL now gives the config file name and line number for
882 JH/04 The default received_header_text now uses the RFC 8314 tls cipher clause.
884 JH/05 DKIM: ensure that dkim_domain elements are lowercased before use.
886 JH/06 Fix buggy handling of autoreply bounce_return_size_limit, and a possible
887 buffer overrun for (non-chunking) other transports.
889 JH/07 GnuTLS: Our use of late (post-handshake) certificate verification, under
890 TLS1.3, means that a server rejecting a client certificate is not visible
891 to the client until the first read of encrypted data (typically the
892 response to EHLO). Add detection for that case and treat it as a failed
893 TLS connection attempt, so that the normal retry-in-clear can work (if
894 suitably configured).
896 JB/01 Bug 2375: fix expansions of 822 addresses having comments in local-part
897 and/or domain. Found and fixed by Jason Betts.
899 JH/08 Add hardening against SRV & TLSA lookups the hit CNAMEs (a nonvalid
900 configuration). If a CNAME target was not a wellformed name pattern, a
903 JH/09 Logging: Fix initial listening-on line for multiple ports for an IP when
904 the OS reports them interleaved with other addresses.
906 JH/10 OpenSSL: Fix aggregation of messages. Previously, when PIPELINING was
907 used both for input and for a verify callout, both encrypted, SMTP
908 responses being sent by the server could be lost. This resulted in
909 dropped connections and sometimes bounces generated by a peer sending
912 JH/11 Harden plaintext authenticator against a badly misconfigured client-send
913 string. Previously it was possible to cause undefined behaviour in a
914 library routine (usually a crash). Found by "zerons".
916 JH/12 Bug 2384: fix "-bP smtp_receive_timeout". Previously it returned no
919 JH/13 Bug 2386: Fix builds with Dane under LibreSSL 2.9.0 onward. Some old
920 API was removed, so update to use the newer ones.
922 JH/14 Bug 1891: Close the log file if receiving a non-smtp message, without
923 any timeout set, is taking a long time. Previously we would hang on to a
924 rotated logfile "forever" if the input was arriving with long gaps
925 (a previous attempt to fix addressed lack, for a long time, of initial
928 HS/01 Bug 2390: Use message_id for tempfile creation to avoid races in a
929 shared (NFS) environment. The length of the tempfile name is now
930 4 + 16 ("hdr.$message_exim_id") which might break on file
931 systems which restrict the file name length to lower values.
934 HS/02 Bug 2390: Use message_id for tempfile creation to avoid races in a
935 shared (NFS) environment.
937 HS/03 Bug 2392: exigrep does case sensitive *option* processing (as it
938 did for all versions <4.90). Notably -M, -m, --invert, -I may be
941 JH/15 Use unsigned when creating bitmasks in macros, to avoid build errors
942 on some platforms for bit 31.
944 JH/16 GnuTLS: rework ciphersuite strings under recent library versions. Thanks
945 to changes apparently associated with TLS1.3 handling some of the APIs
946 previously used were either nonfunctional or inappropriate. Strings
947 like TLS1.3:ECDHE_SECP256R1__RSA_PSS_RSAE_SHA256__AES_256_GCM__AEAD:256
948 and TLS1.2:ECDHE_SECP256R1__RSA_SHA256__AES_128_CBC__SHA256:128 replace
949 the previous TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256 .
950 This affects log line X= elements, the $tls_{in,out}_cipher variables,
951 and the use of specific cipher names in the encrypted= ACL condition.
953 JH/17 OpenSSL: the default openssl_options now disables ssl_v3.
955 JH/18 GnuTLS: fix $tls_out_ocsp under hosts_request_ocsp. Previously the
956 verification result was not updated unless hosts_require_ocsp applied.
958 JH/19 Bug 2398: fix listing of a named-queue. Previously, even with the option
959 queue_list_requires_admin set to false, non-admin users were denied the
962 JH/20 Bug 2389: fix server advertising of usable certificates, under GnuTLS in
963 directory-of-certs mode. Previously they were advertised despite the
966 JH/21 The smtp transport option "hosts_noproxy_tls" is now unset by default.
967 A single TCP connection by a client will now hold a TLS connection open
968 for multiple message deliveries, by default. Previously the default was to
971 JH/22 The smtp transport option "hosts_try_dane" now enables all hosts by
972 default. If built with the facility, DANE will be used. The facility
973 SUPPORT_DANE is now enabled in the prototype build Makefile "EDITME".
975 JH/23 The build default is now for TLS to be included; the SUPPORT_TLS define
976 is replaced with DISABLE_TLS. Either USE_GNUTLS or (the new) USE_OPENSSL
977 must be defined and you must still, unless you define DISABLE_TLS, manage
978 the the include-dir and library-file requirements that go with that
979 choice. Non-TLS builds are still supported.
981 JH/24 Fix duplicated logging of peer name/address, on a transport connection-
984 JH/25 The smtp transport option "hosts_try_fastopen" now enables all hosts by
985 default. If the platform supports and has the facility enabled, it will
986 be requested on all coneections.
988 JH/26 The PIPE_CONNECT facility is promoted from experimental status and is now
989 controlled by the build-time option SUPPORT_PIPE_CONNECT.
991 PP/01 Unbreak heimdal_gssapi, broken in 4.92.
993 JH/27 Bug 2404: Use the main-section configuration option "dsn_from" for
994 success-DSN messages. Previously the From: header was always the default
995 one for these; the option was ignored.
997 JH/28 Fix the timeout on smtp response to apply to the whole response.
998 Previously it was reset for every read, so a teergrubing peer sending
999 single bytes within the time limit could extend the connection for a
1000 long time. Credit to Qualsys Security Advisory Team for the discovery.
1002 JH/29 Fix DSN Final-Recipient: field. Previously it was the post-routing
1003 delivery address, which leaked information of the results of local
1004 forwarding. Change to the original envelope recipient address, per
1007 JH/30 Bug 2411: Fix DSN generation when RFC 3461 failure notification is
1008 requested. Previously not bounce was generated and a log entry of
1009 error ignored was made.
1011 JH/31 Avoid re-expansion in ${sort } expansion. (CVE-2019-13917)
1013 JH/32 Introduce a general tainting mechanism for values read from the input
1014 channel, and values derived from them. Refuse to expand any tainted
1015 values, to catch one form of exploit.
1017 JH/33 Bug 2413: Fix dkim_strict option. Previously the expansion result
1018 was unused and the unexpanded text used for the test. Found and
1019 fixed by Ruben Jenster.
1021 JH/34 Fix crash after TLS shutdown. When the TCP/SMTP channel was left open,
1022 an attempt to use a TLS library read routine dereffed a nul pointer,
1025 JH/35 Bug 2409: filter out-of-spec chars from callout response before using
1026 them in our smtp response.
1028 JH/36 Have the general router option retry_use_local_part default to true when
1029 any of the restrictive preconditions are set (to anything). Previously it
1030 was only for check_local user. The change removes one item of manual
1031 configuration which is required for proper retries when a remote router
1032 handles a subset of addresses for a domain.
1034 JH/37 Appendfile: when evaluating quota use (non-quota_size_regex) take the file
1035 link count into consideration.
1037 HS/04 Fix handling of very log lines in -H files. If a -<key> <value> line
1038 caused the extension of big_buffer, the following lines were ignored.
1040 JH/38 Bug 1395: Teach the DNS negative-cache about TTL value from the SOA in
1041 accordance with RFC 2308. Previously there was no expiry, so a longlived
1042 receive process (eg. due to ACL delays) versus a short SOA value could
1045 HS/05 Handle trailing backslash gracefully. (CVE-2019-15846)
1047 JH/39 Promote DMARC support to mainline.
1049 JH/40 Bug 2452: Add a References: header to DSNs.
1051 JH/41 With GnuTLS 3.6.0 (and later) do not attempt to manage Diffie-Hellman
1052 parameters. The relevant library call is documented as "Deprecated: This
1053 function is unnecessary and discouraged on GnuTLS 3.6.0 or later. Since
1054 3.6.0, DH parameters are negotiated following RFC7919."
1056 HS/06 Change the default of dnssec_request_domains to "*"
1058 JH/42 Bug 2545: Fix CHUNKING for all RCPT commands rejected. Previously we
1059 carried on and emitted a BDAT command, even when PIPELINING was not
1062 JH/43 Bug 2465: Fix taint-handling in dsearch lookup. Previously a nontainted
1063 buffer was used for the filename, resulting in a trap when tainted
1064 arguments (eg. $domain) were used.
1066 JH/44 With OpenSSL 1.1.1 (onwards) disable renegotiation for TLS1.2 and below;
1067 recommended to avoid a possible server-load attack. The feature can be
1068 re-enabled via the openssl_options main cofiguration option.
1070 JH/45 local_scan API: documented the current smtp_printf() call. This changed
1071 for version 4.90 - adding a "more data" boolean to the arguments.
1072 Bumped the ABI version number also, this having been missed previously;
1073 release versions 4.90 to 4.92.3 inclusive were effectively broken in
1074 respect of usage of smtp_printf() by either local_scan code or libraries
1075 accessed via the ${dlfunc } expansion item. Both will need coding
1076 adjustment for any calls to smtp_printf() to match the new function
1077 signature; a FALSE value for the new argument is always safe.
1079 JH/46 FreeBSD: fix use of the sendfile() syscall. The shim was not updating
1080 the file-offset (which the Linux syscall does, and exim expects); this
1081 resulted in an indefinite loop.
1083 JH/47 ARC: fix crash in signing, triggered when a configuration error failed
1084 to do ARC verification. The Authentication-Results: header line added
1085 by the configuration then had no ARC item.
1087 JH/48 Bug 2784: fix shutdown=no in the ${readsocket) expansion item. Previously
1088 an incorrect mode was used for reading the result, resulting in it being
1095 JH/01 Remove code calling the customisable local_scan function, unless a new
1096 definition "HAVE_LOCAL_SCAN=yes" is present in the Local/Makefile.
1098 JH/02 Bug 1007: Avoid doing logging from signal-handlers, as that can result in
1099 non-signal-safe functions being used.
1101 JH/03 Bug 2269: When presented with a received message having a stupidly large
1102 number of DKIM-Signature headers, disable DKIM verification to avoid
1103 a resource-consumption attack. The limit is set at twenty.
1105 JH/04 Add variables $arc_domains, $arc_oldest_pass for ARC verify. Fix the
1106 report of oldest_pass in ${authres } in consequence, and separate out
1107 some descriptions of reasons for verification fail.
1109 JH/05 Bug 2273: Cutthrough delivery left a window where the received messsage
1110 files in the spool were present and unlocked. A queue-runner could spot
1111 them, resulting in a duplicate delivery. Fix that by doing the unlock
1112 after the unlink. Investigation by Tim Stewart. Take the opportunity to
1113 add more error-checking on spoolfile handling while that code is being
1116 PP/01 Refuse to open a spool data file (*-D) if it's a symlink.
1117 No known attacks, no CVE, this is defensive hardening.
1119 JH/06 Bug 2275: The MIME ACL unlocked the received message files early, and
1120 a queue-runner could start a delivery while other operations were ongoing.
1121 Cutthrough delivery was a common victim, resulting in duplicate delivery.
1122 Found and investigated by Tim Stewart. Fix by using the open message data
1123 file handle rather than opening another, and not locally closing it (which
1124 releases a lock) for that case, while creating the temporary .eml format
1125 file for the MIME ACL. Also applies to "regex" and "spam" ACL conditions.
1127 JH/07 Bug 177: Make a random-recipient callout success visible in ACL, by setting
1128 $sender_verify_failure/$recipient_verify_failure to "random".
1130 JH/08 When generating a selfsigned cert, use serial number 1 since zero is not
1133 JH/09 Bug 2274: Fix logging of cmdline args when starting in an unlinked cwd.
1134 Previously this would segfault.
1136 JH/10 Fix ARC signing for case when DKIM signing failed. Previously this would
1139 JH/11 Bug 2264: Exim now only follows CNAME chains one step by default. We'd
1140 like zero, since the resolver should be doing this for us, But we need one
1141 as a CNAME but no MX presence gets the CNAME returned; we need to check
1142 that doesn't point to an MX to declare it "no MX returned" rather than
1143 "error, loop". A new main option is added so the older capability of
1144 following some limited number of chain links is maintained.
1146 JH/12 Add client-ip info to non-pass iprev ${authres } lines.
1148 JH/13 For receent Openssl versions (1.1 onward) use modern generic protocol
1149 methods. These should support TLS 1.3; they arrived with TLS 1.3 and the
1150 now-deprecated earlier definitions used only specified the range up to TLS
1151 1.2 (in the older-version library docs).
1153 JH/14 Bug 2284: Fix DKIM signing for body lines starting with a pair of dots.
1155 JH/15 Rework TLS client-side context management. Stop using a global, and
1156 explicitly pass a context around. This enables future use of TLS for
1157 connections to service-daemons (eg. malware scanning) while a client smtp
1158 connection is using TLS; with cutthrough connections this is quite likely.
1160 JH/16 Fix ARC verification to do AS checks in reverse order.
1162 JH/17 Support a "tls" option on the ${readsocket } expansion item.
1164 JH/18 Bug 2287: Fix the protocol name (eg utf8esmtp) for multiple messages
1165 using the SMTPUTF8 option on their MAIL FROM commands, in one connection.
1166 Previously the "utf8" would be re-prepended for every additional message.
1168 JH/19 Reject MAIL FROM commands with SMTPUTF8 when the facility was not advertised.
1169 Previously thery were accepted, resulting in issues when attempting to
1170 forward messages to a non-supporting MTA.
1172 PP/02 Let -n work with printing macros too, not just options.
1174 JH/20 Bug 2296: Fix cutthrough for >1 address redirection. Previously only
1175 one parent address was copied, and bogus data was used at delivery-logging
1176 time. Either a crash (after delivery) or bogus log data could result.
1177 Discovery and analysis by Tim Stewart.
1179 PP/03 Make ${utf8clean:} expansion operator detect incomplete final character.
1180 Previously if the string ended mid-character, we did not insert the
1181 promised '?' replacement.
1183 PP/04 Documentation: current string operators work on bytes, not codepoints.
1185 JH/21 Change as many as possible of the global flags into one-bit bitfields; these
1186 should pack well giving a smaller memory footprint so better caching and
1187 therefore performance. Group the declarations where this can't be done so
1188 that the byte-sized flag variables are not interspersed among pointer
1189 variables, giving a better chance of good packing by the compiler.
1191 JH/22 Bug 1896: Fix the envelope from for DMARC forensic reports to be possibly
1192 non-null, to avoid issues with sites running BATV. Previously reports were
1193 sent with an empty envelope sender so looked like bounces.
1195 JH/23 Bug 2318: Fix the noerror command within filters. It wasn't working.
1196 The ignore_error flag wasn't being returned from the filter subprocess so
1197 was not set for later routers. Investigation and fix by Matthias Kurz.
1199 JH/24 Bug 2310: Raise a msg:fail:internal event for each undelivered recipient,
1200 and a msg:complete for the whole, when a message is manually removed using
1201 -Mrm. Developement by Matthias Kurz, hacked on by JH.
1203 JH/25 Avoid fixed-size buffers for pathnames in DB access. This required using
1204 a "Gnu special" function, asprintf() in the DB utility binary builds; I
1205 hope that is portable enough.
1207 JH/26 Bug 2311: Fix DANE-TA verification under GnuTLS. Previously it was also
1208 requiring a known-CA anchor certificate; make it now rely entirely on the
1209 TLSA as an anchor. Checking the name on the leaf cert against the name
1210 on the A-record for the host is still done for TA (but not for EE mode).
1212 JH/27 Fix logging of proxy address. Previously, a pointless "PRX=[]:0" would be
1213 included in delivery lines for non-proxied connections, when compiled with
1214 SUPPORT_SOCKS and running with proxy logging enabled.
1216 JH/28 Bug 2314: Fire msg:fail:delivery event even when error is being ignored.
1217 Developement by Matthias Kurz, tweaked by JH. While in that bit of code,
1218 move the existing event to fire before the normal logging of message
1219 failure so that custom logging is bracketed by normal logging.
1221 JH/29 Bug 2322: A "fail" command in a non-system filter (file) now fires the
1222 msg:fail:internal event. Developement by Matthias Kurz.
1224 JH/30 Bug 2329: Increase buffer size used for dns lookup from 2k, which was
1225 far too small for todays use of crypto signatures stored there. Go all
1226 the way to the max DNS message size of 64kB, even though this might be
1227 overmuch for IOT constrained device use.
1229 JH/31 Fix a bad use of a copy function, which could be used to pointlessly
1230 copy a string over itself. The library routine is documented as not
1231 supporting overlapping copies, and on MacOS it actually raised a SIGABRT.
1233 JH/32 For main options check_spool_space and check_inode_space, where the
1234 platform supports 64b integers, support more than the previous 2^31 kB
1235 (i.e. more than 2 TB). Accept E, P and T multipliers in addition to
1236 the previous G, M, k.
1238 JH/33 Bug 2338: Fix the cyrus-sasl authenticator to fill in the
1239 $authenticated_fail_id variable on authentication failure. Previously
1242 JH/34 Increase RSA keysize of autogen selfsign cert from 1024 to 2048. RHEL 8.0
1243 OpenSSL didn't want to use such a weak key. Do for GnuTLS also, and for
1244 more-modern GnuTLS move from GNUTLS_SEC_PARAM_LOW to
1245 GNUTLS_SEC_PARAM_MEDIUM.
1247 JH/35 OpenSSL: fail the handshake when SNI processing hits a problem, server
1248 side. Previously we would continue as if no SNI had been received.
1250 JH/36 Harden the handling of string-lists. When a list consisted of a sole
1251 "<" character, which should be a list-separator specification, we walked
1252 off past the nul-terimation.
1254 JH/37 Bug 2341: Send "message delayed" warning MDNs (restricted to external
1255 causes) even when the retry time is not yet met. Previously they were
1256 not, meaning that when (say) an account was over-quota and temp-rejecting,
1257 and multiple senders' messages were queued, only one sender would get
1258 notified on each configured delay_warning cycle.
1260 JH/38 Bug 2351: Log failures to extract envelope addresses from message headers.
1262 JH/39 OpenSSL: clear the error stack after an SSL_accept(). With anon-auth
1263 cipher-suites, an error can be left on the stack even for a succeeding
1264 accept; this results in impossible error messages when a later operation
1267 AM/01 Bug 2359: GnuTLS: repeat lowlevel read and write operations while they
1268 return error codes indicating retry. Under TLS1.3 this becomes required.
1270 JH/40 Fix the feature-cache refresh for EXPERIMENTAL_PIPE_CONNECT. Previously
1271 it only wrote the new authenticators, resulting in a lack of tracking of
1272 peer changes of ESMTP extensions until the next cache flush.
1274 JH/41 Fix the loop reading a message header line to check for integer overflow,
1275 and more-often against header_maxsize. Previously a crafted message could
1276 induce a crash of the recive process; now the message is cleanly rejected.
1278 JH/42 Bug 2366: Fix the behaviour of the dkim_verify_signers option. It had
1279 been totally disabled for all of 4.91. Discovery and fix by "Mad Alex".
1285 GF/01 DEFER rather than ERROR on redis cluster MOVED response.
1286 When redis_servers is set to a list of > 1 element, and the Redis servers
1287 in that list are in cluster configuration, convert the REDIS_REPLY_ERROR
1288 case of MOVED into a DEFER case instead, thus moving the query onto the
1289 next server in the list. For a cluster of N elements, all N servers must
1290 be defined in redis_servers.
1292 GF/02 Catch and remove uninitialized value warning in exiqsumm
1293 Check for existence of @ARGV before looking at $ARGV[0]
1295 JH/01 Replace the store_release() internal interface with store_newblock(),
1296 which internalises the check required to safely use the old one, plus
1297 the allocate and data copy operations duplicated in both (!) of the
1298 extant use locations.
1300 JH/02 Disallow '/' characters in queue names specified for the "queue=" ACL
1301 modifier. This matches the restriction on the commandline.
1303 JH/03 Fix pgsql lookup for multiple result-tuples with a single column.
1304 Previously only the last row was returned.
1306 JH/04 Bug 2217: Tighten up the parsing of DKIM signature headers. Previously
1307 we assumed that tags in the header were well-formed, and parsed the
1308 element content after inspecting only the first char of the tag.
1309 Assumptions at that stage could crash the receive process on malformed
1312 JH/05 Bug 2215: Fix crash associated with dnsdb lookup done from DKIM ACL.
1313 While running the DKIM ACL we operate on the Permanent memory pool so that
1314 variables created with "set" persist to the DATA ACL. Also (at any time)
1315 DNS lookups that fail create cache records using the Permanent pool. But
1316 expansions release any allocations made on the current pool - so a dnsdb
1317 lookup expansion done in the DKIM ACL releases the memory used for the
1318 DNS negative-cache, and bad things result. Solution is to switch to the
1319 Main pool for expansions.
1320 While we're in that code, add checks on the DNS cache during store_reset,
1321 active in the testsuite.
1322 Problem spotted, and debugging aided, by Wolfgang Breyha.
1324 JH/06 Fix issue with continued-connections when the DNS shifts unreliably.
1325 When none of the hosts presented to a transport match an already-open
1326 connection, close it and proceed with the list. Previously we would
1327 queue the message. Spotted by Lena with Yahoo, probably involving
1330 JH/07 Bug 2214: Fix SMTP responses resulting from non-accept result of MIME ACL.
1331 Previously a spurious "250 OK id=" response was appended to the proper
1334 JH/08 The "support for" informational output now, which built with Content
1335 Scanning support, has a line for the malware scanner interfaces compiled
1336 in. Interface can be individually included or not at build time.
1338 JH/09 The "aveserver", "kavdaemon" and "mksd" interfaces are now not included
1339 by the template makefile "src/EDITME". The "STREAM" support for an older
1340 ClamAV interface method is removed.
1342 JH/10 Bug 2223: Fix mysql lookup returns for the no-data case (when the number of
1343 rows affected is given instead).
1345 JH/11 The runtime Berkeley DB library version is now additionally output by
1346 "exim -d -bV". Previously only the compile-time version was shown.
1348 JH/12 Bug 2230: Fix cutthrough routing for nonfirst messages in an initiating
1349 SMTP connection. Previously, when one had more recipients than the
1350 first, an abortive onward connection was made. Move to full support for
1351 multiple onward connections in sequence, handling cutthrough connection
1352 for all multi-message initiating connections.
1354 JH/13 Bug 2229: Fix cutthrough routing for nonstandard port numbers defined by
1355 routers. Previously, a multi-recipient message would fail to match the
1356 onward-connection opened for the first recipient, and cause its closure.
1358 JH/14 Bug 2174: A timeout on connect for a callout was also erroneously seen as
1359 a timeout on read on a GnuTLS initiating connection, resulting in the
1360 initiating connection being dropped. This mattered most when the callout
1361 was marked defer_ok. Fix to keep the two timeout-detection methods
1364 JH/15 Relax results from ACL control request to enable cutthrough, in
1365 unsupported situations, from error to silently (except under debug)
1366 ignoring. This covers use with PRDR, frozen messages, queue-only and
1369 HS/01 Fix Buffer overflow in base64d() (CVE-2018-6789)
1371 JH/16 Fix bug in DKIM verify: a buffer overflow could corrupt the malloc
1372 metadata, resulting in a crash in free().
1374 PP/01 Fix broken Heimdal GSSAPI authenticator integration.
1375 Broken in f2ed27cf5, missing an equals sign for specified-initialisers.
1376 Broken also in d185889f4, with init system revamp.
1378 JH/17 Bug 2113: Fix conversation closedown with the Avast malware scanner.
1379 Previously we abruptly closed the connection after reading a malware-
1380 found indication; now we go on to read the "scan ok" response line,
1383 JH/18 Bug 2239: Enforce non-usability of control=utf8_downconvert in the mail
1384 ACL. Previously, a crash would result.
1386 JH/19 Speed up macro lookups during configuration file read, by skipping non-
1387 macro text after a replacement (previously it was only once per line) and
1388 by skipping builtin macros when searching for an uppercase lead character.
1390 JH/20 DANE support moved from Experimental to mainline. The Makefile control
1391 for the build is renamed.
1393 JH/21 Fix memory leak during multi-message connections using STARTTLS. A buffer
1394 was allocated for every new TLS startup, meaning one per message. Fix
1395 by only allocating once (OpenSSL) or freeing on TLS-close (GnuTLS).
1397 JH/22 Bug 2236: When a DKIM verification result is overridden by ACL, DMARC
1398 reported the original. Fix to report (as far as possible) the ACL
1399 result replacing the original.
1401 JH/23 Fix memory leak during multi-message connections using STARTTLS under
1402 OpenSSL. Certificate information is loaded for every new TLS startup,
1403 and the resources needed to be freed.
1405 JH/24 Bug 2242: Fix exim_dbmbuild to permit directoryless filenames.
1407 JH/25 Fix utf8_downconvert propagation through a redirect router. Previously it
1410 JH/26 Bug 2253: For logging delivery lines under PRDR, append the overall
1411 DATA response info to the (existing) per-recipient response info for
1412 the "C=" log element. It can have useful tracking info from the
1413 destination system. Patch from Simon Arlott.
1415 JH/27 Bug 2251: Fix ldap lookups that return a single attribute having zero-
1416 length value. Previously this would segfault.
1418 HS/02 Support Avast multiline protoocol, this allows passing flags to
1419 newer versions of the scanner.
1421 JH/28 Ensure that variables possibly set during message acceptance are marked
1422 dead before release of memory in the daemon loop. This stops complaints
1423 about them when the debug_store option is enabled. Discovered specifically
1424 for sender_rate_period, but applies to a whole set of variables.
1425 Do the same for the queue-runner and queue-list loops, for variables set
1426 from spool message files. Do the same for the SMTP per-message loop, for
1427 certain variables indirectly set in ACL operations.
1429 JH/29 Bug 2250: Fix a longstanding bug in heavily-pipelined SMTP input (such
1430 as a multi-recipient message from a mailinglist manager). The coding had
1431 an arbitrary cutoff number of characters while checking for more input;
1432 enforced by writing a NUL into the buffer. This corrupted long / fast
1433 input. The problem was exposed more widely when more pipelineing of SMTP
1434 responses was introduced, and one Exim system was feeding another.
1435 The symptom is log complaints of SMTP syntax error (NUL chars) on the
1436 receiving system, and refused recipients seen by the sending system
1437 (propating to people being dropped from mailing lists).
1438 Discovered and pinpointed by David Carter.
1440 JH/30 The (EXPERIMENTAL_DMARC) variable $dmarc_ar_header is withdrawn, being
1441 replaced by the ${authresults } expansion.
1443 JH/31 Bug 2257: Fix pipe transport to not use a socket-only syscall.
1445 HS/03 Set a handler for SIGTERM and call exit(3) if running as PID 1. This
1446 allows proper process termination in container environments.
1448 JH/32 Bug 2258: Fix spool_wireformat in combination with LMTP transport.
1449 Previously the "final dot" had a newline after it; ensure it is CR,LF.
1451 JH/33 SPF: remove support for the "spf" ACL condition outcome values "err_temp"
1452 and "err_perm", deprecated since 4.83 when the RFC-defined words
1453 "temperror" and "permerror" were introduced.
1455 JH/34 Re-introduce enforcement of no cutthrough delivery on transports having
1456 transport-filters or DKIM-signing. The restriction was lost in the
1457 consolidation of verify-callout and delivery SMTP handling.
1458 Extend the restriction to also cover ARC-signing.
1460 JH/35 Cutthrough: for a final-dot response timeout (and nonunderstood responses)
1461 in defer=pass mode supply a 450 to the initiator. Previously the message
1464 PP/02 DANE: add dane_require_tls_ciphers SMTP Transport option; if unset,
1465 tls_require_ciphers is used as before.
1467 HS/03 Malware Avast: Better match the Avast multiline protocol. Add
1468 "pass_unscanned". Only tmpfails from the scanner are written to
1469 the paniclog, as they may require admin intervention (permission
1470 denied, license issues). Other scanner errors (like decompression
1471 bombs) do not cause a paniclog entry.
1473 JH/36 Fix reinitialisation of DKIM logging variable between messages.
1474 Previously it was possible to log spurious information in receive log
1477 JH/37 Bug 2255: Revert the disable of the OpenSSL session caching. This
1478 triggered odd behaviour from Outlook Express clients.
1480 PP/03 Add util/renew-opendmarc-tlds.sh script for safe renewal of public
1483 JH/38 DKIM: accept Ed25519 pubkeys in SubjectPublicKeyInfo-wrapped form,
1484 since the IETF WG has not yet settled on that versus the original
1485 "bare" representation.
1487 JH/39 Fix syslog logging for syslog_timestamp=no and log_selector +millisec.
1488 Previously the millisecond value corrupted the output.
1489 Fix also for syslog_pid=no and log_selector +pid, for which the pid
1490 corrupted the output.
1496 JH/01 Rework error string handling in TLS interface so that the caller in
1497 more cases is responsible for logging. This permits library-sourced
1498 string to be attached to addresses during delivery, and collapses
1499 pairs of long lines into single ones.
1501 PP/01 Allow PKG_CONFIG_PATH to be set in Local/Makefile and use it correctly
1502 during configuration. Wildcards are allowed and expanded.
1504 JH/02 Rework error string handling in DKIM to pass more info back to callers.
1505 This permits better logging.
1507 JH/03 Rework the transport continued-connection mechanism: when TLS is active,
1508 do not close it down and have the child transport start it up again on
1509 the passed-on TCP connection. Instead, proxy the child (and any
1510 subsequent ones) for TLS via a unix-domain socket channel. Logging is
1511 affected: the continued delivery log lines do not have any DNSSEC, TLS
1512 Certificate or OCSP information. TLS cipher information is still logged.
1514 JH/04 Shorten the log line for daemon startup by collapsing adjacent sets of
1515 identical IP addresses on different listening ports. Will also affect
1518 PP/02 Bug 2070: uClibc defines __GLIBC__ without providing glibc headers;
1519 add noisy ifdef guards to special-case this sillyness.
1520 Patch from Bernd Kuhls.
1522 JH/05 Tighten up the checking in isip4 (et al): dotted-quad components larger
1523 than 255 are no longer allowed.
1525 JH/06 Default openssl_options to include +no_ticket, to reduce load on peers.
1526 Disable the session-cache too, which might reduce our load. Since we
1527 currrectly use a new context for every connection, both as server and
1528 client, there is no benefit for these.
1529 GnuTLS appears to not support tickets server-side by default (we don't
1530 call gnutls_session_ticket_enable_server()) but client side is enabled
1531 by default on recent versions (3.1.3 +) unless the PFS priority string
1534 PP/03 Add $SOURCE_DATE_EPOCH support for reproducible builds, per spec at
1535 <https://reproducible-builds.org/specs/source-date-epoch/>.
1537 JH/07 Fix smtp transport use of limited max_rcpt under mua_wrapper. Previously
1538 the check for any unsuccessful recipients did not notice the limit, and
1539 erroneously found still-pending ones.
1541 JH/08 Pipeline CHUNKING command and data together, on kernels that support
1542 MSG_MORE. Only in-clear (not on TLS connections).
1544 JH/09 Avoid using a temporary file during transport using dkim. Unless a
1545 transport-filter is involved we can buffer the headers in memory for
1546 creating the signature, and read the spool data file once for the
1547 signature and again for transmission.
1549 JH/10 Enable use of sendfile in Linux builds as default. It was disabled in
1550 4.77 as the kernel support then wasn't solid, having issues in 64bit
1551 mode. Now, it's been long enough. Add support for FreeBSD also.
1553 JH/11 Bug 2104: Fix continued use of a transport connection with TLS. In the
1554 case where the routing stage had gathered several addresses to send to
1555 a host before calling the transport for the first, we previously failed
1556 to close down TLS in the old transport process before passing the TCP
1557 connection to the new process. The new one sent a STARTTLS command
1558 which naturally failed, giving a failed delivery and bloating the retry
1559 database. Investigation and fix prototype from Wolfgang Breyha.
1561 JH/12 Fix check on SMTP command input synchronisation. Previously there were
1562 false-negatives in the check that the sender had not preempted a response
1563 or prompt from Exim (running as a server), due to that code's lack of
1564 awareness of the SMTP input buffering.
1566 PP/04 Add commandline_checks_require_admin option.
1567 Exim drops privileges sanely, various checks such as -be aren't a
1568 security problem, as long as you trust local users with access to their
1569 own account. When invoked by services which pass untrusted data to
1570 Exim, this might be an issue. Set this option in main configuration
1571 AND make fixes to the calling application, such as using `--` to stop
1574 JH/13 Do pipelining under TLS. Previously, although safe, no advantage was
1575 taken. Now take care to pack both (client) MAIL,RCPT,DATA, and (server)
1576 responses to those, into a single TLS record each way (this usually means
1577 a single packet). As a side issue, smtp_enforce_sync now works on TLS
1580 PP/05 OpenSSL/1.1: use DH_bits() for more accurate DH param sizes. This
1581 affects you only if you're dancing at the edge of the param size limits.
1582 If you are, and this message makes sense to you, then: raise the
1583 configured limit or use OpenSSL 1.1. Nothing we can do for older
1586 JH/14 For the "sock" variant of the malware scanner interface, accept an empty
1587 cmdline element to get the documented default one. Previously it was
1590 JH/15 Fix a crash in the smtp transport caused when two hosts in succession
1591 are unsuable for non-message-specific reasons - eg. connection timeout,
1592 banner-time rejection.
1594 JH/16 Fix logging of delivery remote port, when specified by router, under
1597 PP/06 Repair manualroute's ability to take options in any order, even if one
1598 is the name of a transport.
1601 HS/01 Cleanup, prevent repeated use of -p/-oMr (CVE-2017-1000369)
1603 JH/17 Change the list-building routines interface to use the expanding-string
1604 triplet model, for better allocation and copying behaviour.
1606 JH/18 Prebuild the data-structure for "builtin" macros, for faster startup.
1607 Previously it was constructed the first time a possibly-matching string
1608 was met in the configuration file input during startup; now it is done
1611 JH/19 Bug 2141: Use the full-complex API for Berkeley DB rather than the legacy-
1612 compatible one, to avoid the (poorly documented) possibility of a config
1613 file in the working directory redirecting the DB files, possibly correpting
1614 some existing file. CVE-2017-10140 assigned for BDB.
1616 JH/20 Bug 2147: Do not defer for a verify-with-callout-and-random which is not
1617 cache-hot. Previously, although the result was properly cached, the
1618 initial verify call returned a defer.
1620 JH/21 Bug 2151: Avoid using SIZE on the MAIL for a callout verify, on any but
1621 the main verify for receipient in uncached-mode.
1623 JH/22 Retire historical build files to an "unsupported" subdir. These are
1624 defined as "ones for which we have no current evidence of testing".
1626 JH/23 DKIM: enforce the DNS pubkey record "h" permitted-hashes optional field,
1627 if present. Previously it was ignored.
1629 JH/24 Start using specified-initialisers in C structure init coding. This is
1630 a C99 feature (it's 2017, so now considered safe).
1632 JH/25 Use one-bit bitfields for flags in the "addr" data structure. Previously
1633 if was a fixed-sized field and bitmask ops via macros; it is now more
1636 PP/07 GitHub PR 56: Apply MariaDB build fix.
1637 Patch provided by Jaroslav Škarvada.
1639 PP/08 Bug 2161: Fix regression in sieve quoted-printable handling introduced
1640 during Coverity cleanups [4.87 JH/47]
1641 Diagnosis and fix provided by Michael Fischer v. Mollard.
1643 JH/26 Fix DKIM bug: when the pseudoheader generated for signing was exactly
1644 the right size to place the terminating semicolon on its own folded
1645 line, the header hash was calculated to an incorrect value thanks to
1646 the (relaxed) space the fold became.
1648 HS/02 Fix Bug 2130: large writes from the transport subprocess were chunked
1649 and confused the parent.
1651 JH/27 Fix SOCKS bug: an unitialized pointer was deref'd by the transport process
1652 which could crash as a result. This could lead to undeliverable messages.
1654 JH/28 Logging: "next input sent too soon" now shows where input was truncated
1657 JH/29 Fix queue_run_in_order to ignore the PID portion of the message ID. This
1658 matters on fast-turnover and PID-randomising systems, which were getting
1659 out-of-order delivery.
1661 JH/30 Fix a logging bug on aarch64: an unsafe routine was previously used for
1662 a possibly-overlapping copy. The symptom was that "Remote host closed
1663 connection in response to HELO" was logged instead of the actual 4xx
1666 JH/31 Fix CHUNKING code to properly flush the unwanted chunk after an error.
1667 Previously only that bufferd was discarded, resulting in SYMTP command
1670 JH/32 DKIM: when a message has multiple signatures matching an identity given
1671 in dkim_verify_signers, run the dkim acl once for each. Previously only
1672 one run was done. Bug 2189.
1674 JH/33 Downgrade an unfound-list name (usually a typo in the config file) from
1675 "panic the current process" to "deliberately defer". The panic log is
1676 still written with the problem list name; the mail and reject logs now
1677 get a temp-reject line for the message that was being handled, saying
1678 something like "domains check lookup or other defer". The SMTP 451
1679 message is still "Temporary local problem".
1681 JH/34 Bug 2199: Fix a use-after-free while reading smtp input for header lines.
1682 A crafted sequence of BDAT commands could result in in-use memory beeing
1683 freed. CVE-2017-16943.
1685 HS/03 Bug 2201: Fix checking for leading-dot on a line during headers reading
1686 from SMTP input. Previously it was always done; now only done for DATA
1687 and not BDAT commands. CVE-2017-16944.
1689 JH/35 Bug 2201: Flush received data in BDAT mode after detecting an error fatal
1690 to the message (such as an overlong header line). Previously this was
1691 not done and we did not exit BDAT mode. Followon from the previous item
1692 though a different problem.
1698 JH/01 Bug 1922: Support IDNA2008. This has slightly different conversion rules
1699 than -2003 did; needs libidn2 in addition to libidn.
1701 JH/02 The path option on a pipe transport is now expanded before use.
1703 PP/01 GitHub PR 50: Do not call ldap_start_tls_s on ldapi:// connections.
1704 Patch provided by "Björn", documentation fix added too.
1706 JH/03 Bug 2003: fix Proxy Protocol v2 handling: the address size field was
1707 missing a wire-to-host endian conversion.
1709 JH/04 Bug 2004: fix CHUNKING in non-PIPELINEING mode. Chunk data following
1710 close after a BDAT command line could be taken as a following command,
1711 giving a synch failure. Fix by only checking for synch immediately
1712 before acknowledging the chunk.
1714 PP/02 GitHub PR 52: many spelling fixes, which include fixing parsing of
1715 no_require_dnssec option and creation of _HAVE_TRANSPORT_APPEND_MAILDIR
1716 macro. Patches provided by Josh Soref.
1718 JH/05 Have the EHLO response advertise VRFY, if there is a vrfy ACL defined.
1719 Previously we did not; the RFC seems ambiguous and VRFY is not listed
1720 by IANA as a service extension. However, John Klensin suggests that we
1723 JH/06 Bug 2017: Fix DKIM verification in -bh test mode. The data feed into
1724 the dkim code may be unix-mode line endings rather than smtp wire-format
1725 CRLF, so prepend a CR to any bare LF.
1727 JH/07 Rationalise the coding for callout smtp conversations and transport ones.
1728 As a side-benfit, callouts can now use PIPELINING hence fewer round-trips.
1730 JH/08 Bug 2016: Fix DKIM verification vs. CHUNKING. Any BDAT commands after
1731 the first were themselves being wrongly included in the feed into dkim
1732 processing; with most chunk sizes in use this resulted in an incorrect
1733 body hash calculated value.
1735 JH/09 Bug 2014: permit inclusion of a DKIM-Signature header in a received
1736 DKIM signature block, for verification. Although advised against by
1737 standards it is specifically not ruled illegal.
1739 JH/10 Bug 2025: Fix reception of (quoted) local-parts with embedded spaces.
1741 JH/11 Bug 2029: Fix crash in DKIM verification when a message signature block is
1742 missing a body hash (the bh= tag).
1744 JH/12 Bug 2018: Re-order Proxy Protocol startup versus TLS-on-connect startup.
1745 It seems that HAProxy sends the Proxy Protocol information in clear and
1746 only then does a TLS startup, so do the same.
1748 JH/13 Bug 2027: Avoid attempting to use TCP Fast Open for non-transport client
1749 TCP connections (such as for Spamd) unless the daemon successfully set
1750 Fast Open mode on its listening sockets. This fixes breakage seen on
1751 too-old kernels or those not configured for Fast Open, at the cost of
1752 requiring both directions being enabled for TFO, and TFO never being used
1753 by non-daemon-related Exim processes.
1755 JH/14 Bug 2000: Reject messages recieved with CHUNKING but with malformed line
1756 endings, at least on the first header line. Try to canonify any that get
1757 past that check, despite the cost.
1759 JH/15 Angle-bracket nesting (an error inserted by broken sendmails) levels are
1760 now limited to an arbitrary five deep, while parsing addresses with the
1761 strip_excess_angle_brackets option enabled.
1763 PP/03 Bug 2018: For Proxy Protocol and TLS-on-connect, do not over-read and
1764 instead leave the unprompted TLS handshake in socket buffer for the
1765 TLS library to consume.
1767 PP/04 Bug 2018: Also handle Proxy Protocol v2 safely.
1769 PP/05 FreeBSD compat: handle that Ports no longer create /usr/bin/perl
1771 JH/16 Drop variables when they go out of scope. Memory management drops a whole
1772 region in one operation, for speed, and this leaves assigned pointers
1773 dangling. Add checks run only under the testsuite which checks all
1774 variables at a store-reset and panics on a dangling pointer; add code
1775 explicitly nulling out all the variables discovered. Fixes one known
1776 bug: a transport crash, where a dangling pointer for $sending_ip_address
1777 originally assigned in a verify callout, is re-used.
1779 PP/06 Drop '.' from @INC in various Perl scripts.
1781 PP/07 Switch FreeBSD iconv to always use the base-system libc functions.
1783 PP/08 Reduce a number of compilation warnings under clang; building with
1784 CC=clang CFLAGS+=-Wno-dangling-else -Wno-logical-op-parentheses
1785 should be warning-free.
1787 JH/17 Fix inbound CHUNKING when DKIM disabled at runtime.
1789 HS/01 Fix portability problems introduced by PP/08 for platforms where
1790 realloc(NULL) is not equivalent to malloc() [SunOS et al].
1792 HS/02 Bug 1974: Fix missing line terminator on the last received BDAT
1793 chunk. This allows us to accept broken chunked messages. We need a more
1794 general solution here.
1796 PP/09 Wrote util/chunking_fixqueue_finalnewlines.pl to help recover
1797 already-broken messages in the queue.
1799 JH/18 Bug 2061: Fix ${extract } corrupting an enclosing ${reduce } $value.
1801 JH/19 Fix reference counting bug in routing-generated-address tracking.
1807 JH/01 Use SIZE on MAIL FROM in a cutthrough connection, if the destination
1808 supports it and a size is available (ie. the sending peer gave us one).
1810 JH/02 The obsolete acl condition "demime" is removed (finally, after ten
1811 years of being deprecated). The replacements are the ACLs
1812 acl_smtp_mime and acl_not_smtp_mime.
1814 JH/03 Upgrade security requirements imposed for hosts_try_dane: previously
1815 a downgraded non-dane trust-anchor for the TLS connection (CA-style)
1816 or even an in-clear connection were permitted. Now, if the host lookup
1817 was dnssec and dane was requested then the host is only used if the
1818 TLSA lookup succeeds and is dnssec. Further hosts (eg. lower priority
1819 MXs) will be tried (for hosts_try_dane though not for hosts_require_dane)
1820 if one fails this test.
1821 This means that a poorly-configured remote DNS will make it incommunicado;
1822 but it protects against a DNS-interception attack on it.
1824 JH/04 Bug 1810: make continued-use of an open smtp transport connection
1825 non-noisy when a race steals the message being considered.
1827 JH/05 If main configuration option tls_certificate is unset, generate a
1828 self-signed certificate for inbound TLS connections.
1830 JH/06 Bug 165: hide more cases of password exposure - this time in expansions
1831 in rewrites and routers.
1833 JH/07 Retire gnutls_require_mac et.al. These were nonfunctional since 4.80
1834 and logged a warning sing 4.83; now they are a configuration file error.
1836 JH/08 Bug 1836: Fix crash in VRFY handling when handed an unqualified name
1837 (lacking @domain). Apply the same qualification processing as RCPT.
1839 JH/09 Bug 1804: Avoid writing msglog files when in -bh or -bhc mode.
1841 JH/10 Support ${sha256:} applied to a string (as well as the previous
1844 JH/11 Cutthrough: avoid using the callout hints db on a verify callout when
1845 a cutthrough deliver is pending, as we always want to make a connection.
1846 This also avoids re-routing the message when later placing the cutthrough
1847 connection after a verify cache hit.
1848 Do not update it with the verify result either.
1850 JH/12 Cutthrough: disable when verify option success_on_redirect is used, and
1851 when routing results in more than one destination address.
1853 JH/13 Cutthrough: expand transport dkim_domain option when testing for dkim
1854 signing (which inhibits the cutthrough capability). Previously only
1855 the presence of an option was tested; now an expansion evaluating as
1856 empty is permissible (obviously it should depend only on data available
1857 when the cutthrough connection is made).
1859 JH/14 Fix logging of errors under PIPELINING. Previously the log line giving
1860 the relevant preceding SMTP command did not note the pipelining mode.
1862 JH/15 Fix counting of empty lines in $body_linecount and $message_linecount.
1863 Previously they were not counted.
1865 JH/16 DANE: treat a TLSA lookup response having all non-TLSA RRs, the same
1866 as one having no matching records. Previously we deferred the message
1867 that needed the lookup.
1869 JH/17 Fakereject: previously logged as a normal message arrival "<="; now
1870 distinguished as "(=".
1872 JH/18 Bug 1867: make the fail_defer_domains option on a dnslookup router work
1873 for missing MX records. Previously it only worked for missing A records.
1875 JH/19 Bug 1850: support Radius libraries that return REJECT_RC.
1877 JH/20 Bug 1872: Ensure that acl_smtp_notquit is run when the connection drops
1878 after the data-go-ahead and data-ack. Patch from Jason Betts.
1880 JH/21 Bug 1846: Send DMARC forensic reports for reject and quarantine results,
1881 even for a "none" policy. Patch from Tony Meyer.
1883 JH/22 Fix continued use of a connection for further deliveries. If a port was
1884 specified by a router, it must also match for the delivery to be
1887 JH/23 Bug 1874: fix continued use of a connection for further deliveries.
1888 When one of the recipients of a message was unsuitable for the connection
1889 (has no matching addresses), we lost track of needing to mark it
1890 deferred. As a result mail would be lost.
1892 JH/24 Bug 1832: Log EHLO response on getting conn-close response for HELO.
1894 JH/25 Decoding ACL controls is now done using a binary search; the source code
1895 takes up less space and should be simpler to maintain. Merge the ACL
1896 condition decode tables also, with similar effect.
1898 JH/26 Fix problem with one_time used on a redirect router which returned the
1899 parent address unchanged. A retry would see the parent address marked as
1900 delivered, so not attempt the (identical) child. As a result mail would
1903 JH/27 Fix a possible security hole, wherein a process operating with the Exim
1904 UID can gain a root shell. Credit to http://www.halfdog.net/ for
1905 discovery and writeup. Ubuntu bug 1580454; no bug raised against Exim
1908 JH/28 Enable {spool,log} filesystem space and inode checks as default.
1909 Main config options check_{log,spool}_{inodes,space} are now
1910 100 inodes, 10MB unless set otherwise in the configuration.
1912 JH/29 Fix the connection_reject log selector to apply to the connect ACL.
1913 Previously it only applied to the main-section connection policy
1916 JH/30 Bug 1897: fix callouts connection fallback from TLS to cleartext.
1918 PP/01 Changed default Diffie-Hellman parameters to be Exim-specific, created
1919 by me. Added RFC7919 DH primes as an alternative.
1921 PP/02 Unbreak build via pkg-config with new hash support when crypto headers
1922 are not in the system include path.
1924 JH/31 Fix longstanding bug with aborted TLS server connection handling. Under
1925 GnuTLS, when a session startup failed (eg because the client disconnected)
1926 Exim did stdio operations after fclose. This was exposed by a recent
1927 change which nulled out the file handle after the fclose.
1929 JH/32 Bug 1909: Fix OCSP proof verification for cases where the proof is
1930 signed directly by the cert-signing cert, rather than an intermediate
1931 OCSP-signing cert. This is the model used by LetsEncrypt.
1933 JH/33 Bug 1914: Ensure socket is nonblocking before draining after SMTP QUIT.
1935 HS/01 Fix leak in verify callout under GnuTLS, about 3MB per recipient on
1936 an incoming connection.
1938 HS/02 Bug 1802: Do not half-close the connection after sending a request
1941 HS/03 Use "auto" as the default EC curve parameter. For OpenSSL < 1.0.2
1942 fallback to "prime256v1".
1944 JH/34 SECURITY: Use proper copy of DATA command in error message.
1945 Could leak key material. Remotely exploitable. CVE-2016-9963.
1951 JH/01 Bug 1664: Disable OCSP for GnuTLS library versions at/before 3.3.16
1952 and 3.4.4 - once the server is enabled to respond to an OCSP request
1953 it does even when not requested, resulting in a stapling non-aware
1954 client dropping the TLS connection.
1956 TF/01 Code cleanup: Overhaul the debug_selector and log_selector machinery to
1957 support variable-length bit vectors. No functional change.
1959 TF/02 Improve the consistency of logging incoming and outgoing interfaces.
1960 The I= interface field on outgoing lines is now after the H= remote
1961 host field, same as incoming lines. There is a separate
1962 outgoing_interface log selector which allows you to disable the
1965 JH/02 Bug 728: Close logfiles after a daemon-process "exceptional" log write.
1966 If not running log_selector +smtp_connection the mainlog would be held
1967 open indefinitely after a "too many connections" event, including to a
1968 deleted file after a log rotate. Leave the per net connection logging
1969 leaving it open for efficiency as that will be quickly detected by the
1970 check on the next write.
1972 HS/01 Bug 1671: Fix post transport crash.
1973 Processing the wait-<transport> messages could crash the delivery
1974 process if the message IDs didn't exist for some reason. When
1975 using 'split_spool_directory=yes' the construction of the spool
1976 file name failed already, exposing the same netto behaviour.
1978 JH/03 Bug 425: Capture substrings in $regex1, $regex2 etc from regex &
1979 mime_regex ACL conditions.
1981 JH/04 Bug 1686: When compiled with EXPERIMENTAL_DSN_INFO: Add extra information
1982 to DSN fail messages (bounces): remote IP, remote greeting, remote response
1983 to HELO, local diagnostic string.
1985 JH/05 Downgrade message for a TLS-certificate-based authentication fail from
1986 log line to debug. Even when configured with a tls authenticator many
1987 client connections are expected to not authenticate in this way, so
1988 an authenticate fail is not an error.
1990 HS/02 Add the Exim version string to the process info. This way exiwhat
1991 gives some more detail about the running daemon.
1993 JH/06 Bug 1395: time-limit caching of DNS lookups, to the TTL value. This may
1994 matter for fast-change records such as DNSBLs.
1996 JH/07 Bug 1678: Always record an interface option value, if set, as part of a
1997 retry record, even if constant. There may be multiple transports with
1998 different interface settings and the retry behaviour needs to be kept
2001 JH/08 Bug 1586: exiqgrep now refuses to run if there are unexpected arguments.
2003 JH/09 Bug 1700: ignore space & tab embedded in base64 during decode.
2005 JH/10 Bug 840: fix log_defer_output option of pipe transport
2007 JH/11 Bug 830: use same host for all RCPTS of a message, even under
2008 hosts_randomize. This matters a lot when combined with mua_wrapper.
2010 JH/12 Bug 1706: percent and underbar characters are no longer escaped by the
2011 ${quote_pgsql:<string>} operator.
2013 JH/13 Bug 1708: avoid misaligned access in cached lookup.
2015 JH/14 Change header file name for freeradius-client. Relevant if compiling
2016 with Radius support; from the Gentoo tree and checked under Fedora.
2018 JH/15 Bug 1712: Introduce $prdr_requested flag variable
2020 JH/16 Bug 1714: Permit an empty string as expansion result for transport
2021 option transport_filter, meaning no filtering.
2023 JH/17 Bug 1713: Fix non-PDKIM_DEBUG build. Patch from Jasen Betts.
2025 JH/18 Bug 1709: When built with TLS support, the tls_advertise_hosts option now
2026 defaults to "*" (all hosts). The variable is now available when not built
2027 with TLS, default unset, mainly to enable keeping the testsuite sane.
2028 If a server certificate is not supplied (via tls_certificate) an error is
2029 logged, and clients will find TLS connections fail on startup. Presumably
2030 they will retry in-clear.
2031 Packagers of Exim are strongly encouraged to create a server certificate
2032 at installation time.
2034 HS/03 Add -bP config_file as a synonym for -bP configure_file, for consistency
2035 with the $config_file variable.
2037 JH/19 Two additional event types: msg:rcpt:defer and msg:rcpt:host:defer. Both
2038 in transport context, after the attempt, and per-recipient. The latter type
2039 is per host attempted. The event data is the error message, and the errno
2040 information encodes the lookup type (A vs. MX) used for the (first) host,
2041 and the trailing two digits of the smtp 4xx response.
2043 GF/01 Bug 1715: Fix for race condition in exicyclog, where exim could attempt
2044 to write to mainlog (or rejectlog, paniclog) in the window between file
2045 creation and permissions/ownership being changed. Particularly affects
2046 installations where exicyclog is run as root, rather than exim user;
2047 result is that the running daemon panics and dies.
2049 JH/20 Bug 1701: For MySQL lookups, support MySQL config file option group names.
2051 JH/21 Bug 1720: Add support for priority groups and weighted-random proxy
2052 selection for the EXPERIMENTAL_SOCKS feature, via new per-proxy options
2053 "pri" and "weight". Note that the previous implicit priority given by the
2054 list order is no longer honoured.
2056 JH/22 Bugs 963, 1721: Fix some corner cases in message body canonicalization
2057 for DKIM processing.
2059 JH/23 Move SOCKS5 support from Experimental to mainline, enabled for a build
2060 by defining SUPPORT_SOCKS.
2062 JH/26 Move PROXY support from Experimental to mainline, enabled for a build
2063 by defining SUPPORT_PROXY. Note that the proxy_required_hosts option
2064 is renamed to hosts_proxy, and the proxy_{host,target}_{address,port}.
2065 variables are renamed to proxy_{local,external}_{address,port}.
2067 JH/27 Move Internationalisation support from Experimental to mainline, enabled
2068 for a build by defining SUPPORT_I18N
2070 JH/28 Bug 1745: Fix redis lookups to handle (quoted) spaces embedded in parts
2071 of the query string, and make ${quote_redis:} do that quoting.
2073 JH/29 Move Events support from Experimental to mainline, enabled by default
2074 and removable for a build by defining DISABLE_EVENT.
2076 JH/30 Updated DANE implementation code to current from Viktor Dukhovni.
2078 JH/31 Fix bug with hosts_connection_nolog and named-lists which were wrongly
2079 cached by the daemon.
2081 JH/32 Move Redis support from Experimental to mainline, enabled for a build
2082 by defining LOOKUP_REDIS. The libhiredis library is required.
2084 JH/33 Bug 1748: Permit ACL dnslists= condition in non-smtp ACLs if explicit
2085 keys are given for lookup.
2087 JH/34 Bug 1192: replace the embedded copy of PolarSSL RSA routines in the DKIM
2088 support, by using OpenSSL or GnuTLS library ones. This means DKIM is
2089 only supported when built with TLS support. The PolarSSL SHA routines
2090 are still used when the TLS library is too old for convenient support.
2092 JH/35 Require SINGLE_DH_USE by default in OpenSSL (main config option
2093 openssl_options), for security. OpenSSL forces this from version 1.1.0
2094 server-side so match that on older versions.
2096 JH/36 Bug 1778: longstanding bug in memory use by the ${run } expansion: A fresh
2097 allocation for $value could be released as the expansion processing
2098 concluded, but leaving the global pointer active for it.
2100 JH/37 Bug 1769: Permit a VRFY ACL to override the default 252 response,
2101 and to use the domains and local_parts ACL conditions.
2103 JH/38 Fix cutthrough bug with body lines having a single dot. The dot was
2104 incorrectly not doubled on cutthrough transmission, hence seen as a
2105 body-termination at the receiving system - resulting in truncated mails.
2106 Commonly the sender saw a TCP-level error, and retransmitted the message
2107 via the normal store-and-forward channel. This could result in duplicates
2108 received - but deduplicating mailstores were liable to retain only the
2109 initial truncated version.
2111 JH/39 Bug 1781: Fix use of DKIM private-keys having trailing '=' in the base-64.
2113 JH/40 Fix crash in queryprogram router when compiled with EXPERIMENTAL_SRS.
2115 JH/41 Bug 1792: Fix selection of headers to sign for DKIM: bottom-up. While
2116 we're in there, support oversigning also; bug 1309.
2118 JH/42 Bug 1796: Fix error logged on a malware scanner connection failure.
2120 HS/04 Add support for keep_environment and add_environment options.
2122 JH/43 Tidy coding issues detected by gcc --fsanitize=undefined. Some remain;
2123 either intentional arithmetic overflow during PRNG, or testing config-
2126 JH/44 Bug 1800: The combination of a -bhc commandline option and cutthrough
2127 delivery resulted in actual delivery. Cancel cutthrough before DATA
2130 JH/45 Fix cutthrough, when connection not opened by verify and target hard-
2131 rejects a recipient: pass the reject to the originator.
2133 JH/46 Multiple issues raised by Coverity. Some were obvious or plausible bugs.
2134 Many were false-positives and ignorable, but it's worth fixing the
2137 JH/47 Fix build on HP-UX and older Solaris, which need (un)setenv now also
2138 for the new environment-manipulation done at startup. Move the routines
2139 from being local to tls.c to being global via the os.c file.
2141 JH/48 Bug 1807: Fix ${extract } for the numeric/3-string case. While preparsing
2142 an extract embedded as result-arg for a map, the first arg for extract
2143 is unavailable so we cannot tell if this is a numbered or keyed
2144 extraction. Accept either.
2150 JH/01 Bug 1545: The smtp transport option "retry_include_ip_address" is now
2153 JH/02 The smtp transport option "multi_domain" is now expanded.
2155 JH/03 The smtp transport now requests PRDR by default, if the server offers
2158 JH/04 Certificate name checking on server certificates, when exim is a client,
2159 is now done by default. The transport option tls_verify_cert_hostnames
2160 can be used to disable this per-host. The build option
2161 EXPERIMENTAL_CERTNAMES is withdrawn.
2163 JH/05 The value of the tls_verify_certificates smtp transport and main options
2164 default to the word "system" to access the system default CA bundle.
2165 For GnuTLS, only version 3.0.20 or later.
2167 JH/06 Verification of the server certificate for a TLS connection is now tried
2168 (but not required) by default. The verification status is now logged by
2169 default, for both outbound TLS and client-certificate supplying inbound
2172 JH/07 Changed the default rfc1413 lookup settings to disable calls. Few
2175 JH/08 The EXPERIMENTAL_DSN compile option is no longer needed; all Delivery
2176 Status Notification (bounce) messages are now MIME format per RFC 3464.
2177 Support for RFC 3461 DSN options NOTIFY,ENVID,RET,ORCPT can be advertised
2178 under the control of the dsn_advertise_hosts option, and routers may
2179 have a dsn_lasthop option.
2181 JH/09 A timeout of 2 minutes is now applied to all malware scanner types by
2182 default, modifiable by a malware= option. The list separator for
2183 the options can now be changed in the usual way. Bug 68.
2185 JH/10 The smtp_receive_timeout main option is now expanded before use.
2187 JH/11 The incoming_interface log option now also enables logging of the
2188 local interface on delivery outgoing connections.
2190 JH/12 The cutthrough-routing facility now supports multi-recipient mails,
2191 if the interface and destination host and port all match.
2193 JH/13 Bug 344: The verify = reverse_host_lookup ACL condition now accepts a
2196 JH/14 Bug 1573: The spam= ACL condition now additionally supports Rspamd.
2197 Patch from Andrew Lewis.
2199 JH/15 Bug 670: The spamd_address main option (for the spam= ACL condition)
2200 now supports optional time-restrictions, weighting, and priority
2201 modifiers per server. Patch originally by <rommer@active.by>.
2203 JH/16 The spamd_address main option now supports a mixed list of local
2204 and remote servers. Remote servers can be IPv6 addresses, and
2205 specify a port-range.
2207 JH/17 Bug 68: The spamd_address main option now supports an optional
2208 timeout value per server.
2210 JH/18 Bug 1581: Router and transport options headers_add/remove can
2211 now have the list separator specified.
2213 JH/19 Bug 392: spamd_address, and clamd av_scanner, now support retry
2216 JH/20 Bug 1571: Ensure that $tls_in_peerdn is set, when verification fails
2219 JH/21 Support for the A6 type of dns record is withdrawn.
2221 JH/22 Bug 608: The result of a QUIT or not-QUIT toplevel ACL now matters
2222 rather than the verbs used.
2224 JH/23 Bug 1572: Increase limit on SMTP confirmation message copy size
2225 from 255 to 1024 chars.
2227 JH/24 Verification callouts now attempt to use TLS by default.
2229 HS/01 DNSSEC options (dnssec_require_domains, dnssec_request_domains)
2230 are generic router options now. The defaults didn't change.
2232 JH/25 Bug 466: Add RFC2322 support for MIME attachment filenames.
2233 Original patch from Alexander Shikoff, worked over by JH.
2235 HS/02 Bug 1575: exigrep falls back to autodetection of compressed
2236 files if ZCAT_COMMAND is not executable.
2238 JH/26 Bug 1539: Add timeout/retry options on dnsdb lookups.
2240 JH/27 Bug 286: Support SOA lookup in dnsdb lookups.
2242 JH/28 Bug 1588: Do not use the A lookup following an AAAA for setting the FQDN.
2243 Normally benign, it bites when the pair was led to by a CNAME;
2244 modern usage is to not canonicalize the domain to a CNAME target
2245 (and we were inconsistent anyway for A-only vs AAAA+A).
2247 JH/29 Bug 1632: Removed the word "rejected" from line logged for ACL discards.
2249 JH/30 Check the forward DNS lookup for DNSSEC, in addition to the reverse,
2250 when evaluating $sender_host_dnssec.
2252 JH/31 Check the HELO verification lookup for DNSSEC, adding new
2253 $sender_helo_dnssec variable.
2255 JH/32 Bug 1397: Enable ECDHE on OpenSSL, just the NIST P-256 curve.
2257 JH/33 Bug 1346: Note MAIL cmd seen in -bS batch, to avoid smtp_no_mail log.
2259 JH/34 Bug 1648: Fix a memory leak seen with "mailq" and large queues.
2261 JH/35 Bug 1642: Fix support of $spam_ variables at delivery time. Was
2262 documented as working, but never had. Support all but $spam_report.
2264 JH/36 Bug 1659: Guard checking of input smtp commands again pseudo-command
2265 added for tls authenticator.
2267 HS/03 Add perl_taintmode main config option
2273 TL/01 When running the test suite, the README says that variables such as
2274 no_msglog_check are global and can be placed anywhere in a specific
2275 test's script, however it was observed that placement needed to be near
2276 the beginning for it to behave that way. Changed the runtest perl
2277 script to read through the entire script once to detect and set these
2278 variables, reset to the beginning of the script, and then run through
2279 the script parsing/test process like normal.
2281 TL/02 The BSD's have an arc4random API. One of the functions to induce
2282 adding randomness was arc4random_stir(), but it has been removed in
2283 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
2284 function when detected.
2286 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
2287 cause callback expansion.
2289 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
2290 syntax errors in an expansion can be treated as a string instead of
2291 logging or causing an error, due to the internal use of bool_lax
2292 instead of bool when processing it.
2294 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
2295 server certificates when making smtp deliveries.
2297 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
2299 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
2301 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
2303 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
2304 Merged patch from Sebastian Wiedenroth.
2306 JH/05 Fix results-pipe from transport process. Several recipients, combined
2307 with certificate use, exposed issues where response data items split
2308 over buffer boundaries were not parsed properly. This eventually
2309 resulted in duplicates being sent. This issue only became common enough
2310 to notice due to the introduction of connection certificate information,
2311 the item size being so much larger. Found and fixed by Wolfgang Breyha.
2313 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
2314 size buffer was used, resulting in syntax errors when an expansion
2317 JH/07 Add support for directories of certificates when compiled with a GnuTLS
2318 version 3.3.6 or later.
2320 JH/08 Rename the TPDA experimental facility to Event Actions. The #ifdef
2321 is EXPERIMENTAL_EVENT, the main-configuration and transport options
2322 both become "event_action", the variables become $event_name, $event_data
2323 and $event_defer_errno. There is a new variable $verify_mode, usable in
2324 routers, transports and related events. The tls:cert event is now also
2325 raised for inbound connections, if the main configuration event_action
2328 TL/06 In test suite, disable OCSP for old versions of openssl which contained
2329 early OCSP support, but no stapling (appears to be less than 1.0.0).
2331 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
2332 server certificate names available under the smtp transport option
2333 "tls_verify_cert_hostname" now do not permit multi-component wildcard
2336 JH/10 Time-related extraction expansions from certificates now use the main
2337 option "timezone" setting for output formatting, and are consistent
2338 between OpenSSL and GnuTLS compilations. Bug 1541.
2340 JH/11 Fix a crash in mime ACL when meeting a zero-length, quoted or RFC2047-
2341 encoded parameter in the incoming message. Bug 1558.
2343 JH/12 Bug 1527: Autogrow buffer used in reading spool files. Since they now
2344 include certificate info, eximon was claiming there were spoolfile
2347 JH/13 Bug 1521: Fix ldap lookup for single-attr request, multiple-attr return.
2349 JH/14 Log delivery-related information more consistently, using the sequence
2350 "H=<name> [<ip>]" wherever possible.
2352 TL/07 Bug 1547: Omit RFCs from release. Draft and RFCs have licenses which
2353 are problematic for Debian distribution, omit them from the release
2356 JH/15 Updates and fixes to the EXPERIMENTAL_DSN feature.
2358 JH/16 Fix string representation of time values on 64bit time_t architectures.
2361 JH/17 Fix a null-indirection in certextract expansions when a nondefault
2362 output list separator was used.
2367 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
2368 checkers that were complaining about end of non-void function with no
2371 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
2372 This was a regression introduced in 4.83 by another bugfix.
2374 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
2376 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
2377 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
2383 TF/01 Correctly close the server side of TLS when forking for delivery.
2385 When a message was received over SMTP with TLS, Exim failed to clear up
2386 the incoming connection properly after forking off the child process to
2387 deliver the message. In some situations the subsequent outgoing
2388 delivery connection happened to have the same fd number as the incoming
2389 connection previously had. Exim would try to use TLS and fail, logging
2390 a "Bad file descriptor" error.
2392 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
2393 utilities have not been installed.
2395 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
2396 temporary space as the ACL may create new global variables.
2398 TL/01 LDAP support uses per connection or global context settings, depending
2399 upon the detected version of the libraries at build time.
2401 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
2402 to extract and use the src ip:port in logging and expansions as if it
2403 were a direct connection from the outside internet. PPv2 support was
2404 updated based on HAProxy spec change in May 2014.
2406 JH/02 Add ${listextract {number}{list}{success}{fail}}.
2408 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
2409 Properly escape header and check for NULL return.
2411 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
2414 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
2416 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
2417 characters in header names, implemented as a verify condition.
2418 Contributed by Michael Fischer v. Mollard.
2420 TL/05 Rename SPF condition results err_perm and err_temp to standardized
2421 results permerror and temperror. Previous values are deprecated but
2422 still accepted. In a future release, err_perm and err_temp will be
2423 completely removed, which will be a backward incompatibility if the
2424 ACL tests for either of these two old results. Patch contributed by
2425 user bes-internal on the mailing list.
2427 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
2429 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
2430 selectors, in both main and reject logs.
2432 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
2435 JH/07 Add malware type "sock" for talking to simple daemon.
2437 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
2439 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
2440 routers/transports under cutthrough routing.
2442 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
2443 numbers. Touch up "bool" conditional to keep the same definition.
2445 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
2447 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
2449 JH/12 Expand items in router/transport headers_add or headers_remove lists
2450 individually rather than the list as a whole. Bug 1452.
2452 Required for reasonable handling of multiple headers_ options when
2453 they may be empty; requires that headers_remove items with embedded
2454 colons must have them doubled (or the list-separator changed).
2456 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
2457 view the policy declared in the DMARC record. Currently, $dmarc_status
2458 is a combined value of both the record presence and the result of the
2461 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
2463 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
2464 dnslookup router and the smtp transport (applying to the forward
2467 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
2468 of ldap servers used for a specific lookup. Patch provided by Heiko
2471 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
2472 New variable $lookup_dnssec_authenticated for observability.
2474 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
2475 Patch submitted by Lars Timman.
2477 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
2479 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
2480 Requires trusted mode and valid format message id, aborts otherwise.
2481 Patch contributed by Heiko Schlichting.
2483 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
2484 certextract with support for various fields. Bug 1358.
2486 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
2487 is requested by default, modifiable by smtp transport option
2490 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
2491 operate on certificate variables to give certificate fingerprints
2492 Also new ${sha256:cert_variable}.
2494 JH/23 The PRDR feature is moved from being Experimental into the mainline.
2496 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
2497 Christian Aistleitner.
2499 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
2501 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
2502 file. Patch from Wolfgang Breyha.
2504 JH/25 Expand the coverage of the delivery $host and $host_address to
2505 client authenticators run in verify callout. Bug 1476.
2507 JH/26 Port service names are now accepted for tls_on_connect_ports, to
2508 align with daemon_smtp_ports. Bug 72.
2510 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
2511 support and error reporting did not work properly.
2513 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
2514 and is readable. Patch from Andrew Colin Kissa.
2516 TL/14 Enhance documentation of ${run expansion and how it parses the
2517 commandline after expansion, particularly in the case when an
2518 unquoted variable expansion results in an empty value.
2520 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
2522 PP/02 Fix internal collision of T_APL on systems which support RFC3123
2523 by renaming away from it. Addresses GH issue 15, reported by
2526 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
2528 TL/15 SECURITY: prevent double expansion in math comparison functions
2529 (can expand unsanitized data). Not remotely exploitable.
2536 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
2538 PP/02 Make -n do something, by making it not do something.
2539 When combined with -bP, the name of an option is not output.
2541 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
2544 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
2545 $sender_host_name and config options to manage this, and basic check
2548 PP/05 DSCP support for outbound connections and control modifier for inbound.
2550 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
2551 (Only plugin which currently uses this is kerberos4, which nobody should
2552 be using, but we should make it available and other future plugins might
2553 conceivably use it, even though it would break NAT; stuff *should* be
2554 using channel bindings instead).
2556 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
2557 name; added for Sendmail compatibility; requires admin caller.
2558 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
2559 just ignore it); requires trusted caller.
2560 Also parse but ignore: -Ac -Am -X<logfile>
2563 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
2565 TL/02 Add +smtp_confirmation as a default logging option.
2567 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
2568 Patch by Magnus Holmgren from 2007-02-20.
2570 TL/04 Bugzilla 1281 - Spec typo.
2571 Bugzilla 1283 - Spec typo.
2572 Bugzilla 1290 - Spec grammar fixes.
2574 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
2576 TL/06 Add Experimental DMARC support using libopendmarc libraries.
2578 TL/07 Fix an out of order global option causing a segfault. Reported to dev
2579 mailing list by by Dmitry Isaikin.
2581 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
2583 JH/02 Support "G" suffix to numbers in ${if comparisons.
2585 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
2587 NM/01 Bugzilla 1197 - Spec typo
2588 Bugzilla 1196 - Spec examples corrections
2590 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
2592 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
2593 gnutls_enable_pkcs11, but renamed to more accurately indicate its
2596 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
2597 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
2599 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
2600 "acl {{name}{arg}...}", and optional args on acl condition
2603 JH/05 Permit multiple router/transport headers_add/remove lines.
2605 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
2607 JH/07 Avoid using a waiting database for a single-message-only transport.
2608 Performance patch from Paul Fisher. Bugzilla 1262.
2610 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
2613 JH/09 Add $headers_added variable, with content from use of ACL modifier
2614 add_header (but not yet added to the message). Bugzilla 199.
2616 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
2617 Pulled from Bugzilla 817 by Wolfgang Breyha.
2619 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
2621 (nb: this is the same fix as in Exim 4.80.1)
2623 JH/11 Add A= logging on delivery lines, and a client_set_id option on
2626 JH/12 Add optional authenticated_sender logging to A= and a log_selector
2629 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
2631 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
2632 advertise SMTP AUTH mechanism to us, instead of a generic
2633 protocol violation error. Also, make Exim more robust to bad
2634 data from the Dovecot auth socket.
2636 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
2638 When a queue runner is handling a message, Exim first routes the
2639 recipient addresses, during which it prunes them based on the retry
2640 hints database. After that it attempts to deliver the message to
2641 any remaining recipients. It then updates the hints database using
2644 So if a recipient address works intermittently, it can get repeatedly
2645 deferred at routing time. The retry hints record remains fresh so the
2646 address never reaches the final cutoff time.
2648 This is a fairly common occurrence when a user is bumping up against
2649 their storage quota. Exim had some logic in its local delivery code
2650 to deal with this. However it did not apply to per-recipient defers
2651 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
2653 This change adds a proper retry rule check during routing so that the
2654 final cutoff time is checked against the message's age. We only do
2655 this check if there is an address retry record and there is not a
2656 domain retry record; this implies that previous attempts to handle
2657 the address had the retry_use_local_parts option turned on. We use
2658 this as an approximation for the destination being like a local
2659 delivery, as in LMTP.
2661 I suspect this new check makes the old local delivery cutoff check
2662 redundant, but I have not verified this so I left the code in place.
2664 TF/02 Correct gecos expansion when From: is a prefix of the username.
2666 Test 0254 submits a message to Exim with the header
2670 When I ran the test suite under the user fanf2, Exim expanded
2671 the header to contain my full name, whereas it should have added
2672 a Resent-Sender: header. It erroneously treats any prefix of the
2673 username as equal to the username.
2675 This change corrects that bug.
2677 GF/01 DCC debug and logging tidyup
2678 Error conditions log to paniclog rather than rejectlog.
2679 Debug lines prefixed by "DCC: " to remove any ambiguity.
2681 TF/03 Avoid unnecessary rebuilds of lookup-related code.
2683 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
2684 Bug spotted by Jeremy Harris; was flawed since initial commit.
2685 Would have resulted in OCSP responses post-SNI triggering an Exim
2686 NULL dereference and crash.
2688 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
2690 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
2691 Bug detection, analysis and fix by Samuel Thibault.
2692 Bugzilla 1331, Debian bug #698092.
2694 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
2696 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
2697 Server implementation by Todd Lyons, client by JH.
2698 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
2699 config variable "prdr_enable" controls whether the server
2700 advertises the facility. If the client requests PRDR a new
2701 acl_data_smtp_prdr ACL is called once for each recipient, after
2702 the body content is received and before the acl_smtp_data ACL.
2703 The client is controlled by both of: a hosts_try_prdr option
2704 on the smtp transport, and the server advertisement.
2705 Default client logging of deliveries and rejections involving
2706 PRDR are flagged with the string "PRDR".
2708 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
2709 fclose(). Diagnosis by Todd Lyons.
2711 PP/17 Update configure.default to handle IPv6 localhost better.
2712 Patch by Alain Williams (plus minor tweaks).
2715 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
2716 This is now consistent with GnuTLS, and is now documented: the
2717 previous undocumented portable approach to treating the option as
2718 unset was to force an expansion failure. That still works, and
2719 an empty string is now equivalent.
2721 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
2722 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
2723 not performing validation itself.
2725 PP/20 Added force_command boolean option to pipe transport.
2726 Patch from Nick Koston, of cPanel Inc.
2728 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
2731 TF/04 Added udpsend ACL modifier and hexquote expansion operator
2733 PP/21 Fix eximon continuous updating with timestamped log-files.
2734 Broken in a format-string cleanup in 4.80, missed when I repaired the
2735 other false fix of the same issue.
2736 Report and fix from Heiko Schlichting.
2739 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
2740 Report from Prashanth Katuri.
2742 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
2743 It's SecureTransport, so affects any MacOS clients which use the
2744 system-integrated TLS libraries, including email clients.
2746 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
2747 using a MIME ACL for non-SMTP local injection.
2748 Report and assistance in diagnosis by Warren Baker.
2750 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
2752 JH/16 Fix comparisons for 64b. Bugzilla 1385.
2754 TL/09 Add expansion variable $authenticated_fail_id to keep track of
2755 last id that failed so it may be referenced in subsequent ACL's.
2757 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
2760 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
2761 ldap library initialization, allowing self-signed CA's to be
2762 used. Also properly sets require_cert option later in code by
2763 using NULL (global ldap config) instead of ldap handle (per
2764 session). Bug diagnosis and testing by alxgomz.
2766 TL/12 Enhanced documentation in the ratelimit.pl script provided in
2767 the src/util/ subdirectory.
2769 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
2770 renamed to Transport Post Delivery Action by Jeremy Harris, as
2773 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
2774 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
2775 redis_servers = needs to be configured which will be used by the redis
2776 lookup. Patch from Warren Baker, of The Packet Hub.
2778 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
2780 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
2781 hostname or reverse DNS when processing a host list. Used suggestions
2782 from multiple comments on this bug.
2784 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
2786 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
2787 Missed a few lines, added it to make the runtest require no keyboard
2790 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
2791 contains upper case chars. Make router use caseful_local_part.
2793 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
2794 support when GnuTLS has been built with p11-kit.
2800 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
2802 This, or similar/improved, will also be change PP/11 of 4.82.
2808 PP/01 Handle short writes when writing local log-files.
2809 In practice, only affects FreeBSD (8 onwards).
2810 Bugzilla 1053, with thanks to Dmitry Isaikin.
2812 NM/01 Bugzilla 949 - Documentation tweak
2814 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
2817 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
2819 PP/02 Implemented gsasl authenticator.
2821 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
2823 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
2824 `pkg-config foo` for cflags/libs.
2826 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
2827 with rest of GSASL and with heimdal_gssapi.
2829 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
2830 `pkg-config foo` for cflags/libs for the TLS implementation.
2832 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
2833 properties get this fed in as external SSF. A number of robustness
2834 and debugging improvements to the cyrus_sasl authenticator.
2836 PP/08 cyrus_sasl server now expands the server_realm option.
2838 PP/09 Bugzilla 1214 - Log authentication information in reject log.
2839 Patch by Jeremy Harris.
2841 PP/10 Added dbmjz lookup type.
2843 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
2845 PP/12 MAIL args handles TAB as well as SP, for better interop with
2846 non-compliant senders.
2847 Analysis and variant patch by Todd Lyons.
2849 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
2850 Bug report from Lars Müller <lars@samba.org> (via SUSE),
2851 Patch from Dirk Mueller <dmueller@suse.com>
2853 PP/13 tls_peerdn now print-escaped for spool files.
2854 Observed some $tls_peerdn in wild which contained \n, which resulted
2855 in spool file corruption.
2857 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
2858 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
2859 or write after TLS renegotiation, which otherwise led to messages
2862 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
2863 as a tracking header (ie: a signed header comes before the signature).
2864 Patch from Wolfgang Breyha.
2866 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
2867 comma-sep list; embedded commas doubled.
2869 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
2871 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
2873 Report and patch from Dmitry Banschikov.
2875 PP/16 Removed "dont_insert_empty_fragments" from "openssl_options".
2876 Removed SSL_clear() after SSL_new() which led to protocol negotiation
2877 failures. We appear to now support TLS1.1+ with Exim.
2879 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
2880 lets Exim select keys and certificates based upon TLS SNI from client.
2881 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
2882 before an outbound SMTP session. New log_selector, +tls_sni.
2884 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
2885 NULL dereference. Report and patch from Alun Jones.
2887 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
2888 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
2889 Not seeing resolver debug output on NetBSD, but suspect this is a
2890 resolver implementation change.
2892 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
2893 Left warnings. Added "eximon gdb" invocation mode.
2895 PP/21 Defaulting "accept_8bitmime" to true, not false.
2897 PP/22 Added -bw for inetd wait mode support.
2899 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
2900 locate the relevant includes and libraries. Made this the default.
2902 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
2903 Bugzilla 1246, report and most of solution from Tomasz Kusy.
2905 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
2906 This may cause build issues on older platforms.
2908 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
2909 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
2910 gnutls_require_mac & gnutls_require_protocols (no longer supported).
2911 Added SNI support via GnuTLS too.
2912 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
2914 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
2916 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
2917 Applied second patch from Janne, implementing suggestion to default
2918 multiple-strings-in-record handling to match SPF spec.
2920 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
2922 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
2923 read-only, out of scope).
2924 Patch from Wolfgang Breyha, report from Stuart Northfield.
2926 PP/29 Fix three issues highlighted by clang analyser static analysis.
2927 Only crash-plausible issue would require the Cambridge-specific
2928 iplookup router and a misconfiguration.
2929 Report from Marcin Mirosław.
2931 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
2933 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
2934 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
2935 As part of this, removing so much warning spew let me fix some minor
2936 real issues in debug logging.
2938 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
2939 assignment on my part. Fixed.
2941 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
2942 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
2943 Janne Snabb (who went above and beyond: thank you).
2945 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
2946 string otherwise requires a connection and a bunch more work and it's
2947 relatively easy to get wrong. Should also expose TLS library linkage
2950 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
2951 64-bit ${eval} (JH/03).
2953 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
2954 GNU libc to support some of the 64-bit stuff, should not lead to
2955 conflicts. Defined before os.h is pulled in, so if a given platform
2956 needs to override this, it can.
2958 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
2959 protection layer was required, which is not implemented.
2960 Bugzilla 1254, patch from Wolfgang Breyha.
2962 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
2963 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
2964 tls_dhparam take prime identifiers. Also unbreak combination of
2965 OpenSSL+DH_params+TLSSNI.
2967 PP/39 Disable SSLv2 by default in OpenSSL support.
2973 PP/01 Solaris build fix for Oracle's LDAP libraries.
2974 Bugzilla 1109, patch from Stephen Usher.
2976 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
2978 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
2981 TF/02 Fix a couple more cases where we did not log the error message
2982 when unlink() failed. See also change 4.74-TF/03.
2984 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
2985 lock up or crash if it happened to be inside a call to libc when it
2986 got a SIGUSR1 from exiwhat.
2988 The SIGUSR1 handler appends the current process status to the process
2989 log which is later printed by exiwhat. It used to use the general
2990 purpose logging code to do this, but several functions it calls are
2991 not safe for signals.
2993 The new output code in the SIGUSR1 handler is specific to the process
2994 log, and simple enough that it's easy to inspect for signal safety.
2995 Removing some special cases also simplifies the general logging code.
2996 Removing the spurious timestamps from the process log simplifies
2999 TF/04 Improved ratelimit ACL condition.
3001 The /noupdate option has been deprecated in favour of /readonly which
3002 has clearer semantics. The /leaky, /strict, and /readonly update modes
3003 are mutually exclusive. The update mode is no longer included in the
3004 database key; it just determines when the database is updated. (This
3005 means that when you upgrade Exim will forget old rate measurements.)
3007 Exim now checks that the per_* options are used with an update mode that
3008 makes sense for the current ACL. For example, when Exim is processing a
3009 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
3010 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
3011 must specify per_mail/readonly. If you omit the update mode it defaults to
3012 /leaky where that makes sense (as before) or /readonly where required.
3014 The /noupdate option is now undocumented but still supported for
3015 backwards compatibility. It is equivalent to /readonly except that in
3016 ACLs where /readonly is required you may specify /leaky/noupdate or
3017 /strict/noupdate which are treated the same as /readonly.
3019 A useful new feature is the /count= option. This is a generalization
3020 of the per_byte option, so that you can measure the throughput of other
3021 aggregate values. For example, the per_byte option is now equivalent
3022 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
3024 The per_rcpt option has been generalized using the /count= mechanism
3025 (though it's more complicated than the per_byte equivalence). When it is
3026 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
3027 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
3028 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
3029 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
3030 Note that using per_rcpt with a non-readonly update mode in more than
3031 one ACL will cause the recipients to be double-counted. (The per_mail
3032 and per_byte options don't have this problem.)
3034 The handling of very low rates has changed slightly. If the computed rate
3035 is less than the event's count (usually one) then this event is the first
3036 after a long gap. In this case the rate is set to the same as this event's
3037 count, so that the first message of a spam run is counted properly.
3039 The major new feature is a mechanism for counting the rate of unique
3040 events. The new per_addr option counts the number of different
3041 recipients that someone has sent messages to in the last time period. It
3042 behaves like per_rcpt if all the recipient addresses are different, but
3043 duplicate recipient addresses do not increase the measured rate. Like
3044 the /count= option this is a general mechanism, so the per_addr option
3045 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
3046 example, measure the rate that a client uses different sender addresses
3047 with the options per_mail/unique=$sender_address. There are further
3048 details in the main documentation.
3050 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
3052 TF/06 Removed a few PCRE remnants.
3054 TF/07 Automatically extract Exim's version number from tags in the git
3055 repository when doing development or release builds.
3057 PP/02 Raise smtp_cmd_buffer_size to 16kB.
3058 Bugzilla 879. Patch from Paul Fisher.
3060 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
3061 Heavily based on revision 40f9a89a from Simon Arlott's tree.
3064 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
3066 PP/05 Variable $av_failed, true if the AV scanner deferred.
3067 Bugzilla 1078. Patch from John Horne.
3069 PP/06 Stop make process more reliably on build failure.
3070 Bugzilla 1087. Patch from Heiko Schlittermann.
3072 PP/07 Make maildir_use_size_file an _expandable_ boolean.
3073 Bugzilla 1089. Patch from Heiko Schlittermann.
3075 PP/08 Handle ${run} returning more data than OS pipe buffer size.
3076 Bugzilla 1131. Patch from Holger Weiß.
3078 PP/09 Handle IPv6 addresses with SPF.
3079 Bugzilla 860. Patch from Wolfgang Breyha.
3081 PP/10 GnuTLS: support TLS 1.2 & 1.1.
3083 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
3086 PP/11 match_* no longer expand right-hand-side by default.
3087 New compile-time build option, EXPAND_LISTMATCH_RHS.
3088 New expansion conditions, "inlist", "inlisti".
3090 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
3092 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
3094 PP/14 fix log_write() format string regression from TF/03.
3095 Bugzilla 1152. Patch from Dmitry Isaikin.
3101 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
3103 PP/02 Harmonised TLS library version reporting; only show if debugging.
3104 Layout now matches that introduced for other libraries in 4.74 PP/03.
3106 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
3108 PP/04 New "dns_use_edns0" global option.
3110 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
3113 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
3114 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
3116 TK/01 Updated PolarSSL code to 0.14.2.
3117 Bugzilla 1097. Patch from Andreas Metzler.
3119 PP/07 Catch divide-by-zero in ${eval:...}.
3120 Fixes bugzilla 1102.
3122 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
3125 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
3126 format-string attack -- SECURITY: remote arbitrary code execution.
3128 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
3129 time unintentionally subject to list matching rules, letting the header
3130 cause arbitrary Exim lookups (of items which can occur in lists, *not*
3131 arbitrary string expansion). This allowed for information disclosure.
3133 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
3134 INT_MIN/-1 -- value coerced to INT_MAX.
3140 NM/01 Workaround for PCRE version dependency in version reporting
3143 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
3144 This fixes portability to compilers other than gcc, notably
3145 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
3147 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
3148 makefiles for portability to HP-UX and POSIX correctness.
3150 PP/01 Permit LOOKUP_foo enabling on the make command-line.
3151 Also via indented variable definition in the Makefile.
3152 (Debugging by Oliver Heesakkers).
3154 PP/02 Restore caching of spamd results with expanded spamd_address.
3155 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
3157 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
3158 Improves build reliability. Fix from: Frank Elsner
3160 NM/02 Fix wide character breakage in the rfc2047 coding
3161 Fixes bug 1064. Patch from Andrey N. Oktyabrski
3163 NM/03 Allow underscore in dnslist lookups
3164 Fixes bug 1026. Patch from Graeme Fowler
3166 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
3167 Code patches from Adam Ciarcinski of NetBSD.
3169 NM/04 Fixed exiqgrep to cope with mailq missing size issue
3172 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
3173 is logged, to avoid truncation. Patch from John Horne.
3175 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
3176 Patch from Jakob Hirsch.
3178 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
3179 SQL string expansion failure details.
3180 Patch from Andrey Oktyabrski.
3182 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
3183 Patch from Simon Arlott.
3185 PP/09 New lookups functionality failed to compile on old gcc which rejects
3186 extern declarations in function scope.
3187 Patch from Oliver Fleischmann
3189 PP/10 Use sig_atomic_t for flags set from signal handlers.
3190 Check getgroups() return and improve debugging.
3191 Fixed developed for diagnosis in bug 927 (which turned out to be
3194 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
3195 Patch from Mark Zealey.
3197 PP/12 Bugzilla 1056: Improved spamd server selection.
3198 Patch from Mark Zealey.
3200 PP/13 Bugzilla 1086: Deal with maildir quota file races.
3201 Based on patch from Heiko Schlittermann.
3203 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
3204 Patch from Uwe Doering, sign-off by Michael Haardt.
3206 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
3207 variable declaration deep within a block. Bug and patch from
3210 PP/15 lookups-Makefile IRIX compatibility coercion.
3212 PP/16 Make DISABLE_DKIM build knob functional.
3214 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
3215 Patch by Simon Arlott
3217 TF/03 Fix valgrind.h portability to C89 compilers that do not support
3218 variable argument macros. Our copy now differs from upstream.
3224 TF/01 Failure to get a lock on a hints database can have serious
3225 consequences so log it to the panic log.
3227 TF/02 Log LMTP confirmation messages in the same way as SMTP,
3228 controlled using the smtp_confirmation log selector.
3230 TF/03 Include the error message when we fail to unlink a spool file.
3232 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
3233 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
3234 for maintaining out-of-tree patches for some time.
3236 PP/01 Bugzilla 139: Documentation and portability issues.
3237 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
3238 Handle per-OS dynamic-module compilation flags.
3240 PP/02 Let /dev/null have normal permissions.
3241 The 4.73 fixes were a little too stringent and complained about the
3242 permissions on /dev/null. Exempt it from some checks.
3243 Reported by Andreas M. Kirchwitz.
3245 PP/03 Report version information for many libraries, including
3246 Exim version information for dynamically loaded libraries. Created
3247 version.h, now support a version extension string for distributors
3248 who patch heavily. Dynamic module ABI change.
3250 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
3251 privilege escalation vulnerability whereby the Exim run-time user
3252 can cause root to append content of the attacker's choosing to
3255 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
3258 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
3259 If dropping privileges for untrusted macros, we disabled normal logging
3260 on the basis that it would fail; for the Exim run-time user, this is not
3261 the case, and it resulted in successful deliveries going unlogged.
3262 Fixed. Reported by Andreas Metzler.
3268 PP/01 Date: & Message-Id: revert to normally being appended to a message,
3269 only prepend for the Resent-* case. Fixes regression introduced in
3270 Exim 4.70 by NM/22 for Bugzilla 607.
3272 PP/02 Include check_rfc2047_length in configure.default because we're seeing
3273 increasing numbers of administrators be bitten by this.
3275 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
3277 PP/03 Bugzilla 994: added openssl_options main configuration option.
3279 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
3281 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
3283 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
3285 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
3286 without a peer certificate, leading to a segfault because of an
3287 assumption that peers always have certificates. Be a little more
3288 paranoid. Problem reported by Martin Tscholak.
3290 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
3291 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
3292 NB: ClamAV planning to remove STREAM in "middle of 2010".
3293 CL also introduces -bmalware, various -d+acl logging additions and
3294 more caution in buffer sizes.
3296 PP/09 Implemented reverse_ip expansion operator.
3298 PP/10 Bugzilla 937: provide a "debug" ACL control.
3300 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
3302 PP/12 Bugzilla 973: Implement --version.
3304 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
3306 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
3308 PP/15 Bugzilla 816: support multiple condition rules on Routers.
3310 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
3311 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
3312 ignore trailing whitespace.
3314 JJ/02 prevent non-panic DKIM error from being sent to paniclog
3316 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
3319 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
3320 Notification from Dr Andrew Aitchison.
3322 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
3323 ExtendedDetectionInfo response format.
3324 Notification from John Horne.
3326 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
3329 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
3330 XSL and documented dependency on system catalogs, with examples of how
3333 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
3336 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
3337 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
3338 configuration file which is writeable by the Exim user or group.
3340 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
3341 of configuration files to cover files specified with the -C option if
3342 they are going to be used with root privileges, not just the default
3345 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
3346 option (effectively making it always true).
3348 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
3349 files to be used while preserving root privileges.
3351 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
3352 that rogue child processes cannot use them.
3354 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
3355 run-time user, instead of root.
3357 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
3358 Exim run-time user without dropping privileges.
3360 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
3361 result string, instead of calling string_vformat() twice with the same
3364 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
3365 for other users. Others should always drop root privileges if they use
3366 -C on the command line, even for a whitelisted configure file.
3368 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
3370 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
3376 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
3377 $data_path, and $header_path variables; fixed documentation bugs and
3380 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
3381 exipick to access non-standard spools, including the "frozen" queue
3384 NM/01 Bugzilla 965: Support mysql stored procedures.
3385 Patch from Alain Williams
3387 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
3389 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
3390 Patch from Andreas Metzler
3392 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
3393 Patch from Kirill Miazine
3395 NM/05 Bugzilla 671: Added umask to procmail example.
3397 JJ/03 installed exipick 20100323.0, fixing doc bug
3399 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
3400 directory. Notification and patch from Dan Rosenberg.
3402 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
3404 TK/02 Improve log output when DKIM signing operation fails.
3406 MH/01 Treat the transport option dkim_domain as a colon separated
3407 list, not as a single string, and sign the message with each element,
3408 omitting multiple occurences of the same signer.
3410 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
3411 Bugzilla 985, 986. Patch by Simon Arlott
3413 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
3414 Patch by Simon Arlott
3416 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
3417 MBX locking. Notification from Dan Rosenberg.
3423 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
3425 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
3427 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
3429 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
3431 NM/04 Bugzilla 915: Flag broken perl installation during build.
3437 TK/01 Added patch by Johannes Berg that expands the main option
3438 "spamd_address" if it starts with a dollar sign.
3440 TK/02 Write list of recipients to X-Envelope-Sender header when building
3441 the mbox-format spool file for content scanning (suggested by Jakob
3444 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
3445 (http://www.dcc-servers.net/) support via dccifd. Activated by
3446 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
3448 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
3449 by Mark Daniel Reidel <mr@df.eu>.
3451 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
3452 When building exim an external PCRE library is now needed -
3453 PCRE is a system library on the majority of modern systems.
3454 See entry on PCRE_LIBS in EDITME file.
3456 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
3457 conversation. Added nologin parameter to request.
3458 Patch contributed by Kirill Miazine.
3460 TF/01 Do not log submission mode rewrites if they do not change the address.
3462 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
3464 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
3465 log files in place. Contributed by Roberto Lima.
3467 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
3469 TF/03 Bugzilla 615: When checking the local_parts router precondition
3470 after a local_part_suffix or local_part_prefix option, Exim now
3471 does not use the address's named list lookup cache, since this
3472 contains cached lookups for the whole local part.
3474 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
3475 Robert Millan. Documentation is in experimental-spec.txt.
3477 TF/04 Bugzilla 668: Fix parallel build (make -j).
3479 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
3481 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
3482 Patch provided by Jan Srzednicki.
3484 TF/05 Leading white space used to be stripped from $spam_report which
3485 wrecked the formatting. Now it is preserved.
3487 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
3488 that they are available at delivery time.
3490 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
3492 TF/08 TLS error reporting now respects the incoming_interface and
3493 incoming_port log selectors.
3495 TF/09 Produce a more useful error message if an SMTP transport's hosts
3496 setting expands to an empty string.
3498 NM/06 Bugzilla 744: EXPN did not work under TLS.
3499 Patch provided by Phil Pennock.
3501 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
3502 Patch provided by Richard Godbee.
3504 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
3505 acl_smtp_notquit, added index entry.
3507 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
3508 Patch provided by Eugene Bujak.
3510 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
3511 accept(). Patch provided by Maxim Dounin.
3513 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
3514 Patch provided by Phil Pennock.
3516 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
3518 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
3519 Patch provided by Brad "anomie" Jorsch.
3521 NM/14 Bugzilla 622: Added timeout setting to transport filter.
3522 Patch provided by Dean Brooks.
3524 TK/05 Add native DKIM support (does not depend on external libraries).
3526 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
3527 Patch provided by Graeme Fowler.
3529 NM/16 Bugzilla 851: Documentation example syntax fix.
3531 NM/17 Changed NOTICE file to remove references to embedded PCRE.
3533 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
3536 NM/19 Bugzilla 745: TLS version reporting.
3537 Patch provided by Phil Pennock.
3539 NM/20 Bugzilla 167: bool: condition support.
3540 Patch provided by Phil Pennock.
3542 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
3543 clients. Patch provided by Phil Pennock.
3545 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
3546 Patch provided by Brad "anomie" Jorsch.
3548 NM/23 Bugzilla 687: Fix misparses in eximstats.
3549 Patch provided by Heiko Schlittermann.
3551 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
3552 Patch provided by Heiko Schlittermann.
3554 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
3555 plus update to original patch.
3557 NM/26 Bugzilla 799: Documentation correction for ratelimit.
3559 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
3560 Patch provided by David Brownlee.
3562 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
3564 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
3566 NM/30 Bugzilla 888: TLS documentation bugfixes.
3568 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
3570 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
3571 Unlike the original bugzilla I have changed all shell scripts in src tree.
3573 NM/33 Bugzilla 898: Transport filter timeout fix.
3574 Patch by Todd Rinaldo.
3576 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
3577 Patch by Serge Demonchaux.
3579 NM/35 Bugzilla 39: Base64 decode bug fixes.
3580 Patch by Jakob Hirsch.
3582 NM/36 Bugzilla 909: Correct connect() call in dcc code.
3584 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
3586 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
3588 NM/39 Bugzilla 911: Fixed MakeLinks build script.
3594 TK/01 Add preliminary DKIM support. Currently requires a forked version of
3595 ALT-N's libdkim that I have put here:
3596 http://duncanthrax.net/exim-experimental/
3598 Note to Michael Haardt: I had to rename some vars in sieve.c. They
3599 were called 'true' and it seems that C99 defines that as a reserved
3600 keyword to be used with 'bool' variable types. That means you could
3601 not include C99-style headers which use bools without triggering
3602 build errors in sieve.c.
3604 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
3605 as mailq or other aliases. Changed the --help handling significantly
3606 to do whats expected. exim_usage() emits usage/help information.
3608 SC/01 Added the -bylocaldomain option to eximstats.
3610 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
3612 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
3614 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
3620 PH/01 Another patch from the Sieve maintainer.
3622 PH/02 When an IPv6 address is converted to a string for single-key lookup
3623 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
3624 dots are used instead of colons so that keys in lsearch files need not
3625 contain colons. This was done some time before quoting was made available
3626 in lsearch files. However, iplsearch files do require colons in IPv6 keys
3627 (notated using the quote facility) so as to distinguish them from IPv4
3628 keys. This meant that lookups for IP addresses in host lists did not work
3629 for iplsearch lookups.
3631 This has been fixed by arranging for IPv6 addresses to be expressed with
3632 colons if the lookup type is iplsearch. This is not incompatible, because
3633 previously such lookups could never work.
3635 The situation is now rather anomalous, since one *can* have colons in
3636 ordinary lsearch keys. However, making the change in all cases is
3637 incompatible and would probably break a number of configurations.
3639 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
3642 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
3643 conversion specification without a maximum field width, thereby enabling
3644 a rogue spamd server to cause a buffer overflow. While nobody in their
3645 right mind would setup Exim to query an untrusted spamd server, an
3646 attacker that gains access to a server running spamd could potentially
3647 exploit this vulnerability to run arbitrary code as the Exim user.
3649 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
3650 $primary_hostname instead of what libspf2 thinks the hosts name is.
3652 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
3653 a directory entry by the name of the lookup key. Previously, if a
3654 symlink pointed to a non-existing file or a file in a directory that
3655 Exim lacked permissions to read, a lookup for a key matching that
3656 symlink would fail. Now it is enough that a matching directory entry
3657 exists, symlink or not. (Bugzilla 503.)
3659 PH/03 The body_linecount and body_zerocount variables are now exported in the
3662 PH/04 Added the $dnslist_matched variable.
3664 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
3665 This means they are set thereafter only if the connection becomes
3668 PH/06 Added the client_condition to authenticators so that some can be skipped
3669 by clients under certain conditions.
3671 PH/07 The error message for a badly-placed control=no_multiline_responses left
3672 "_responses" off the end of the name.
3674 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
3676 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
3677 (without spaces) instead of just copying the configuration text.
3679 PH/10 Added the /noupdate option to the ratelimit ACL condition.
3681 PH/11 Added $max_received_linelength.
3683 PH/12 Added +ignore_defer and +include_defer to host lists.
3685 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
3686 way in which PCRE > 7.0 is built.
3688 PH/14 Implemented queue_only_load_latch.
3690 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
3691 MAIL command. The effect was to mangle the value on 64-bit systems.
3693 PH/16 Another patch from the Sieve maintainer.
3695 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
3697 PH/18 If a system quota error occurred while trying to create the file for
3698 a maildir delivery, the message "Mailbox is full" was not appended to the
3699 bounce if the delivery eventually timed out. Change 4.67/27 below applied
3700 only to a quota excession during the actual writing of the file.
3702 PH/19 It seems that peer DN values may contain newlines (and other non-printing
3703 characters?) which causes problems in log lines. The DN values are now
3704 passed through string_printing() before being added to log lines.
3706 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
3707 and InterBase are left for another time.)
3709 PH/21 Added message_body_newlines option.
3711 PH/22 Guard against possible overflow in moan_check_errorcopy().
3713 PH/23 POSIX allows open() to be a macro; guard against that.
3715 PH/24 If the recipient of an error message contained an @ in the local part
3716 (suitably quoted, of course), incorrect values were put in $domain and
3717 $local_part during the evaluation of errors_copy.
3723 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
3724 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
3727 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
3728 issue a MAIL command.
3730 PH/02 In an ACL statement such as
3732 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
3734 if a client was not listed at all, or was listed with a value other than
3735 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
3736 the condition was not true (as it should be), so access was not denied.
3737 The bug was that the ! inversion was incorrectly passed on to the second
3738 item. This has been fixed.
3740 PH/03 Added additional dnslists conditions == and =& which are different from
3741 = and & when the dns lookup returns more than one IP address.
3743 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
3744 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
3746 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
3747 FSYNC, which compiles an option called disable_fsync that allows for
3748 bypassing fsync(). The documentation is heavily laced with warnings.
3750 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
3752 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
3753 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
3754 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
3755 including adding "make clean"; (3) Added -fPIC when compiling the test
3756 dynamically loaded module, to get rid of a warning.
3758 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
3759 message fails, move_frozen_messages = true and ignore_bounce_errors_after
3760 = 0s. The bug is otherwise harmless.
3762 PH/07 There was a bug in the dovecot authenticator such that the value of
3763 $auth1 could be overwritten, and so not correctly preserved, after a
3764 successful authentication. This usually meant that the value preserved by
3765 the server_setid option was incorrect.
3767 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
3769 PH/09 Installed PCRE release 7.0.
3771 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
3772 run for batched SMTP input. It is now run at the start of every message
3773 in the batch. While fixing this I discovered that the process information
3774 (output by running exiwhat) was not always getting set for -bs and -bS
3775 input. This is fixed, and it now also says "batched" for BSMTP.
3777 PH/11 Added control=no_pipelining.
3779 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
3780 patch, slightly modified), and move the expansion of helo_data till after
3781 the connection is made in the smtp transport (so it can use these
3784 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
3786 PH/14 Added log_selector = +pid.
3788 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
3790 PH/16 Add ${if forany and ${if forall.
3792 PH/17 Added dsn_from option to vary the From: line in DSNs.
3794 PH/18 Flush SMTP output before performing a callout, unless control =
3795 no_callout_flush is set.
3797 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
3798 was true (the default) a successful delivery failed to delete the retry
3799 item, thus causing premature timeout of the address. The bug is now
3802 PH/20 Added hosts_avoid_pipelining to the smtp transport.
3804 PH/21 Long custom messages for fakedefer and fakereject are now split up
3805 into multiline responses in the same way that messages for "deny" and
3806 other ACL rejections are.
3808 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
3809 with slight modification.
3811 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
3812 draft, changing the syntax and factoring some duplicate code".
3814 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
3815 for deliveries of the second and subsequent messages over the same SMTP
3818 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
3819 ${reduce, with only minor "tidies".
3821 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
3823 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
3824 expansion side effects.
3826 PH/27 When a message times out after an over-quota error from an Exim-imposed
3827 quota, the bounce message says "mailbox is full". This message was not
3828 being given when it was a system quota that was exceeded. It now should
3831 MH/03 Made $recipients available in local_scan(). local_scan() already has
3832 better access to the recipient list through recipients_list[], but
3833 $recipients can be useful in postmaster-provided expansion strings.
3835 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
3836 in the case of a MAIL command with additional options following the
3837 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
3838 were accidentally chopped off.
3840 PH/29 SMTP synchronization checks are implemented when a command is read -
3841 there is a check that no more input is waiting when there shouldn't be
3842 any. However, for some commands, a delay in an ACL can mean that it is
3843 some time before the response is written. In this time, more input might
3844 arrive, invalidly. So now there are extra checks after an ACL has run for
3845 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
3846 pipelining has not been advertised.
3848 PH/30 MH's patch to allow iscntrl() characters to be list separators.
3850 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
3851 returned in the SMTP response when smtp_return_error_details was false.
3852 This has been fixed.
3854 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
3855 instead of the C I/O that was originally supplied, because problems were
3856 reported on Solaris.
3858 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
3859 Exim which did not show up earlier: it was assuming that a call to
3860 SSL_CTX_set_info_callback() might give an error value. In fact, there is
3861 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
3862 was a macro that became an assignment, so it seemed to work. This has
3863 changed to a proper function call with a void return, hence the compile
3864 error. Exim's code has been fixed.
3866 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
3869 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
3871 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
3873 PH/37 If a message is not accepted after it has had an id assigned (e.g.
3874 because it turns out to be too big or there is a timeout) there is no
3875 "Completed" line in the log. When some messages of this type were
3876 selected by exigrep, they were listed as "not completed". Others were
3877 picked up by some special patterns. I have improved the selection
3878 criteria to be more general.
3880 PH/38 The host_find_failed option in the manualroute router can now be set
3881 to "ignore", to completely ignore a host whose IP address cannot be
3882 found. If all hosts are ignored, the behaviour is controlled by the new
3883 host_all_ignored option.
3885 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
3886 homing or because of multiple MX records with /mx) generated more than
3887 one IP address, and the following item turned out to be the local host,
3888 all the secondary addresses of the first item were incorrectly removed
3889 from the list, along with the local host and any following hosts (which
3890 is what is supposed to happen).
3892 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
3893 whoever called Exim into the -H file. In the case of the daemon it was
3894 behaving confusingly. When first started, it used values for whoever
3895 started the daemon, but after a SIGHUP it used the Exim user (because it
3896 calls itself on a restart). I have changed the code so that it now always
3899 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
3900 message are rejected with the same error (e.g. no authentication or bad
3901 sender address), and a DATA command is nevertheless sent (as can happen
3902 with PIPELINING or a stupid MUA), the error message that was given to the
3903 RCPT commands is included in the rejection of the DATA command. This is
3904 intended to be helpful for MUAs that show only the final error to their
3907 PH/42 Another patch from the Sieve maintainer.
3909 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
3910 Eximstats - Fixed some broken HTML links and added missing column headers
3912 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
3913 columns for Rejects, Temp Rejects, Ham, and Spam rows.
3915 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
3917 PH/43 Yet another patch from the Sieve maintainer.
3919 PH/44 I found a way to check for a TCP/IP connection going away before sending
3920 the response to the final '.' that terminates a message, but only in the
3921 case where the client has not sent further data following the '.'
3922 (unfortunately, this is allowed). However, in many cases there won't be
3923 any further data because there won't be any more messages to send. A call
3924 to select() can be used: if it shows that the input is "ready", there is
3925 either input waiting, or the socket has been closed. An attempt to read
3926 the next input character can distinguish the two cases. Previously, Exim
3927 would have sent an OK response which the client would never have see.
3928 This could lead to message repetition. This fix should cure that, at
3929 least in a lot of common cases.
3931 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
3932 advertised in response to EHLO.
3938 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
3939 fixed by 4.65/MH/01 (is this a record?) are fixed:
3941 (i) An empty string was always treated as zero by the numeric comparison
3942 operators. This behaviour has been restored.
3944 (ii) It is documented that the numeric comparison operators always treat
3945 their arguments as decimal numbers. This was broken in that numbers
3946 starting with 0 were being interpreted as octal.
3948 While fixing these problems I realized that there was another issue that
3949 hadn't been noticed. Values of message_size_limit (both the global option
3950 and the transport option) were treated as octal if they started with 0.
3951 The documentation was vague. These values are now always treated as
3952 decimal, and I will make that clear in the documentation.
3958 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
3959 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
3962 MH/01 Don't check that the operands of numeric comparison operators are
3963 integers when their expansion is in "skipping" mode (fixes bug
3964 introduced by 4.64-PH/07).
3966 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
3967 child addresses, Exim now panics and dies. Previously, because the count
3968 is held in a short int, deliveries were likely to be lost. As such a
3969 large number of recipients for a single message is ridiculous
3970 (performance will be very, very poor), I have chosen to impose a limit
3971 rather than extend the field.
3977 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
3978 leftover -K file (the existence of which was triggered by #402).
3979 While we were at it, introduced process PID as part of the -K
3980 filename. This should rule out race conditions when creating
3983 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
3984 processing considerably. Previous code took too long for large mails,
3985 triggering a timeout which in turn triggers #401.
3987 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
3988 in the DK code in transports.c. sendfile() is not really portable,
3989 hence the _LINUX specificness.
3991 TF/01 In the add_headers option to the mail command in an Exim filter,
3992 there was a bug that Exim would claim a syntax error in any
3993 header after the first one which had an odd number of characters
3996 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
3997 callout verification, Exim cached a "reject" for the entire domain. This
3998 is correct for most verifications, but it is not correct for a recipient
3999 verification with use_sender or use_postmaster set, because in that case
4000 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
4001 case of MAIL FROM:<> rejection from other early rejections (e.g.
4002 rejection of HELO). When verifying a recipient using a non-null MAIL
4003 address, the cache is ignored if it shows MAIL FROM:<> rejection.
4004 Whatever the result of the callout, the value of the domain cache is
4005 left unchanged (for any other kind of callout, getting as far as trying
4006 RCPT means that the domain itself is ok).
4008 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
4011 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
4012 manifest itself as EPIPE rather than ECONNECT. When tidying away a
4013 session, the daemon ignores ECONNECT errors and logs others; it now
4014 ignores EPIPE as well.
4016 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
4017 (quoted-printable decoding).
4019 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
4020 later the small subsequent patch to fix an introduced bug.
4022 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
4024 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
4026 PH/08 An error is now given if message_size_limit is specified negative.
4028 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
4029 to be given (somewhat) arbitrary names.
4031 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
4034 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
4035 miscellaneous code fixes
4037 PH/10 Added the log_reject_target ACL modifier to specify where to log
4040 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
4041 hostname. This is wrong, because it relates to the incoming message (and
4042 probably the interface on which it is arriving) and not to the outgoing
4043 callout (which could be using a different interface). This has been
4044 changed to use the value of the helo_data option from the smtp transport
4045 instead - this is what is used when a message is actually being sent. If
4046 there is no remote transport (possible with a router that sets up host
4047 addresses), $smtp_active_hostname is used.
4049 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
4050 tweaks were necessary in order to get it to work (see also 21 below):
4051 (a) The code assumed that strncpy() returns a negative number on buffer
4052 overflow, which isn't the case. Replaced with Exim's string_format()
4054 (b) There were several signed/unsigned issues. I just did the minimum
4055 hacking in of casts. There is scope for a larger refactoring.
4056 (c) The code used strcasecmp() which is not a standard C function.
4057 Replaced with Exim's strcmpic() function.
4058 (d) The code set only $1; it now sets $auth1 as well.
4059 (e) A simple test gave the error "authentication client didn't specify
4060 service in request". It would seem that Dovecot has changed its
4061 interface. Fortunately there's a specification; I followed it and
4062 changed what the client sends and it appears to be working now.
4064 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
4067 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
4068 address A is aliased to B and C, where B exists and C does not. Without
4069 -v the output is "A verified" because verification stops after a
4070 successful redirection if more than one address is generated. However,
4071 with -v the child addresses are also verified. Exim was outputting "A
4072 failed to verify" and then showing the successful verification for C,
4073 with its parentage. It now outputs "B failed to verify", showing B's
4074 parentage before showing the successful verification of C.
4076 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
4077 look up a TXT record in a specific list after matching in a combined
4080 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
4081 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
4082 they consult the DNS. I had assumed they would set it the way they
4083 wanted; and indeed my experiments on Linux seem to show that in some
4084 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
4085 To be on the safe side, however, I have now made the interface to
4086 host_find_byname() similar to host_find_bydns(), with an argument
4087 containing the DNS resolver options. The host_find_byname() function now
4088 sets these options at its start, just as host_find_bydns() does. The smtp
4089 transport options dns_qualify_single and dns_search_parents are passed to
4090 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
4091 of host_find_byname() use the default settings of RES_DEFNAMES
4092 (qualify_single) but not RES_DNSRCH (search_parents).
4094 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
4095 spool_read_header() do less string testing, by means of a preliminary
4096 switch on the second character of optional "-foo" lines. (This is
4097 overdue, caused by the large number of possibilities that now exist.
4098 Originally there were few.) While I was there, I also converted the
4099 str(n)cmp tests so they don't re-test the leading "-" and the first
4100 character, in the hope this might squeeze out yet more improvement.
4102 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
4103 flag allowing group syntax was set by the header_syntax check but not
4104 turned off, possible causing trouble later; (2) The flag was not being
4105 set at all for the header_verify test, causing "group"-style headers to
4106 be rejected. I have now set it in this case, and also caused header_
4107 verify to ignore an empty address taken from a group. While doing this, I
4108 came across some other cases where the code for allowing group syntax
4109 while scanning a header line wasn't quite right (mostly, not resetting
4110 the flag correctly in the right place). These bugs could have caused
4111 trouble for malformed header lines. I hope it is now all correct.
4113 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
4114 with the "reply" argument non-NULL. The code, however (which originally
4115 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
4116 but it didn't always do it. This confused somebody who was copying the
4117 code for some other use. I have removed all the tests.
4119 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
4120 feature that was used to support insecure browsers during the U.S. crypto
4121 embargo. It requires special client support, and Exim is probably the
4122 only MTA that supported it -- and would never use it because real RSA is
4123 always available. This code has been removed, because it had the bad
4124 effect of slowing Exim down by computing (never used) parameters for the
4125 RSA_EXPORT functionality.
4127 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
4128 authenticator to fail if there's a tab character in the incoming data
4129 (there should never be unless someone is messing about, as it's supposed
4130 to be base64-encoded). Also added, on Timo's advice, the "secured" option
4131 if the connection is using TLS or if the remote IP is the same as the
4132 local IP, and the "valid-client-cert option" if a client certificate has
4135 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
4136 authenticators. This can be used for authorization after authentication
4137 succeeds. (In the case of plaintext, it servers for both authentication
4140 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
4141 if any retry times were supplied.
4143 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
4144 connection, where there is no client IP address to check. In this
4145 situation, the verify now always succeeds.
4147 PH/25 Applied John Jetmore's -Mset patch.
4149 PH/26 Added -bem to be like -Mset, but loading a message from a file.
4151 PH/27 In a string expansion for a processed (not raw) header when multiple
4152 headers of the same name were present, leading whitespace was being
4153 removed from all of them, but trailing whitespace was being removed only
4154 from the last one. Now trailing whitespace is removed from each header
4155 before concatenation. Completely empty headers in a concatenation (as
4156 before) are ignored.
4158 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
4159 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
4161 PH/29 [Removed. This was a change that I later backed out, and forgot to
4162 correct the ChangeLog entry (that I had efficiently created) before
4163 committing the later change.]
4165 PH/30 Exim was sometimes attempting to deliver messages that had suffered
4166 address errors (4xx response to RCPT) over the same connection as other
4167 messages routed to the same hosts. Such deliveries are always "forced",
4168 so retry times are not inspected. This resulted in far too many retries
4169 for the affected addresses. The effect occurred only when there were more
4170 hosts than the hosts_max_try setting in the smtp transport when it had
4171 the 4xx errors. Those hosts that it had tried were not added to the list
4172 of hosts for which the message was waiting, so if all were tried, there
4173 was no problem. Two fixes have been applied:
4175 (i) If there are any address or message errors in an SMTP delivery, none
4176 of the hosts (tried or untried) are now added to the list of hosts
4177 for which the message is waiting, so the message should not be a
4178 candidate for sending over the same connection that was used for a
4179 successful delivery of some other message. This seems entirely
4180 reasonable: after all the message is NOT "waiting for some host".
4181 This is so "obvious" that I'm not sure why it wasn't done
4182 previously. Hope I haven't missed anything, but it can't do any
4183 harm, as the worst effect is to miss an optimization.
4185 (ii) If, despite (i), such a delivery is accidentally attempted, the
4186 routing retry time is respected, so at least it doesn't keep
4187 hammering the server.
4189 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
4190 in ${readsocket because some servers need this prod.
4192 PH/32 Added some extra debug output when updating a wait-xxx database.
4194 PH/33 The hint "could be header name not terminated by colon", which has been
4195 given for certain expansion errors for a long time, was not being given
4196 for the ${if def:h_colon_omitted{... case.
4198 PH/34 The spec says: "With one important exception, whenever a domain list is
4199 being scanned, $domain contains the subject domain." There was at least
4200 one case where this was not true.
4202 PH/35 The error "getsockname() failed: connection reset by peer" was being
4203 written to the panic log as well as the main log, but it isn't really
4204 panic-worthy as it just means the connection died rather early on. I have
4205 removed the panic log writing for the ECONNRESET error when getsockname()
4208 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
4209 runs only) independently of the message's sender address. This meant
4210 that, if the 4xx error was in fact related to the sender, a different
4211 message to the same recipient with a different sender could confuse
4212 things. In particular, this can happen when sending to a greylisting
4213 server, but other circumstances could also provoke similar problems.
4214 I have changed the default so that the retry time for these errors is now
4215 based a combination of the sender and recipient addresses. This change
4216 can be overridden by setting address_retry_include_sender=false in the
4219 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
4220 remote server are returned as part of bounce messages. This was not
4221 happening for LMTP over a pipe (the lmtp transport), but now it is the
4222 same for both kinds of LMTP.
4224 PH/38 Despite being documented as not happening, Exim was rewriting addresses
4225 in header lines that were in fact CNAMEs. This is no longer the case.
4227 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
4228 and queue runs started by the daemon processed all messages. This has
4229 been fixed so that -R and -S can now usefully be given with -q<time>.
4231 PH/40 Import PCRE release 6.7 (fixes some bugs).
4233 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
4235 PH/42 Give an error if -q is specified more than once.
4237 PH/43 Renamed the variables $interface_address and $interface_port as
4238 $received_ip_address and $received_port, to make it clear that these
4239 values apply to message reception, and not to the outgoing interface when
4240 a message is delivered. (The old names remain recognized, of course.)
4242 PH/44 There was no timeout on the connect() call when using a Unix domain
4243 socket in the ${readsocket expansion. There now is.
4245 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
4246 be meaningful with "accept".
4248 SC/01 Eximstats V1.43
4249 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
4251 SC/02 Eximstats V1.44
4252 Use a glob alias rather than an array ref in the generated
4253 parser. This improves both readability and performance.
4255 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
4256 Collect SpamAssassin and rejection statistics.
4257 Don't display local sender or destination tables unless
4258 there is data to show.
4259 Added average volumes into the top table text output.
4261 SC/04 Eximstats V1.46
4262 Collect data on the number of addresses (recipients)
4263 as well as the number of messages.
4265 SC/05 Eximstats V1.47
4266 Added 'Message too big' to the list of mail rejection
4267 reasons (thanks to Marco Gaiarin).
4269 SC/06 Eximstats V1.48
4270 Mainlog lines which have GMT offsets and are too short to
4271 have a flag are now skipped.
4273 SC/07 Eximstats V1.49 (Alain Williams)
4274 Added the -emptyok flag.
4276 SC/08 Eximstats V1.50
4277 Fixes for obtaining the IP address from reject messages.
4279 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
4280 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
4281 whitespace changes from 4.64-PH/27
4283 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
4286 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
4287 are found, allow negative numbers in numeric criteria)
4289 JJ/06 exipick.20061117.2, added new $message_body_missing variable
4291 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
4292 to match changes made in 4.64-PH/43
4294 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
4296 PH/47 Put in an explicit test for a DNS lookup of an address record where the
4297 "domain" is actually an IP address, and force a failure. This locks out
4298 those revolvers/nameservers that support "A-for-A" lookups, in
4299 contravention of the specifications.
4301 PH/48 When a host name was looked up from an IP address, and the subsequent
4302 forward lookup of the name timed out, the host name was left in
4303 $sender_host_name, contrary to the specification.
4305 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
4306 restricted to single-key lookups, Exim was not diagnosing an error if
4307 * or *@ was used with a query-style lookup.
4309 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
4311 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
4312 long ago, but noone interested enough thought of it. Let's just say that
4313 the "1.1" means that there are some new functions that weren't there at
4314 some point in the past.
4316 PH/51 Error processing for expansion failure of helo_data from an smtp
4317 transport during callout processing was broken.
4319 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
4320 tested/used via the -bh/-bhc/-bs options.
4322 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
4323 bug, fixed in subsequent PCRE releases).
4325 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
4326 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
4328 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
4334 SC/01 Use a glob alias rather than an array ref in eximstats generated
4335 parser. This improves both readability and performance.
4337 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
4338 Don't display local sender or destination tables in eximstats unless
4339 there is data to show.
4340 Added average volumes into the eximstats top table text output.
4342 SC/03 Collect data on the number of addresses (recipients) as well
4343 as the number of messages in eximstats.
4345 TF/01 Correct an error in the documentation for the redirect router. Exim
4346 does (usually) call initgroups() when daemonizing.
4348 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
4349 with consistent privilege compared to when running as a daemon.
4351 TF/03 Note in the spec that $authenticated_id is not set for local
4352 submissions from trusted users.
4354 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
4355 Thanks to Dean Brooks <dean@iglou.com> for the patch.
4357 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
4358 by adding some example configuration directives to the default
4359 configuration file. A little bit of work is required to uncomment the
4360 directives and define how usernames and passwords are checked, but
4361 there is now a framework to start from.
4363 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
4364 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
4365 without this. I don't know how relevant this is to other LDAP libraries.
4367 PH/02 Add the verb name to the "unknown ACL verb" error.
4369 PH/03 Magnus Holmgren's patch for filter_prepend_home.
4371 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
4373 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
4374 directory not expanded when it should be if an expanded home directory
4375 was set for the address (which is overridden by the transport).
4377 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
4380 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
4381 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
4382 because it is too late at that time, and has no effect.
4384 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
4385 security issue with \' (bugzilla #107). I could not use the
4386 PQescapeStringConn() function, because it needs a PGconn value as one of
4389 PH/08 When testing addresses using -bt, indicate those final addresses that
4390 are duplicates that would not cause an additional delivery. At least one
4391 person was confused, thinking that -bt output corresponded to deliveries.
4392 (Suppressing duplicates isn't a good idea as you lose the information
4393 about possibly different redirections that led to the duplicates.)
4395 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
4396 systems where poll() doesn't work, in particular OS X.
4398 PH/10 Added more information to debugging output for retry time not reached.
4400 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
4401 operations in malware.c.
4403 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
4406 PH/13 If write_rejectlog was set false when logging was sent to syslog with
4407 syslog_duplication set false, log lines that would normally be written
4408 both the the main log and to the reject log were not written to syslog at
4411 PH/14 In the default configuration, change the use of "message" in ACL warn
4412 statements to "add_header".
4414 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
4415 not followed by a command (e.g. "seen endif").
4417 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
4418 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
4421 PH/17 Added extra conditions to the default value of delay_warning_condition
4425 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
4426 { match{$h_precedence:}{(?i)bulk|list|junk} } \
4427 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
4430 The Auto-Submitted: and various List- headers are standardised, whereas I
4431 don't think Precedence: ever was.
4433 PH/18 Refactored debugging code in route_finduser() to show more information,
4434 in particular, the error code if getpwnam() issues one.
4436 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
4437 This is apparently needed in addition to the PH/07 change above to avoid
4438 any possible encoding problems.
4440 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
4441 but not after initializing Perl.
4443 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
4444 output them only if debugging. By default they are written stderr,
4445 apparently, which is not desirable.
4447 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
4450 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
4453 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
4455 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
4456 authenticated or an ident call has been made. Suppress the default
4457 values for $authenticated_id and $authenticated_sender (but permit -oMai
4458 and -oMas) when testing with -bh.
4460 PH/24 Re-jigged the order of the tests in the default configuration so that the
4461 tests for valid domains and recipients precede the DNS black list and CSA
4462 tests, on the grounds that those ones are more expensive.
4464 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
4465 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
4466 This bug exists in every version of Exim that I still have, right back to
4469 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
4470 However, an attempt to turn on case-sensitivity in a regex key by
4471 including (?-i) didn't work because the subject string was already
4472 lowercased, and the effects were non-intuitive. It turns out that a
4473 one-line patch can be used to allow (?-i) to work as expected.
4479 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
4480 other effects) broke the use of negated acl sub-conditions.
4482 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
4485 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
4486 "Deny" causes Exim to reject the incoming connection with a 554 error.
4487 Unfortunately, if there is a major crisis, such as a disk failure,
4488 tcp-wrappers gives "deny", whereas what one would like would be some
4489 kind of temporary error. A kludge has been added to help with this.
4490 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
4491 554 error is used if errno is still zero or contains ENOENT (which occurs
4492 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
4495 PH/03 Add -lutil to the default FreeBSD LIBS setting.
4497 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
4498 errors. Otherwise a message that provokes a temporary error (when other
4499 messages do not) can cause a whole host to time out.
4501 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
4502 addresses were routed directly to files or pipes from a redirect router.
4503 File deliveries just didn't batch; pipe deliveries might have suffered
4506 PH/06 A failure to get a lock for a hints database would erroneously always say
4507 "Failed to get write lock", even when it was really a read lock.
4509 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
4510 of 0600. This has been changed to use the value of the lockfile_mode
4511 option (which defaults to 0600).
4513 PH/08 Applied small patch from the Sieve maintainer.
4515 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
4516 folder from quota calculations, a direct delivery into this folder messed
4517 up the contents of the maildirsize file. This was because the regex was
4518 used only to exclude .Trash (or whatever) when the size of the mailbox
4519 was calculated. There was no check that a delivery was happening into an
4520 excluded directory. This bug has been fixed by ignoring all quota
4521 processing for deliveries into excluded directories.
4523 PH/10 Added the maildirfolder_create_regex option to appendfile.
4529 PH/01 The code for finding all the local interface addresses on a FreeBSD
4530 system running IPv6 was broken. This may well have applied to all BSD
4531 systems, as well as to others that have similar system calls. The broken
4532 code found IPv4 interfaces correctly, but gave incorrect values for the
4533 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
4534 that it would not match correctly against @[] and not recognize the IPv6
4537 PH/02 The ipliteral router was not recognizing addresses of the form user@
4538 [ipv6:....] because it didn't know about the "ipv6:" prefix.
4540 PH/03 Added disable_ipv6.
4542 PH/04 Changed $reply_address to use the raw form of the headers instead of the
4543 decoded form, because it is most often used to construct To: headers
4544 lines in autoreplies, and the decoded form may well be syntactically
4545 invalid. However, $reply_address has leading white space removed, and all
4546 newlines turned into spaces so that the autoreply transport does not
4549 PH/05 If group was specified without a user on a router, and no group or user
4550 was specified on a transport, the group from the router was ignored.
4552 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
4553 for visible compile-time settings that can be used to change these
4554 numbers, for those that want even more. Backwards compatibility with old
4555 spool files has been maintained. However, going back to a previous Exim
4556 release will lost any variables that are in spool files.
4558 PH/07 Two small changes when running in the test harness: increase delay when
4559 passing a TCP/IP connection to a new process, in case the original
4560 process has to generate a bounce, and remove special handling of
4561 127.0.0.2 (sic), which is no longer necessary.
4563 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
4564 be the same on different OS.
4566 PH/09 Moved a debug statement in filter processing to avoid a race problem when
4569 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
4570 whether --show-vars was specified or not
4572 JJ/02 exipick: Added support for new ACL variable spool format introduced
4575 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
4576 syntactically invalid From: or Reply-to: line, and a filter used this to
4577 generate an autoreply, and therefore failed to obtain an address for the
4578 autoreply, Exim could try to deliver to a non-existent relative file
4579 name, causing unrelated and misleading errors. What now happens is that
4580 it logs this as a hard delivery error, but does not attempt to create a
4583 PH/11 The exinext utility has a -C option for testing purposes, but although
4584 the given file was scanned by exinext itself; it wasn't being passed on
4585 when Exim was called.
4587 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
4588 an end-of-file indication when reading a command response.
4590 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
4591 compiled. In many other places in Exim, IPv6 addresses are always
4592 recognized, so I have changed this. It also means that IPv4 domain
4593 literals of the form [IPV4:n.n.n.n] are now always recognized.
4595 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
4596 used if the router is not running as root, for example, when verifying at
4597 ACL time, or when using -bh. The debugging output from this situation was
4598 non-existent - all you got was a failure to exec. I have made two
4601 (a) Failures to set uid/gid, the current directory, or a process leader
4602 in a subprocess such as that created by queryprogram now generate
4603 suitable debugging output when -d is set.
4605 (b) The queryprogram router detects when it is not running as root,
4606 outputs suitable debugging information if -d is set, and then runs
4607 the subprocess without attempting to change uid/gid.
4609 PH/15 Minor change to Makefile for building test_host (undocumented testing
4612 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
4613 additional section of a DNS packet that returns MX or SRV records.
4614 Instead, it always explicitly searches for A/AAAA records. This avoids
4615 major problems that occur when a DNS server includes only records of one
4616 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
4617 fixed another bug: if SRV records were looked up and the corresponding
4618 address records were *not* found in the additional section, the port
4619 values from the SRV records were lost.
4621 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
4622 using the correct key (the original address) when searching the retry
4623 rules in order to find which one to use for generating the retry hint.
4625 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
4626 adding the default one. Similarly, if it contains a Reply-To: header, the
4627 errors_reply_to option, if set, is not used.
4629 PH/19 When calculating a retry time, Exim used to measure the "time since
4630 failure" by looking at the "first failed" field in the retry record. Now
4631 it does not use this if it is later than than the arrival time of the
4632 message. Instead it uses the arrival time. This makes for better
4633 behaviour in cases where some deliveries succeed, thus re-setting the
4634 "first failed" field. An example is a quota failure for a huge message
4635 when small messages continue to be delivered. Without this change, the
4636 "time since failure" will always be short, possible causing more frequent
4637 delivery attempts for the huge message than are intended.
4638 [Note: This change was subsequently modified - see PH/04 for 4.62.]
4640 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
4641 $1, $2, $3) because the numerical variables can be reset during some
4642 expansion items (e.g. "match"), thereby losing the authentication data.
4644 PH/21 Make -bV show the size of off_t variables so that the test suite can
4645 decide whether to run tests for quotas > 2G.
4647 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
4648 mailbox_size, and mailbox_filecount in the appendfile transport. If a
4649 filecount value is greater than 2G or if a quota value is greater than 2G
4650 on a system where the size of off_t is not greater than 4, a panic error
4653 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
4654 never match. The debug and -bh output now contains an explicit error
4655 message indicating a malformed IPv4 address or mask.
4657 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
4658 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
4659 PH/23 above applies.
4661 PH/25 Do not write to syslog when running in the test harness. The only
4662 occasion when this arises is a failure to open the main or panic logs
4663 (for which there is an explicit test).
4665 PH/26 Added the /no_tell option to "control=freeze".
4667 PH/27 If a host name lookup failed very early in a connection, for example, if
4668 the IP address matched host_lookup and the reverse lookup yielded a name
4669 that did not have a forward lookup, an error message of the form "no IP
4670 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
4671 could be logged. Now it outputs the IP address instead of "NULL".
4673 PH/28 An enabling patch from MH: add new function child_open_exim2() which
4674 allows the sender and the authenticated sender to be set when
4675 submitting a message from within Exim. Since child_open_exim() is
4676 documented for local_scan(), the new function should be too.
4678 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
4679 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
4680 results in an empty string is now treated as unset.
4682 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
4684 PH/31 Added sender_verify_fail logging option.
4686 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
4687 needed by RFC 822 but not by RFC 2822 was commented out. I have now
4688 tidied the source and removed it altogether.
4690 PH/33 When a queue run was abandoned because the load average was too high, a
4691 log line was always written; now it is written only if the queue_run log
4692 selector is set. In addition, the log line for abandonment now contains
4693 information about the queue run such as the pid. This is always present
4694 in "start" and "stop" lines but was omitted from the "abandon" line.
4696 PH/34 Omit spaces between a header name and the colon in the error message that
4697 is given when verify = headers_syntax fails (if there are lots of them,
4698 the message gets confusing).
4700 PH/35 Change the default for dns_check_names_pattern to allow slashes within
4701 names, as there are now some PTR records that contain slashes. This check
4702 is only to protect against broken name servers that fall over on strange
4703 characters, so the fact that it applies to all lookups doesn't matter.
4705 PH/36 Now that the new test suite is complete, we can remove some of the
4706 special code in Exim that was needed for the old test suite. For example,
4707 sorting DNS records because real resolvers return them in an arbitrary
4708 order. The new test suite's fake resolver always returns records in the
4711 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
4712 bounces) except when queue_only is set, to avoid logging races between
4713 the different processes.
4715 PH/38 Panic-die if .include specifies a non-absolute path.
4717 PH/39 A tweak to the "H" retry rule from its user.
4719 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
4720 a label. They prevented compilation on older perls.
4722 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
4723 a warning to be raised on newish perls.
4725 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
4726 on queue. Changes to match documented behaviour of showing count of
4727 messages matching specified criteria.
4729 PH/40 Changed the default ident timeout from 30s to 5s.
4731 PH/41 Added support for the use of login_cap features, on those BSD systems
4732 that have them, for controlling the resources used by pipe deliveries.
4734 PH/42 The content-scanning code uses fopen() to create files in which to put
4735 message data. Previously it was not paying any attention to the mode of
4736 the files. Exim runs with umask(0) because the rest of the code creates
4737 files with open(), and sets the required mode explicitly. Thus, these
4738 files were ending up world-writeable. This was not a big issue, because,
4739 being within the spool directory, they were not world-accessible. I have
4740 created a function called modefopen, which takes an additional mode
4741 argument. It sets umask(777), creates the file, chmods it to the required
4742 mode, then resets the umask. All the relevant calls to fopen() in the
4743 content scanning code have been changed to use this function.
4745 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
4746 to 24 hours. This avoids potential overflow problems when processing G
4747 and H retry rules. I suspect nobody ever tinkers with this value.
4749 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
4751 PH/45 When the plaintext authenticator is running as a client, the server's
4752 challenges are checked to ensure they are valid base64 strings. By
4753 default, the authentication attempt is cancelled if an invalid string is
4754 received. Setting client_ignore_invalid_base64 true ignores these errors.
4755 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
4756 they are received. Thus, the responses can be made to depend on the
4757 challenges. If an invalid string is ignored, an empty string is placed in
4760 PH/46 Messages that are created by the autoreply transport now contains a
4761 References: header, in accordance with RFCs 2822 and 3834.
4763 PH/47 Added authenticated_sender_force to the smtp transport.
4765 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
4767 PH/49 Installed latest patch from the Sieve maintainer.
4769 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
4770 was also set, the appendfile transport was unnecessarily scanning a
4771 directory of message files (e.g. for maildir delivery) to find the count
4772 of files (along with the size), even though it did not need this
4773 information. It now does the scan only if it needs to find either the
4774 size of the count of files.
4776 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
4778 PH/52 Two bugs concerned with error handling when the smtp transport is
4781 (i) Exim was not creating retry information for temporary errors given
4782 for individual recipients after the DATA command when the smtp transport
4783 was used in LMTP mode. This meant that they could be retried too
4784 frequently, and not timed out correctly.
4786 (ii) Exim was setting the flag that allows error details to be returned
4787 for LMTP errors on RCPT commands, but not for LMTP errors for individual
4788 recipients that were returned after the DATA command.
4790 PH/53 This is related to PH/52, but is more general: for any failing address,
4791 when detailed error information was permitted to be returned to the
4792 sender, but the error was temporary, then after the final timeout, only
4793 "retry timeout exceeded" was returned. Now it returns the full error as
4794 well as "retry timeout exceeded".
4796 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
4797 do this, and (what is worse) MTAs that accept it.
4799 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
4800 will now be deprecated.
4802 PH/56 New os.c-cygwin from the Cygwin maintainer.
4804 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
4805 formats (previously only available in exim formats via -bpr, -bpru,
4806 and -bpra. Now also available in native and exiqgrep formats)
4808 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
4809 with very large, slow to parse queues
4811 JJ/08 exipick: added ! as generic prefix to negate any criteria format
4813 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
4815 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
4816 responses to authentication challenges, though it was showing the
4817 challenges; (ii) I've removed the CR characters from the debug output for
4820 PH/58 Allow for the insertion of a newline as well as a space when a string
4821 is turned into more than one encoded-word during RFC 2047 encoding. The
4822 Sieve code now uses this.
4824 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
4825 data_4xx, lost_connection, tls_required.
4827 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
4828 message was being sent as an SMTP response.
4830 PH/61 Add -l and -k options to exicyclog.
4832 PH/62 When verifying, if an address was redirected to one new address, so that
4833 verification continued, and the new address failed or deferred after
4834 having set something in $address_data, the value of $address_data was not
4835 passed back to the ACL. This was different to the case when no
4836 redirection occurred. The value is now passed back in both cases.
4838 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
4839 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
4840 use login_cap.h, so on its own it isn't the distinguishing feature. The
4841 new name refers directly to the setclassresources() function.
4843 PH/65 Added configuration files for NetBSD3.
4845 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
4847 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
4848 is preferred over IPv4.
4850 PH/68 The bounce_return_message and bounce_return_body options were not being
4851 honoured for bounces generated during the reception of non-SMTP messages.
4852 In particular, this applied to messages rejected by the ACL. This bug has
4853 been fixed. However, if bounce_return_message is true and bounce_return_
4854 body is false, the headers that are returned for a non-SMTP message
4855 include only those that have been read before the error was detected.
4856 (In the case of an ACL rejection, they have all been read.)
4858 PH/69 The HTML version of the specification is now built in a directory called
4859 spec_html instead of spec.html, because the latter looks like a path with
4860 a MIME-type, and this confuses some software.
4862 PH/70 Catch two compiler warnings in sieve.c.
4864 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
4865 function verify_get_ident() calls ip_connect() to connect a socket, but
4866 if the "connect()" function timed out, ip_connect() used to close the
4867 socket. However, verify_get_ident() also closes the socket later, and in
4868 between Exim writes to the log, which may get opened at this point. When
4869 the socket was closed in ip_connect(), the log could get the same file
4870 descriptor number as the socket. This naturally causes chaos. The fix is
4871 not to close the socket in ip_connect(); the socket should be closed by
4872 the function that creates it. There was only one place in the code where
4873 this was missing, in the iplookup router, which I don't think anybody now
4874 uses, but I've fixed it anyway.
4876 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
4877 well as to direct DNS lookups. Otherwise the handling of names in host
4878 lists is inconsistent and therefore confusing.
4884 PH/01 Two changes to the default runtime configuration:
4886 (1) Move the checks for relay_from_hosts and authenticated clients from
4887 after to before the (commented out) DNS black list checks.
4889 (2) Add control=submission to the relay_from_hosts and authenticated
4890 clients checks, on the grounds that messages accepted by these
4891 statements are most likely to be submissions.
4893 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
4895 (1) Generate an error if the third argument for the ${prvs expansion is
4898 (2) Treat a missing third argument of ${prvscheck as if it were an empty
4901 (3) Reset the variables that are obtained from the first argument of
4902 ${prvscheck and used in the second argument before leaving the code,
4903 because their memory is reclaimed, so using them afterwards may do
4906 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
4907 one (it's much easier than Tom thought :-).
4909 (5) Because of (4), we can now allow for the use of $prvscheck_result
4910 inside the third argument.
4912 PH/03 For some reason, the default setting of PATH when running a command from
4913 a pipe transport was just "/usr/bin". I have changed it to
4916 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
4917 anything to be listed in the output from -bV.
4919 PH/05 When a filter generated an autoreply, the entire To: header line was
4920 quoted in the delivery log line, like this:
4922 => >A.N.Other <ano@some.domain> <original@ddress> ...
4924 This has been changed so that it extracts the operative address. There
4925 may be more than one such address. If so, they are comma-separated, like
4928 => >ano@some.domain,ona@other.domain <original@ddress> ...
4930 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
4931 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
4932 not being looked up in the rDNS to get a host name, Exim was showing the
4933 IP address twice in Received: lines, even though the IP addresses were
4934 identical. For example:
4936 Received: from [1.2.3.4] (helo=[1.2.3.4])
4938 However, if the real host name was known, it was omitting the HELO data
4939 if it matched the actual IP address. This has been tidied up so that it
4940 doesn't show the same IP address twice.
4942 PH/07 When both +timestamp and +memory debugging was on, the value given by
4943 $tod_xxx expansions could be wrong, because the tod_stamp() function was
4944 called by the debug printing, thereby overwriting the timestamp buffer.
4945 Debugging no longer uses the tod_stamp() function when +timestamp is set.
4947 PH/08 When the original message was included in an autoreply transport, it
4948 always said "this is a copy of the message, including all the headers",
4949 even if body_only or headers_only was set. It now gives an appropriate
4952 PH/09 Applied a patch from the Sieve maintainer which:
4954 o fixes some comments
4955 o adds the (disabled) notify extension core
4956 o adds some debug output for the result of if/elsif tests
4957 o points to the current vacation draft in the documentation
4958 and documents the missing references header update
4962 o fixes a bug in processing the envelope test (when testing
4963 multiple envelope elements, the last element determined the
4966 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
4967 Electronic Mail") by including:
4969 Auto-submitted: auto-generated
4971 in the messages that it generates (bounce messages and others, such as
4972 warnings). In the case of bounce messages for non-SMTP messages, there was
4973 also a typo: it was using "Auto_submitted" (underscore instead of
4974 hyphen). Since every message generated by Exim is necessarily in response
4975 to another message, thes have all been changed to:
4977 Auto-Submitted: auto-replied
4979 in accordance with these statements in the RFC:
4981 The auto-replied keyword:
4983 - SHOULD be used on messages sent in direct response to another
4984 message by an automatic process,
4986 - MUST NOT be used on manually-generated messages,
4988 - MAY be used on Delivery Status Notifications (DSNs) and Message
4989 Disposition Notifications (MDNs),
4991 - MUST NOT be used on messages generated by automatic or periodic
4992 processes, except for messages which are automatic responses to
4995 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
4996 to the default Received: header definition.
4998 PH/12 Added log selector acl_warn_skipped (default on).
5000 PH/13 After a successful wildlsearch lookup, discard the values of numeric
5001 variables because (a) they are in the wrong storage pool and (b) even if
5002 they were copied, it wouldn't work properly because of the caching.
5004 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
5005 checking when decoding. Apparently there are clients that generate
5006 overlong encoded strings. Why am I not surprised?
5008 PH/15 If the first argument of "${if match_address" was not empty, but did not
5009 contain an "@" character, Exim crashed. Now it writes a panic log message
5010 and treats the condition as false.
5012 PH/16 In autoreply, treat an empty string for "once" the same as unset.
5014 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
5015 extension "envelope-auth". The code is finished and in agreement with
5016 other implementations, but there is no documentation so far and in fact,
5017 nobody wrote the draft yet. This extension is currently #undef'ed, thus
5018 not changing the active code.
5020 Print executed "if" and "elsif" statements when debugging is used. This
5021 helps a great deal to understand what a filter does.
5023 Document more things not specified clearly in RFC3028. I had all this
5024 sorted out, when out of a sudden new issues came to my mind. Oops."
5026 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
5029 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
5030 canonical form (as documented). However, after a host name lookup from
5031 the IP address, check_host() was doing a simple string comparison with
5032 addresses acquired from the DNS when checking that the found name did
5033 have the original IP as one of its addresses. Since any found IPv6
5034 addresses are likely to be in abbreviated form, the comparison could
5035 fail. Luckily, there already exists a function for doing the comparison
5036 by converting both addresses to binary, so now that is used instead of
5037 the text comparison.
5039 PH/20 There was another similar case to PH/19, when a complete host name was
5040 given in a host list; looking up its IP address could give an abbreviated
5041 form, whereas the current host's name might or might not be abbreviated.
5042 The same fix has been applied.
5048 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
5049 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
5052 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
5053 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
5055 PH/03 Typo: missing ".o" in src/pcre/Makefile.
5057 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
5058 header line, restrict the check to what is listed in RFCs 2369 and 2929.
5059 Also, for "Auto-Submitted", treat anything other than "no" as
5060 non-personal, in accordance with RFC 3834. (Previously it treated
5061 anything starting "auto-" as non-personal.)
5063 TF/01 The control=submission/name=... option had a problem with syntax
5064 errors if the name included a slash character. The /name= option
5065 now slurps the rest of the string, so it can include any characters
5066 but it must come last in the list of options (after /sender_retain
5069 PH/05 Some modifications to the interface to the fake nameserver for the new
5077 TK/01 Added the "success_on_redirect" address verification option. See
5078 NewStuff for rationale and an example.
5080 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
5082 PH/02 Patch to exigrep to allow it to work on syslog lines.
5084 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
5085 fread() to skip over the body file's header line, because in Cygwin the
5086 header line is locked and is inaccessible.
5088 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
5089 co-exist for some time) to make it clear that it is the Exim ID that is
5090 referenced, not the Message-ID: header line.
5092 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
5093 string_format() function, because snprintf() does not exist on all
5096 PH/06 The use of forbid_filter_existstest now also locks out the use of the
5097 ${stat: expansion item.
5099 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
5100 protocol synchronization error", to keep the pedants happy.
5102 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
5103 well as for IRIX systems, when gcc is being used. See the host.c source
5106 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
5108 PH/10 Named domain lists were not working if used in a queue_smtp_domains
5111 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
5112 transport and to the smtp transport in LMTP mode.
5114 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
5116 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
5117 run a filter in a subprocess. This could lead to confusion in subsequent
5118 lookups in the parent process. There should also be a search_tidyup() at
5119 the end of the subprocess.
5121 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
5122 only if the host matched helo_try_verify_hosts, which caused the
5123 verification to occur when the EHLO/HELO command was issued. The ACL just
5124 tested the remembered result. Now, if a previous verification attempt has
5125 not happened, "verify = helo" does it there and then.
5127 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
5129 TK/03 Fix log output including CR from clamd.
5131 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
5132 exist provoked a memory error which could cause a segfault.
5134 PH/15 Installed PCRE 6.2
5136 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
5138 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
5139 of the problem. Specifically, suggested +O2 rather than +O1 for the
5142 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
5144 PH/20 If a delivery was routed to a non-standard port by means of an SRV
5145 record, the port was not correctly logged when the outgoing_port log
5146 selector was set (it logged the transort's default port).
5148 PH/21 Added support for host-specific ports to manualroute, queryprogram,
5149 fallback_hosts, and "hosts" in the smtp transport.
5151 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
5152 host errors such as "Connection refused".
5154 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
5155 authentication with radiusclient 0.4.9:
5157 - Error returned from rc_read_config was caught wrongly
5158 - Username/password not passed on to radius server due to wrong length.
5160 The presumption is that some radiusclient API changes for 4.51/PH/17
5161 were not taken care of correctly. The code is still untested by me (my
5162 Linux distribution still has 0.3.2 of radiusclient), but it was
5163 contributed by a Radius user.
5165 PH/24 When doing a callout, the value of $domain wasn't set correctly when
5166 expanding the "port" option of the smtp transport.
5168 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
5169 while reading a MIME header. Thanks to Tom Hughes for a patch.
5171 PH/24 Include config.h inside local_scan.h so that configuration settings are
5174 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
5175 that in an ACL for RCPT (for example), you can examine exactly what was
5178 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
5179 commands, but it was not correctly comparing the address with the actual
5180 client host address. Thus, it would show the EHLO address in Received:
5181 header lines when this was not necessary.
5183 PH/27 Added the % operator to ${eval:}.
5185 PH/28 Exim tries to create and chdir to its spool directory when it starts;
5186 it should be ignoring failures (because with -C, for example, it has lost
5187 privilege). It wasn't ignoring creation failures other than "already
5190 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
5193 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
5194 because an input line was too long, either on its own, or by virtue of
5195 too many continuations, the temporary file was not being removed, and the
5196 return code was incorrect.
5198 PH/31 Missing "BOOL" in function definition in filtertest.c.
5200 PH/32 Applied Sieve patches from the maintainer.
5202 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
5204 PH/33 Added "verify = not_blind".
5206 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
5207 Local/Makefile (with some defaults set). These are used in built scripts
5208 such as exicyclog, but they have never been used in the exim_install
5209 script (though there are many overriding facilities there). I have
5210 arranged that the exim_install script now takes note of these two
5213 PH/35 Installed configuration files for Dragonfly.
5215 PH/36 When a locally submitted message by a trusted user did not contain a
5216 From: header, and the sender address was obtained from -f or from an SMTP
5217 MAIL command, and the trusted user did not use -F to supply a sender
5218 name, $originator_name was incorrectly used when constructing a From:
5219 header. Furthermore, $originator_name was used for submission mode
5220 messages from external hosts without From: headers in a similar way,
5221 which is clearly wrong.
5223 PH/37 Added control=suppress_local_fixups.
5225 PH/38 When log_selector = +received_sender was set, and the addition of the
5226 sender made the log line's construction buffer exactly full, or one byte
5227 less than full, an overflow happened when the terminating "\n" was
5230 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
5231 when the result of a list match is failure because a DNS lookup failed.
5233 PH/40 RM_COMMAND is now used in the building process.
5235 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
5236 the "build-* directories that it finds.
5238 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
5239 address in a domain literal was a prefix of an interface address.
5241 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
5242 when verifying a sender address, unless rewrite_headers is false.
5244 PH/44 Wrote a long comment about why errors_to addresses are verified as
5245 recipients, not senders.
5247 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
5248 the ratelimit ACL was added.
5250 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
5252 PH/46 Added extra information about PostgreSQL errors to the error string.
5254 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
5255 avoiding the need to install special zones in a real server. This is
5256 backwards compatible; if it can't find the fake resolver, it drops back.
5257 Thus, both old and new test suites can be run.
5259 TF/02 Added util/ratelimit.pl
5261 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
5262 clock is set back in time.
5264 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
5265 Candler <B.Candler@pobox.com>.
5267 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
5268 OK for addresses that are the result of redirections.
5270 PH/48 A number of further additions for the benefit of the new test suite,
5271 including a fake gethostbyname() that interfaces to the fake DNS resolver
5274 TF/06 The fix for widen_domains has also been applied to qualify_single and
5275 search_parents which are the other dnslookup options that can cause
5278 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
5281 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
5283 TF/07 Exim produced the error message "an SRV record indicated no SMTP
5284 service" if it encountered an MX record with an empty target hostname.
5285 The message is now "an MX or SRV record indicated no SMTP service".
5287 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
5288 if the DNS of the sending site is misconfigured. This is quite a
5289 common situation. This change restores the behaviour of treating a
5290 helo verification defer as a failure.
5292 PH/51 If self=fail was set on a router, the bounce message did not include the
5293 actual error message.
5299 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
5301 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
5302 command itself ended in error, the underlying message about the transport
5303 filter timeout was being overwritten with the pipe command error. Now the
5304 underlying error message should be appended to the second error message.
5306 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
5308 PH/02 Reduce the amount of output that "make" produces by default. Full output
5309 can still be requested.
5311 PH/03 The warning log line about a condition test deferring for a "warn" verb
5312 was being output only once per connection, rather than after each
5313 occurrence (because it was using the same function as for successful
5314 "warn" verbs). This seems wrong, so I have changed it.
5316 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
5317 it should not have, which might have caused a crash in the right
5318 circumstances, but probably never did.
5320 PH/04 Installed a modified version of Tony Finch's patch to make submission
5321 mode fix the return path as well as the Sender: header line, and to
5322 add a /name= option so that you can make the user's friendly name appear
5325 TF/03 Added the control = fakedefer ACL modifier.
5327 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
5328 Mark Lowes for thorough testing.
5330 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
5332 TK/03 Merged latest SRS patch from Miles Wilton.
5334 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
5335 with the definition in sysexits.h (which is #included earlier).
5336 Fortunately, Exim does not actually use EX_OK. The code used to try to
5337 preserve the sysexits.h value, by assuming that macro definitions were
5338 scanned for macro replacements. I have been disabused of this notion,
5339 so now the code just undefines EX_OK before #including unistd.h.
5341 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
5342 in the smtp transport. When a block could not be written in a single
5343 write() function, the timeout was being re-applied to each part-write.
5344 This seems wrong - if the receiver was accepting one byte at a time it
5345 would take for ever. The timeout is now adjusted when this happens. It
5346 doesn't have to be particularly precise.
5348 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
5349 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
5351 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
5352 without a domain if the check to <postmaster@domain> fails.
5354 SC/01 Eximstats: added -xls and the ability to specify output files
5355 (patch written by Frank Heydlauf).
5357 SC/02 Eximstats: use FileHandles for outputting results.
5359 SC/03 Eximstats: allow any combination of xls, txt, and html output.
5361 SC/04 Eximstats: fixed display of large numbers with -nvr option
5363 SC/05 Eximstats: fixed merging of reports with empty tables.
5365 SC/06 Eximstats: added the -include_original_destination flag
5367 SC/07 Eximstats: removed tabs and trailing whitespace.
5369 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
5371 TK/06 MBOX spool code: Add real "From " MBOX separator line
5372 so the .eml file is really in mbox format (even though
5373 most programs do not really care). Patch from Alex Miller.
5375 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
5376 The latter is generated from $received_to and is only set if the
5377 message has one envelope recipient. SA can use these headers,
5378 obviously out-of-the-box. Patch from Alex Miller.
5380 PH/08 The ${def test on a variable was returning false if the variable's
5381 value was "0", contrary to what the specification has always said!
5382 The result should be true unless the variable is empty.
5384 PH/09 The syntax error of a character other than { following "${if
5385 def:variable_name" (after optional whitespace) was not being diagnosed.
5386 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
5387 accidental colon was present, for example, could give incorrect results.
5389 PH/10 Tidied the code in a number of places where the st_size field of a stat()
5390 result is used (not including appendfile, where other changes are about
5393 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
5394 This involved changing a lot of size variables from int to off_t. It
5395 should work with maildirs and everything.
5397 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
5398 spamd dying while we are connected to it.
5400 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
5403 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
5404 function for BDB 4.3.
5406 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
5408 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
5409 the sources for PCRE have been re-arranged and more files are now
5412 PH/15 The code I had for printing potentially long long variables in PH/11
5413 above was not the best (it lost precision). The length of off_t variables
5414 is now inspected at build time, and an appropriate printing format (%ld
5415 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
5416 to be "long long int" or "long int". This is needed for the internal
5417 formatting function string_vformat().
5419 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
5420 the configuration file to be ":syslog", then the script "guesses" where
5421 the logs files are, rather than using the compiled in default. In our
5422 case the guess is not the same as the compiled default, so the script
5423 suddenly stopped working when I started to use syslog. The patch checks
5424 to see if log_file_path is "". If so, it attempts to read it from exim
5425 with no configuration file to get the compiled in version, before it
5426 falls back to the previous guessing code."
5428 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
5429 implementing BATV in an Exim configuration. See NewStuff for the gory
5432 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
5433 Makefile that are specific to HP-UX.
5435 PH/18 If the "use_postmaster" option was set for a recipient callout together
5436 with the "random" option, the postmaster address was used as the MAIL
5437 FROM address for the random test, but not for the subsequent recipient
5438 test. It is now used for both.
5440 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
5441 patch removes a few documentation additions to RFC 3028, because the
5442 latest draft now contains them. It adds the new en;ascii-case comparator
5443 and a new error check for 8bit text in MIME parts. Comparator and
5444 require names are now matched exactly. I enabled the subaddress
5445 extension, but it is not well tested yet (read: it works for me)."
5447 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
5448 rework some of the code of TK/09 above to avoid the hardwired use of
5449 "%lld" and "long long". Replaced the call to snprintf() with a call to
5452 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
5453 records point to non-existent hosts", "retry timeout exceeded", and
5454 "retry time not reached for any host after a long failure period".
5456 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
5457 experimental DomainKeys support:
5459 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
5460 (2) On an error such as an illegally used "control", the wrong name for
5461 the control was given.
5463 These problems did NOT occur unless DomainKeys support was compiled.
5465 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
5467 PH/24 Added ${if match_ip condition.
5469 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
5470 databases so that it will be absolutely obvious if a crash occurs in the
5471 DB library. This is a regular occurrence (often caused by mis-matched
5474 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
5475 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
5476 picked up on a user's system that detects such things. There doesn't seem
5477 to be a gcc warning option for this - only an attribute that has to be
5478 put on the function's prototype. It seems that in Fedora Core 4 they have
5479 set this on a number of new functions. No doubt there will be more in due
5482 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
5483 specify a transport. However, if an address that was verified by such a
5484 router was the subject of a callout, Exim crashed because it tried to
5485 read the rcpt_include_affixes from the non-existent transport. Now it
5486 just assumes that the setting of that option is false. This bug was
5487 introduced by 4.51/PH/31.
5489 PH/28 Changed -d+all to exclude +memory, because that information is very
5490 rarely of interest, but it makes the output a lot bigger. People tend to
5491 do -d+all out of habit.
5493 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
5494 code in os-type was giving problems when libc.so lives in lib64, like on
5497 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
5498 aren't the modern standard, and it seems that some systems' include files
5499 don't always have them. Exim was already checking for some of the newer
5500 ones like T_AAAA, and defining it itself. I've added checks for all the
5501 record types that Exim uses.
5503 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
5504 not automatically generating a new one, as it is supposed to. This
5505 prevented TLS from working. If the file did exist, but contained invalid
5506 data, a new version was generated, as expected. It was only the case of a
5507 non-existent file that was broken.
5509 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
5510 with a change in libdomainkeys > 0.64.
5512 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
5513 from DNS. If the selector record carries the flag, it now has
5514 precedence over the domain-wide flag.
5516 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
5518 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
5519 the use of a port name that isn't defined in /etc/services) occurred, the
5520 message was deferred as in a normal delivery, and thus remained on the
5521 spool, instead of being failed because of the mua_wrapper setting. This
5522 is now fixed, and I tidied up some of the mua_wrapper messages at the
5525 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
5526 the messages in a hash of arrays rather than using individual hashes.
5527 This is a bit cleaner and results in dramatic memory savings, albeit
5528 at a slight CPU cost.
5530 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
5531 as requested by Marc Sherman.
5533 SC/10 Eximstats: added histograms for user specified patterns as requested
5536 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
5538 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
5539 fopen() in the content-scanning modules that did not already have it.
5545 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
5546 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
5548 TK/02 Fix ACL "control" statement not being available in MIME ACL.
5550 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
5552 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
5553 to test Sieve filters that use "vacation".
5555 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
5556 that changes the way the GnuTLS parameters are stored in the cache file.
5557 The new format can be generated externally. For backward compatibility,
5558 if the data in the cache doesn't make sense, Exim assumes it has read an
5559 old-format file, and it generates new data and writes a new file. This
5560 means that you can't go back to an older release without removing the
5563 PH/03 A redirect router that has both "unseen" and "one_time" set does not
5564 work if there are any delivery delays because "one_time" forces the
5565 parent to be marked "delivered", so its unseen clone is never tried
5566 again. For this reason, Exim now forbids the simultaneous setting of
5569 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
5570 redirected to themselves ("homonym" addresses). Read the long ChangeLog
5571 entry if you want to know the details. The fix, however, neglected to
5572 consider the case when local delivery batching is involved. The test for
5573 "previously delivered" was not happening when checking to see if an
5574 address could be batched with a previous (undelivered) one; under
5575 certain circumstances this could lead to multiple deliveries to the same
5578 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
5579 in its include files, and this causes problems building Exim.
5581 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
5582 header_syntax) but Exim was just ignoring anything given after a slash.
5583 In particular, this caused confusion with an attempt to use "verify =
5584 reverse_host_lookup/defer_ok". An error is now given when options are
5585 supplied for verify items that do not have them. (Maybe reverse_host_
5586 lookup should have a defer_ok option, but that's a different point.)
5588 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
5589 defined by RFC 821) to 2048, because there were problems with some AUTH
5590 commands, and RFC 1869 says the size should be increased for extended
5591 SMTP commands that take arguments.
5593 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
5596 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
5597 "unknown" error; now it says that the functionality isn't in the binary.
5599 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
5600 an address' error message when a string expansion fails (syntax or
5601 whatever). Otherwise the password may appear in the log. Following change
5602 PH/42 below, there is no longer a chance of it appearing in a bounce
5605 PH/11 Installed exipick version 20050225.0 from John Jetmore.
5607 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
5608 of its addresses was ever tried. (Bugzilla bug #2.)
5610 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
5611 the result incorrectly in the debug output. (It correctly added a newline
5612 to what was transported.)
5614 TF/01 Added $received_time.
5616 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
5617 commented out examples of how to interface to a virus scanner and to
5618 SpamAssassin. Also added commented examples of av_scanner and
5619 spamd_address settings.
5621 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
5622 and controls are allowed in which ACLs. There were a couple of minor
5623 errors. Some of the entries in the conditions table (which is a table of
5624 where they are NOT allowed) were getting very unwieldy; rewrote them as a
5625 negation of where the condition IS allowed.
5627 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
5629 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
5630 header file does not have a version number, so I've had to invent a new
5631 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
5632 API. The code is untested by me (my Linux distribution still has 0.3.2 of
5633 radiusclient), but it was contributed by a Radius user.
5635 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
5636 files or directories, for OpenSSL.
5638 PH/19 When an Exim process that is running as root has to create an Exim log
5639 file, it does so in a subprocess that runs as exim:exim so as to get the
5640 ownership right at creation (otherwise, other Exim processes might see
5641 the file with the wrong ownership). There was no test for failure of this
5642 fork() call, which would lead to the process getting stuck as it waited
5643 for a non-existent subprocess. Forks do occasionally fail when resources
5644 run out. I reviewed all the other calls to fork(); they all seem to check
5647 PH/20 When checking for unexpected SMTP input at connect time (before writing
5648 the banner), Exim was not dealing correctly with a non-positive return
5649 from the read() function. If the client had disconnected by this time,
5650 the result was a log entry for a synchronization error with an empty
5651 string after "input=" when read() returned zero. If read() returned -1
5652 (an event I could not check), uninitialized data bytes were printed.
5653 There were reports of junk text (parts of files, etc) appearing after
5656 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
5658 PH/22 Added support for macro redefinition, and (re)definition in between
5659 driver and ACL definitions.
5661 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
5662 forgetting to use the resulting value; it was using the unexpanded value.
5664 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
5665 hadn't been configured. The fix is from Juergen Kreileder, who
5666 understands it better than I do:
5668 "Here's what I see happening with three configured cyrus_sasl
5669 authenticators configured (plain, login, cram-md5):
5671 On startup auth_cyrus_sasl_init() gets called for each of these.
5672 This means three calls to sasl_listmech() without a specified mech_list.
5673 => SASL tests which mechs of all available mechs actually work
5674 => three warnings about OTP not working
5675 => the returned list contains: plain, login, cram-md5, digest-md5, ...
5677 With the patch, sasl_listmech() also gets called three times. But now
5678 SASL's mech_list option is set to the server_mech specified in the the
5679 authenticator. Or in other words, the answer from sasl_listmech()
5680 gets limited to just the mech you're testing for (which is different
5682 => the return list contains just 'plain' or 'login', 'cram-md5' or
5683 nothing depending on the value of ob->server_mech.
5685 I've just tested the patch: Authentication still works fine,
5686 unavailable mechs specified in the exim configuration are still
5687 caught, and the auth.log warnings about OTP are gone."
5689 PH/25 When debugging is enabled, the contents of the command line are added
5690 to the debugging output, even when log_selector=+arguments is not
5693 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
5694 answer is "GNU", and only if the return is "GNU/something" is the answer
5697 PH/27 $acl_verify_message is now set immediately after the failure of a
5698 verification in an ACL, and so is available in subsequent modifiers. In
5699 particular, the message can be preserved by coding like this:
5701 warn !verify = sender
5702 set acl_m0 = $acl_verify_message
5704 Previously, $acl_verify_message was set only while expanding "message"
5705 and "log_message" when a very denied access.
5707 PH/28 Modified OS/os.c-Linux with
5709 -#ifndef OS_LOAD_AVERAGE
5710 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
5712 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
5713 nomenclature these days.)
5715 PH/29 Installed patch from the Sieve maintainer that adds the options
5716 sieve_useraddress and sieve_subaddress to the redirect router.
5718 PH/30 In these circumstances:
5719 . Two addresses routed to the same list of hosts;
5720 . First host does not offer TLS;
5721 . First host accepts first address;
5722 . First host gives temporary error to second address;
5723 . Second host offers TLS and a TLS session is established;
5724 . Second host accepts second address.
5725 Exim incorrectly logged both deliveries with the TLS parameters (cipher
5726 and peerdn, if requested) that were in fact used only for the second
5729 PH/31 When doing a callout as part of verifying an address, Exim was not paying
5730 attention to any local part prefix or suffix that was matched by the
5731 router that accepted the address. It now behaves in the same way as it
5732 does for delivery: the affixes are removed from the local part unless
5733 rcpt_include_affixes is set on the transport.
5735 PH/32 Add the sender address, as F=<...>, to the log line when logging a
5736 timeout during the DATA phase of an incoming message.
5738 PH/33 Sieve envelope tests were broken for match types other than :is. I have
5739 applied a patch sanctioned by the Sieve maintainer.
5741 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
5742 the uid or gid is negative. A case of a negative gid caused this to be
5743 noticed. The fix allows for either to be negative.
5745 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
5746 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
5749 PH/36 The change PH/12 above was broken. Fixed it.
5751 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
5752 the grounds that routing the same address twice would always produce the
5753 same answer. This might have been true once, but it is certainly no
5754 longer true now. Routing a child address may depend on the previous
5755 routing that produced that child. Some complicated redirection strategies
5756 went wrong when messages had multiple recipients, and made Exim's
5757 behaviour dependent on the order in which the addresses were given.
5759 I have moved the duplicate checking until after the routing is complete.
5760 Exim scans the addresses that are assigned to local and remote
5761 transports, and removes any duplicates. This means that more work will be
5762 done, as duplicates will always all be routed, but duplicates are
5763 presumably rare, so I don't expect this is of any significance.
5765 For deliveries to pipes, files, and autoreplies, the duplicate checking
5766 still happens during the routing process, since they are not going to be
5769 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
5770 It corrects a timeout issue with spamd. This is Ian's comment: "The
5771 background is that sometimes spamd either never reads data from a
5772 connection it has accepted, or it never writes response data. The exiscan
5773 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
5774 blindly assumes that writes won't block so it may never time out."
5776 PH/39 Allow G after quota size as well as K and M.
5778 PH/40 The value set for $authenticated_id in an authenticator may not contain
5779 binary zeroes or newlines because the value is written to log lines and
5780 to spool files. There was no check on this. Now the value is run through
5781 the string_printing() function so that such characters are converted to
5782 printable escape sequences.
5784 PH/41 $message_linecount is a new variable that contains the total number of
5785 lines in the message. Compare $body_linecount, which is the count for the
5788 PH/42 Exim no longer gives details of delivery errors for specific addresses in
5789 bounce and delay warning messages, except in certain special cases, which
5792 (a) An SMTP error message from a remote host;
5793 (b) A message specified in a :fail: redirection;
5794 (c) A message specified in a "fail" command in a system filter;
5795 (d) A message specified in a FAIL return from the queryprogram router;
5796 (e) A message specified by the cannot_route_message router option.
5798 In these cases only, Exim does include the error details in bounce and
5799 warning messages. There are also a few cases where bland messages such
5800 as "unrouteable address" or "local delivery error" are given.
5802 PH/43 $value is now also set for the "else" part of a ${run expansion.
5804 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
5805 being worked on, but at least Exim now implements the latest version to
5808 PH/45 In a pipe transport, although a timeout while waiting for the pipe
5809 process to complete was treated as a delivery failure, a timeout while
5810 writing the message to the pipe was logged, but erroneously treated as a
5811 successful delivery. Such timeouts include transport filter timeouts. For
5812 consistency with the overall process timeout, these timeouts are now
5813 treated as errors, giving rise to delivery failures by default. However,
5814 there is now a new Boolean option for the pipe transport called
5815 timeout_defer, which, if set TRUE, converts the failures into defers for
5816 both kinds of timeout. A transport filter timeout is now identified in
5819 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
5820 systems where "make" and "gmake" are different, calling "gmake" at top
5821 level broke things. I've arranged for the value of $(MAKE) to be passed
5822 from the Makefile to this script so that it can call the same version of
5826 A note about Exim versions 4.44 and 4.50
5827 ----------------------------------------
5829 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
5830 changes of various kinds. As a consequence, a big documentation update was
5831 needed. This delayed the release for rather longer than seemed good, especially
5832 in the light of a couple of (minor) security issues. Therefore, the changes
5833 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
5834 release. So 4.44 and 4.50 are in effect two different branches that both start
5837 I have left the 4.50 change log unchanged; it contains all the changes since
5838 4.43. The change log for 4.44 is below; many of its items are identical to
5839 those for 4.50. This seems to be the most sensible way to preserve the
5840 historical information.
5846 1. Minor wording change to the doc/README.SIEVE file.
5848 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
5849 computation of the current number of files was incorrect.
5851 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
5852 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
5855 4. Give more explanation in the error message when the command for a transport
5856 filter fails to execute.
5858 5. There are several places where Exim runs a non-Exim command in a
5859 subprocess. The SIGUSR1 signal should be disabled for these processes. This
5860 was being done only for the command run by the queryprogram router. It is
5861 now done for all such subprocesses. The other cases are: ${run, transport
5862 filters, and the commands run by the lmtp and pipe transports.
5864 6. Added CONFIGURE_GROUP build-time option.
5866 7. Some older OS have a limit of 256 on the maximum number of file
5867 descriptors. Exim was using setrlimit() to set 1000 as a large value
5868 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
5869 systems. I've change it so that if it can't get 1000, it tries for 256.
5871 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
5872 was an oversight, and furthermore, ever since the addition of extra
5873 controls (e.g. 4.43/32), the checks on when to allow different forms of
5874 "control" were broken. There should now be diagnostics for all cases when a
5875 control that does not make sense is encountered.
5877 9. Added the /retain_sender option to "control=submission".
5879 10. $recipients is now available in the predata ACL (oversight).
5881 11. Tidy the search cache before the fork to do a delivery from a message
5882 received from the command line. Otherwise the child will trigger a lookup
5883 failure and thereby defer the delivery if it tries to use (for example) a
5884 cached ldap connection that the parent has called unbind on.
5886 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
5887 of $address_data from the recipient verification was clobbered by the
5888 sender verification.
5890 13. The value of address_data from a sender verification is now available in
5891 $sender_address_data in subsequent conditions in the ACL statement.
5893 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
5895 15. Added a new option "connect=<time>" to callout options, to set a different
5898 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
5899 was its contents. (It was OK if the option was not defined at all.)
5901 17. A "Completed" log line is now written for messages that are removed from
5902 the spool by the -Mrm option.
5904 18. New variables $sender_verify_failure and $recipient_verify_failure contain
5905 information about exactly what failed.
5907 19. Added -dd to debug only the daemon process.
5909 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
5910 handles timeouts, both on the server side and network timeouts. Renamed the
5911 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
5913 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
5914 It is now set to "smtps".
5916 22. $host_address is now set to the target address during the checking of
5917 ignore_target_hosts.
5919 23. When checking ignore_target_hosts for an ipliteral router, no host name was
5920 being passed; this would have caused $sender_host_name to have been used if
5921 matching the list had actually called for a host name (not very likely,
5922 since this list is usually IP addresses). A host name is now passed as
5925 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
5926 code that specifies a non-restarting handler (typically sigaction() in
5927 modern systems) in an attempt to fix a rare and obscure crash bug.
5929 25. Narrowed the window for a race in the daemon that could cause it to ignore
5930 SIGCHLD signals. This is not a major problem, because they are used only to
5931 wake it up if nothing else does.
5933 26. A malformed maildirsize file could cause Exim to calculate negative values
5934 for the mailbox size or file count. Odd effects could occur as a result.
5935 The maildirsize information is now recalculated if the size or filecount
5938 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
5939 support for a long time. Removed HAVE_SYS_VFS_H.
5941 28. Installed the latest version of exipick from John Jetmore.
5943 29. In an address list, if the pattern was not a regular expression, an empty
5944 subject address (from a bounce message) matched only if the pattern was an
5945 empty string. Non-empty patterns were not even tested. This was the wrong
5946 because it is perfectly reasonable to use an empty address as part of a
5947 database query. An empty address is now tested by patterns that are
5948 lookups. However, all the other forms of pattern expect the subject to
5949 contain a local part and a domain, and therefore, for them, an empty
5950 address still always fails if the pattern is not itself empty.
5952 30. Exim went into a mad DNS loop when attempting to do a callout where the
5953 host was specified on an smtp transport, and looking it up yielded more
5954 than one IP address.
5956 31. Re-factored the code for checking spool and log partition space into a
5957 function that finds that data and another that does the check. The former
5958 is then used to implement four new variables: $spool_space, $log_space,
5959 $spool_inodes, and $log_inodes.
5961 32. The RFC2047 encoding function was originally intended for short strings
5962 such as real names; it was not keeping to the 75-character limit for
5963 encoded words that the RFC imposes. It now respects the limit, and
5964 generates multiple encoded words if necessary. To be on the safe side, I
5965 have increased the buffer size for the ${rfc2047: expansion operator from
5968 33. It is now permitted to omit both strings after an "if" condition; if the
5969 condition is true, the result is "true". As before, when the second string
5970 is omitted, a false condition yields an empty string. This makes it less
5971 cumbersome to write custom ACL and router conditions.
5973 34. Failure to deliver a bounce message always caused it to be frozen, even if
5974 there was an errors_to setting on the router. The errors_to setting is now
5977 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
5978 canonical form (fully expanded) before being placed in
5979 $sender_host_address.
5981 36. The table in the code that translates DNS record types into text (T_A to
5982 "A" for instance) was missing entries for NS and CNAME. It is just possible
5983 that this could have caused confusion if both these types were looked up
5984 for the same domain, because the text type is used as part of Exim's
5985 per-process caching. But the chance of anyone hitting this buglet seems
5988 37. The dnsdb lookup has been extended in a number of ways.
5990 (1) There is a new type, "zns", which walks up the domain tree until it
5991 finds some nameserver records. It should be used with care.
5993 (2) There is a new type, "mxh", which is like "mx" except that it returns
5994 just the host names, not the priorities.
5996 (3) It is now possible to give a list of domains (or IP addresses) to be
5997 looked up. The behaviour when one of the lookups defers can be
5998 controlled by a keyword.
6000 (4) It is now possible to specify the separator character for use when
6001 multiple records are returned.
6003 38. The dnslists ACL condition has been extended: it is now possible to supply
6004 a list of IP addresses and/or domains to be looked up in a particular DNS
6007 39. Added log_selector=+queue_time_overall.
6009 40. When running the queue in the test harness, wait just a tad after forking a
6010 delivery process, to get repeatability of debugging output.
6012 41. Include certificate and key file names in error message when GnuTLS fails
6013 to set them up, because the GnuTLS error message doesn't include the name
6014 of the failing file when there is a problem reading it.
6016 42. Allow both -bf and -bF in the same test run.
6018 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
6020 44. The "Exiscan patch" is now merged into the mainline Exim source.
6022 45. Sometimes the final signoff response after QUIT could fail to get
6023 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
6024 < 0 before doing a fflush(). This bug looks as though it goes back to the
6025 introduction of TLS in release 3.20, but "sometimes" must have been rare
6026 because the tests only now provoked it.
6028 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
6029 (this can affect the format of dates).
6031 47. exim_tidydb, when checking for the continued existence of a message for
6032 which it has found a message-specific retry record, was not finding
6033 messages that were in split spool directories. Consequently, it was
6034 deleting retry records that should have stayed in existence.
6036 48. Steve fixed some bugs in eximstats.
6038 49. The SPA authentication driver was not abandoning authentication and moving
6039 on to the next authenticator when an expansion was forced to fail,
6040 contradicting the general specification for all authenticators. Instead it
6041 was generating a temporary error. It now behaves as specified.
6043 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
6044 (the order specifies the preference for clients). The order is now AES256,
6045 AES128, 3DES, ARCFOUR128.
6047 51. Small patch to Sieve code - explicitly set From: when generating an
6050 52. Exim crashed if a remote delivery caused a very long error message to be
6051 recorded - for instance if somebody sent an entire SpamAssassin report back
6052 as a large number of 550 error lines. This bug was coincidentally fixed by
6053 increasing the size of one of Exim's internal buffers (big_buffer) that
6054 happened as part of the Exiscan merge. However, to be on the safe side, I
6055 have made the code more robust (and fixed the comments that describe what
6058 53. Now that there can be additional text after "Completed" in log lines (if
6059 the queue_time_overall log selector is set), a one-byte patch to exigrep
6060 was needed to allow it to recognize "Completed" as not the last thing in
6063 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
6064 patch that reportedly fixes this has been added. I am not expert enough to
6065 create a test for it. This is what the patch creator wrote:
6067 "I found a little strange behaviour of ldap code when working with
6068 Windows 2003 AD Domain, where users was placed in more than one
6069 Organization Units. When I tried to give exim partial DN, the exit code
6070 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
6071 But simultaneously result of request was absolutely normal ldap result,
6072 so I produce this patch..."
6074 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
6075 REFERENCE, so I have modified the code to exclude the patch when that macro
6078 55. Some experimental protocols are using DNS PTR records for new purposes. The
6079 keys for these records are domain names, not reversed IP addresses. The
6080 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
6081 leaves it alone. Component reversal etc. now happens only for IP addresses.
6084 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
6086 57. Double the size of the debug message buffer (to 2048) so that more of very
6087 long debug lines gets shown.
6089 58. The exicyclog utility now does better if the number of log files to keep
6090 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
6092 59. Two changes related to the smtp_active_hostname option:
6094 (1) $smtp_active_hostname is now available as a variable.
6095 (2) The default for smtp_banner uses $smtp_active_hostname instead
6096 of $primary_hostname.
6098 60. The host_aton() function is supposed to be passed a string that is known
6099 to be a valid IP address. However, in the case of IPv6 addresses, it was
6100 not checking this. This is a hostage to fortune. Exim now panics and dies
6101 if the condition is not met. A case was found where this could be provoked
6102 from a dnsdb PTR lookup with an IPv6 address that had more than 8
6103 components; fortuitously, this particular loophole had already been fixed
6104 by change 4.50/55 above.
6106 If there are any other similar loopholes, the new check in host_aton()
6107 itself should stop them being exploited. The report I received stated that
6108 data on the command line could provoke the exploit when Exim was running as
6109 exim, but did not say which command line option was involved. All I could
6110 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
6111 running as the user.
6114 61. There was a buffer overflow vulnerability in the SPA authentication code
6115 (which came originally from the Samba project). I have added a test to the
6116 spa_base64_to_bits() function which I hope fixes it.
6119 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
6120 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
6122 63. The daemon start-up calls getloadavg() while still root for those OS that
6123 need the first call to be done as root, but it missed one case: when
6124 deliver_queue_load_max is set with deliver_drop_privilege. This is
6125 necessary for the benefit of the queue runner, because there is no re-exec
6126 when deliver_drop_privilege is set.
6128 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
6129 This has been fixed.
6131 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
6132 was in use, was not putting the data itself into the right store pool;
6133 consequently, it could be overwritten for a subsequent message in the same
6134 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
6137 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
6139 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
6140 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
6141 the function were treating the return as a boolean value, which happened to
6142 work because 0=false and not-0=true, but is not correct code.
6144 68. The host_aton() function was not handling scoped IPv6 addresses (those
6145 with, for example, "%eth0" on the end) correctly.
6147 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
6148 negated items (that is, ~something) in unsigned ints. Some compilers
6149 apparently mutter when there is no cast.
6151 70. If an address verification called from an ACL failed, and did not produce a
6152 user-specific message (i.e. there was only a "system" message), nothing was
6153 put in $acl_verify_message. In this situation, it now puts the system
6156 71. Change 4.23/11 added synchronization checking at the start of an SMTP
6157 session; change 4.31/43 added the unwanted input to the log line - except
6158 that it did not do this in the start of session case. It now does.
6160 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
6161 This is wrong and can cause the other end to generate a synchronization
6162 error if it is another Exim or anything else that does the synchronization
6163 check. A QUIT command is no longer sent after a timeout.
6165 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
6166 during host lookups.
6168 74. The defer_ok option of callout verification was not working if it was used
6169 when verifying addresses in header lines, that is, for this case:
6171 verify = header_sender/callout=defer_ok
6173 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
6174 those file descriptors could be used for SMTP connections. If anything
6175 wrote to stderr (the example that came up was "warn" in embedded Perl), it
6176 could be sent to the SMTP client, causing chaos. The daemon now opens
6177 stdin, stdout, and stderr to /dev/null when it puts itself into the
6180 76. Arrange for output from Perl's "warn" command to be written to Exim's main
6181 log by default. The user can override this with suitable Perl magic.
6183 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
6184 the log message when discard triggers, was not working for the DATA ACL or
6185 for the non-SMTP ACL.
6187 78. Error message wording change in sieve.c.
6189 79. If smtp_accept_max_per_host was set, the number of connections could be
6190 restricted to fewer than expected, because the daemon was trying to set up
6191 a new connection before checking whether the processes handling previous
6192 connections had finished. The check for completed processes is now done
6193 earlier. On busy systems, this bug wouldn't be noticed because something
6194 else would have woken the daemon, and it would have reaped the completed
6197 80. If a message was submitted locally by a user whose login name contained one
6198 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
6199 It caused a spool format error. I have fixed the spool reading code. A
6200 related problem was that the "from" clause in the Received: line became
6201 illegal because of the space(s). It is now covered by ${quote_local_part.
6203 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
6206 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
6208 83. Patch from Sieve maintainer to fix unterminated string problem in
6209 "vacation" handling.
6211 84. Some minor changes to the Linux configuration files to help with other
6212 OS variants using glibc.
6214 85. One more patch for Sieve to update vacation handling to latest spec.
6217 ----------------------------------------------------
6218 See the note above about the 4.44 and 4.50 releases.
6219 ----------------------------------------------------
6225 1. Change 4.43/35 introduced a bug that caused file counts to be
6226 incorrectly computed when quota_filecount was set in an appendfile
6229 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
6230 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
6233 3. Give more explanation in the error message when the command for a transport
6234 filter fails to execute.
6236 4. There are several places where Exim runs a non-Exim command in a
6237 subprocess. The SIGUSR1 signal should be disabled for these processes. This
6238 was being done only for the command run by the queryprogram router. It is
6239 now done for all such subprocesses. The other cases are: ${run, transport
6240 filters, and the commands run by the lmtp and pipe transports.
6242 5. Some older OS have a limit of 256 on the maximum number of file
6243 descriptors. Exim was using setrlimit() to set 1000 as a large value
6244 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
6245 systems. I've change it so that if it can't get 1000, it tries for 256.
6247 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
6248 was an oversight, and furthermore, ever since the addition of extra
6249 controls (e.g. 4.43/32), the checks on when to allow different forms of
6250 "control" were broken. There should now be diagnostics for all cases when a
6251 control that does not make sense is encountered.
6253 7. $recipients is now available in the predata ACL (oversight).
6255 8. Tidy the search cache before the fork to do a delivery from a message
6256 received from the command line. Otherwise the child will trigger a lookup
6257 failure and thereby defer the delivery if it tries to use (for example) a
6258 cached ldap connection that the parent has called unbind on.
6260 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
6261 of $address_data from the recipient verification was clobbered by the
6262 sender verification.
6264 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
6265 was its contents. (It was OK if the option was not defined at all.)
6267 11. A "Completed" log line is now written for messages that are removed from
6268 the spool by the -Mrm option.
6270 12. $host_address is now set to the target address during the checking of
6271 ignore_target_hosts.
6273 13. When checking ignore_target_hosts for an ipliteral router, no host name was
6274 being passed; this would have caused $sender_host_name to have been used if
6275 matching the list had actually called for a host name (not very likely,
6276 since this list is usually IP addresses). A host name is now passed as
6279 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
6280 code that specifies a non-restarting handler (typically sigaction() in
6281 modern systems) in an attempt to fix a rare and obscure crash bug.
6283 15. Narrowed the window for a race in the daemon that could cause it to ignore
6284 SIGCHLD signals. This is not a major problem, because they are used only to
6285 wake it up if nothing else does.
6287 16. A malformed maildirsize file could cause Exim to calculate negative values
6288 for the mailbox size or file count. Odd effects could occur as a result.
6289 The maildirsize information is now recalculated if the size or filecount
6292 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
6293 support for a long time. Removed HAVE_SYS_VFS_H.
6295 18. Updated exipick to current release from John Jetmore.
6297 19. Allow an empty sender to be matched against a lookup in an address list.
6298 Previously the only cases considered were a regular expression, or an
6301 20. Exim went into a mad DNS lookup loop when doing a callout where the
6302 host was specified on the transport, if the DNS lookup yielded more than
6305 21. The RFC2047 encoding function was originally intended for short strings
6306 such as real names; it was not keeping to the 75-character limit for
6307 encoded words that the RFC imposes. It now respects the limit, and
6308 generates multiple encoded words if necessary. To be on the safe side, I
6309 have increased the buffer size for the ${rfc2047: expansion operator from
6312 22. Failure to deliver a bounce message always caused it to be frozen, even if
6313 there was an errors_to setting on the router. The errors_to setting is now
6316 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
6317 canonical form (fully expanded) before being placed in
6318 $sender_host_address.
6320 24. Updated eximstats to version 1.33
6322 25. Include certificate and key file names in error message when GnuTLS fails
6323 to set them up, because the GnuTLS error message doesn't include the name
6324 of the failing file when there is a problem reading it.
6326 26. Expand error message when OpenSSL has problems setting up cert/key files.
6329 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
6330 (this can affect the format of dates).
6332 28. exim_tidydb, when checking for the continued existence of a message for
6333 which it has found a message-specific retry record, was not finding
6334 messages that were in split spool directories. Consequently, it was
6335 deleting retry records that should have stayed in existence.
6337 29. eximstats updated to version 1.35
6338 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
6339 1.35 - bugfix such that pie charts by volume are generated correctly
6341 30. The SPA authentication driver was not abandoning authentication and moving
6342 on to the next authenticator when an expansion was forced to fail,
6343 contradicting the general specification for all authenticators. Instead it
6344 was generating a temporary error. It now behaves as specified.
6346 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
6347 (the order specifies the preference for clients). The order is now AES256,
6348 AES128, 3DES, ARCFOUR128.
6350 31. Small patch to Sieve code - explicitly set From: when generating an
6353 32. Exim crashed if a remote delivery caused a very long error message to be
6354 recorded - for instance if somebody sent an entire SpamAssassin report back
6355 as a large number of 550 error lines. This bug was coincidentally fixed by
6356 increasing the size of one of Exim's internal buffers (big_buffer) that
6357 happened as part of the Exiscan merge. However, to be on the safe side, I
6358 have made the code more robust (and fixed the comments that describe what
6361 33. Some experimental protocols are using DNS PTR records for new purposes. The
6362 keys for these records are domain names, not reversed IP addresses. The
6363 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
6364 leaves it alone. Component reversal etc. now happens only for IP addresses.
6367 34. The host_aton() function is supposed to be passed a string that is known
6368 to be a valid IP address. However, in the case of IPv6 addresses, it was
6369 not checking this. This is a hostage to fortune. Exim now panics and dies
6370 if the condition is not met. A case was found where this could be provoked
6371 from a dnsdb PTR lookup with an IPv6 address that had more than 8
6372 components; fortuitously, this particular loophole had already been fixed
6373 by change 4.50/55 or 4.44/33 above.
6375 If there are any other similar loopholes, the new check in host_aton()
6376 itself should stop them being exploited. The report I received stated that
6377 data on the command line could provoke the exploit when Exim was running as
6378 exim, but did not say which command line option was involved. All I could
6379 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
6380 running as the user.
6383 35. There was a buffer overflow vulnerability in the SPA authentication code
6384 (which came originally from the Samba project). I have added a test to the
6385 spa_base64_to_bits() function which I hope fixes it.
6388 36. The daemon start-up calls getloadavg() while still root for those OS that
6389 need the first call to be done as root, but it missed one case: when
6390 deliver_queue_load_max is set with deliver_drop_privilege. This is
6391 necessary for the benefit of the queue runner, because there is no re-exec
6392 when deliver_drop_privilege is set.
6394 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
6395 was in use, was not putting the data itself into the right store pool;
6396 consequently, it could be overwritten for a subsequent message in the same
6397 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
6400 38. Sometimes the final signoff response after QUIT could fail to get
6401 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
6402 < 0 before doing a fflush(). This bug looks as though it goes back to the
6403 introduction of TLS in release 3.20, but "sometimes" must have been rare
6404 because the tests only now provoked it.
6410 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
6411 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
6412 Now it returns an integer. A number of calls were still expecting a T/F
6413 return. Fortuitously, in all cases, the tests worked in OK situations,
6414 which is the norm. However, things would have gone wrong on any write
6415 failures on the smtp file descriptor. This function is used when sending
6416 messages over SMTP and also when doing verify callouts.
6418 2. When Exim is called to do synchronous delivery of a locally submitted
6419 message (the -odf or -odi options), it no longer closes stderr before doing
6422 3. Implemented the mua_wrapper option.
6424 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
6426 5. Implemented the functions header_remove(), header_testname(),
6427 header_add_at_position(), and receive_remove_recipient(), and exported them
6430 6. If an ACL "warn" statement specified the addition of headers, Exim already
6431 inserted X-ACL-Warn: at the start if there was no header name. However, it
6432 was not making this test for the second and subsequent header lines if
6433 there were newlines in the string. This meant that an invalid header could
6434 be inserted if Exim was badly configured.
6436 7. Allow an ACL "warn" statement to add header lines at the start or after all
6437 the Received: headers, as well as at the end.
6439 8. Added the rcpt_4xx retry error code.
6441 9. Added postmaster_mailfrom=xxx to callout verification option.
6443 10. Added mailfrom=xxxx to the callout verification option, for verify=
6446 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
6447 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
6449 12. Inserted some casts to stop certain compilers warning when using pointer
6450 differences as field lengths or precisions in printf-type calls (mostly
6451 affecting debugging statements).
6453 13. Added optional readline() support for -be (dynamically loaded).
6455 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
6456 same clock tick as a message's arrival, so that its received time was the
6457 same as the "first fail" time on the retry record, and that message
6458 remained on the queue past the ultimate address timeout, every queue runner
6459 would try a delivery (because it was past the ultimate address timeout) but
6460 after another failure, the ultimate address timeout, which should have then
6461 bounced the address, did not kick in. This was a "< instead of <=" error;
6462 in most cases the first failure would have been in the next clock tick
6463 after the received time, and all would be well.
6465 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
6466 being recognized when the domain list was tested by the match_domain
6467 condition in an expansion string.
6469 16. Added the ${str2b64: operator.
6471 17. Exim was always calling setrlimit() to set a large limit for the number of
6472 processes, without checking whether the existing limit was already
6473 adequate. (It did check for the limit on file descriptors.) Furthermore,
6474 errors from getrlimit() and setrlimit() were being ignored. Now they are
6475 logged to the main and panic logs, but Exim does carry on, to try to do its
6476 job under whatever limits there are.
6478 18. Imported PCRE 5.0.
6480 19. Trivial typo in log message " temporarily refused connection" (the leading
6483 20. If the log selector return_path_on_delivery was set and an address was
6484 redirected to /dev/null, the delivery process crashed because it assumed
6485 that a return path would always be set for a "successful" delivery. In this
6486 case, the whole delivery is bypassed as an optimization, and therefore no
6489 21. Internal re-arrangement: the function for sending a challenge and reading
6490 a response while authentication was assuming a zero-terminated challenge
6491 string. It's now changed to take a pointer and a length, to allow for
6492 binary data in such strings.
6494 22. Added the cyrus_sasl authenticator (code supplied by MBM).
6496 23. Exim was not respecting finduser_retries when seeking the login of the
6497 uid under which it was called; it was always trying 10 times. (The default
6498 setting of finduser_retries is zero.) Also, it was sleeping after the final
6499 failure, which is pointless.
6501 24. Implemented tls_on_connect_ports.
6503 25. Implemented acl_smtp_predata.
6505 26. If the domain in control=submission is set empty, Exim assumes that the
6506 authenticated id is a complete email address when it generates From: or
6507 Sender: header lines.
6509 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
6510 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
6511 chown and chgrp in /bin and hostname in /usr/bin.
6513 28. Exim was keeping the "process log" file open after each use, just as it
6514 does for the main log. This opens the possibility of it remaining open for
6515 long periods when the USR1 signal hits a daemon. Occasional processlog
6516 errors were reported, that could have been caused by this. Anyway, it seems
6517 much more sensible not to leave this file open at all, so that is what now
6520 29. The long-running daemon process does not normally write to the log once it
6521 has entered its main loop, and it closes the log before doing so. This is
6522 so that log files can straightforwardly be renamed and moved. However,
6523 there are a couple of unusual error situations where the daemon does write
6524 log entries, and I had neglected to close the log afterwards.
6526 30. The text of an SMTP error response that was received during a remote
6527 delivery was being truncated at 512 bytes. This is too short for some of
6528 the long messages that one sometimes sees. I've increased the limit to
6531 31. It is now possible to make retry rules that apply only when a message has a
6532 specific sender, in particular, an empty sender.
6534 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
6535 it possible to be selective about when SMTP synchronization is enforced.
6537 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
6539 32. Implemented hosts_connection_nolog.
6541 33. Added an ACL for QUIT.
6543 34. Setting "delay_warning=" to disable warnings was not working; it gave a
6546 35. Added mailbox_size and mailbox_filecount to appendfile.
6548 36. Added control = no_multiline_responses to ACLs.
6550 37. There was a bug in the logic of the code that waits for the clock to tick
6551 in the case where the clock went backwards by a substantial amount such
6552 that the microsecond fraction of "now" was more than the microsecond
6553 fraction of "then" (but the whole seconds number was less).
6555 38. Added support for the libradius Radius client library this is found on
6556 FreeBSD (previously only the radiusclient library was supported).
6562 1. When certain lookups returned multiple values in the form name=value, the
6563 quoting of the values was not always being done properly. Specifically:
6564 (a) If the value started with a double quote, but contained no whitespace,
6566 (b) If the value contained whitespace other than a space character (i.e.
6567 tabs or newlines or carriage returns) it was not quoted.
6568 This fix has been applied to the mysql and pgsql lookups by writing a
6569 separate quoting function and calling it from the lookup code. The fix
6570 should probably also be applied to nisplus, ibase and oracle lookups, but
6571 since I cannot test any of those, I have not disturbed their existing code.
6573 2. A hit in the callout cache for a specific address caused a log line with no
6574 reason for rejecting RCPT. Now it says "Previous (cached) callout
6575 verification failure".
6577 3. There was an off-by-one bug in the queryprogram router. An over-long
6578 return line was truncated at 256 instead of 255 characters, thereby
6579 overflowing its buffer with the terminating zero. As well as fixing this, I
6580 have increased the buffer size to 1024 (and made a note to document this).
6582 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
6583 when Exim is waiting for an SMTP response from a remote server, Exim
6584 restarts its select() call on the socket, thereby resetting its timeout.
6585 This is not a problem when such interrupts are rare. Somebody set up a cron
6586 job to run exiwhat every 2 minutes, which is less than the normal select()
6587 timeout (5 or 10 minutes). This meant that the select() timeout never
6588 kicked in because it was always reset. I have fixed this by comparing the
6589 time when an interrupt arrives with the time at the start of the first call
6590 to select(). If more time than the timeout has elapsed, the interrupt is
6591 treated as a timeout.
6593 5. Some internal re-factoring in preparation for the addition of Sieve
6594 extensions (by MH). In particular, the "personal" test is moved to a
6595 separate function, and given an option for scanning Cc: and Bcc: (which is
6596 not set for Exim filters).
6598 6. When Exim created an email address using the login of the caller as the
6599 local part (e.g. when creating a From: or Sender: header line), it was not
6600 quoting the local part when it contained special characters such as @.
6602 7. Installed new OpenBSD configuration files.
6604 8. Reworded some messages for syntax errors in "and" and "or" conditions to
6605 try to make them clearer.
6607 9. Callout options, other than the timeout value, were being ignored when
6608 verifying sender addresses in header lines. For example, when using
6610 verify = header_sender/callout=no_cache
6612 the cache was (incorrectly) being used.
6614 10. Added a missing instance of ${EXE} to the exim_install script; this affects
6615 only the Cygwin environment.
6617 11. When return_path_on_delivery was set as a log selector, if different remote
6618 addresses in the same message used different return paths and parallel
6619 remote delivery occurred, the wrong values would sometimes be logged.
6620 (Whenever a remote delivery process finished, the return path value from
6621 the most recently started remote delivery process was logged.)
6623 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
6624 lines when AUTH and/or TLS are in use. This is the "received protocol"
6625 field. Exim used to use "asmtp" for authenticated SMTP, without any
6626 indication (in the protocol name) for TLS use. Now it follows the RFC and
6627 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
6628 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
6629 names appear in log lines as well as in Received: header lines.
6631 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
6632 extensions, and comparison tests, and to fix some bugs.
6634 14. Changes to the "personal" filter test:
6636 (1) The test was buggy in that it was just doing the equivalent of
6637 "contains" tests on header lines. For example, if a user's address was
6638 anne@some.where, the "personal" test would incorrectly be true for
6640 To: susanne@some.where
6642 This test is now done by extracting each address from the header in turn,
6643 and checking the entire address. Other tests that are part of "personal"
6644 are now done using regular expressions (for example, to check local parts
6645 of addresses in From: header lines).
6647 (2) The list of non-personal local parts in From: addresses has been
6648 extended to include "listserv", "majordomo", "*-request", and "owner-*",
6649 taken from the Sieve specification recommendations.
6651 (3) If the message contains any header line starting with "List-" it is
6652 treated as non-personal.
6654 (4) The test for "circular" in the Subject: header line has been removed
6655 because it now seems ill-conceived.
6657 15. Minor typos in src/EDITME comments corrected.
6659 16. Installed latest exipick from John Jetmore.
6661 17. If headers_add on a router specified a text string that was too long for
6662 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
6663 of string_sprintf() is now avoided.
6665 18. $message_body_size was not set (it was always zero) when running the DATA
6666 ACL and the local_scan() function.
6668 19. For the "mail" command in an Exim filter, no default was being set for
6669 the once_repeat time, causing a random time value to be used if "once" was
6670 specified. (If the value happened to be <= 0, no repeat happened.) The
6671 default is now 0s, meaning "never repeat". The "vacation" command was OK
6672 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
6673 (I found it when inspecting the code).
6675 20. There is now an overall timeout for performing a callout verification. It
6676 defaults to 4 times the callout timeout, which applies to individual SMTP
6677 commands during the callout. The overall timeout applies when there is more
6678 than one host that can be tried. The timeout is checked before trying the
6679 next host. This prevents very long delays if there are a large number of
6680 hosts and all are timing out (e.g. when the network connections are timing
6681 out). The value of the overall timeout can be changed by specifying an
6682 additional sub-option for "callout", called "maxwait". For example:
6684 verify = sender/callout=5s,maxwait=20s
6686 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
6687 to the end before writing, but this should make it even safer).
6689 22. Exim was forgetting that it had advertised PIPELINING for the second and
6690 subsequent messages on an SMTP connection. It was also not resetting its
6691 memory on STARTTLS and an internal HELO.
6693 23. When Exim logs an SMTP synchronization error within a session, it now
6694 records whether PIPELINING has been advertised or not.
6696 24. Added 3 instances of "(long int)" casts to time_t variables that were being
6697 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
6698 rather than long int.
6700 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
6702 26. Added the never_mail option to autoreply.
6708 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
6709 crash if the getsockname() call failed; this can happen if a connection is
6710 closed very soon after it is established. The problem was simply in the
6711 order in which certain operations were done, causing Exim to try to write
6712 to the SMTP stream before it had set up the file descriptor. The bug has
6713 been fixed by making things happen in the correct order.
6719 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
6720 before the connection was closed, thus losing the rejection response.
6722 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
6723 some early Solaris releases, but causes trouble in current releases where
6724 socklen_t is defined.
6726 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
6729 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
6732 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
6733 that releasing the top of it at the end releases what was used for sub-
6734 expansions (unless the block got too big). However, discard this block if
6735 the first thing is a variable or header, so that we can use its block when
6736 it is dynamic (useful for very large $message_headers, for example).
6738 6. Lookups now cache *every* query, not just the most recent. A new, separate
6739 store pool is used for this. It can be recovered when all lookup caches are
6740 flushed. Lookups now release memory at the end of their result strings.
6741 This has involved some general refactoring of the lookup sources.
6743 7. Some code has been added to the store_xxx() functions to reduce the amount
6744 of flapping under certain conditions.
6746 8. log_incoming_interface used to affect only the <= reception log lines. Now
6747 it causes the local interface and port to be added to several more SMTP log
6748 lines, for example "SMTP connection from", and rejection lines.
6750 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
6752 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
6754 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
6755 could be overwritten at the end of the current message (or the start of a
6756 new message if it was set in a HELO ACL). The value is now preserved for
6757 the duration of the SMTP connection.
6759 12. If a transport had a headers_rewrite setting, and a matching header line
6760 contained an unqualified address, that address was qualified, even if it
6761 did not match any rewriting rules. The underlying bug was that the values
6762 of the flags that permit the existence of unqualified sender and recipient
6763 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
6764 non-local messages, and by -bnq for local messages) were not being
6765 preserved with the message after it was received.
6767 13. When Exim was logging an SMTP synchronization error, it could sometimes log
6768 "next input=" as part of the text comprising the host identity instead of
6769 the correct text. The code was using the same buffer for two different
6770 strings. However, depending on which order the printing function evaluated
6771 its arguments, the bug did not always show up. Under Linux, for example, my
6772 test suite worked just fine.
6774 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
6775 doesn't work with some older versions of Perl. It has been changed to "my",
6776 which in any case is probably the better facility to use.
6778 15. A really picky compiler found some instances of statements for creating
6779 error messages that either had too many or two few arguments for the format
6782 16. The size of the buffer for calls to the DNS resolver has been increased
6783 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
6784 for addresses that have a lot of PTR records. This alleviates a problem; it
6785 does not fully solve it.
6787 17. A dnsdb lookup for PTR records that receives more data than will fit in the
6788 buffer now truncates the list and logs the incident, which is the same
6789 action as happens when Exim is looking up a host name and its aliases.
6790 Previously in this situation something unpredictable would happen;
6791 sometimes it was "internal error: store_reset failed".
6793 18. If a server dropped the connection unexpectedly when an Exim client was
6794 using GnuTLS and trying to read a response, the client delivery process
6795 crashed while trying to generate an error log message.
6797 19. If a "warn" verb in an ACL added multiple headers to a message in a single
6798 string, for example:
6800 warn message = H1: something\nH2: something
6802 the text was added as a single header line from Exim's point of view
6803 though it ended up OK in the delivered message. However, searching for the
6804 second and subsequent header lines using $h_h2: did not work. This has been
6805 fixed. Similarly, if a system filter added multiple headers in this way,
6806 the routers could not see them.
6808 20. Expanded the error message when iplsearch is called with an invalid key to
6809 suggest using net-iplsearch in a host list.
6811 21. When running tests using -bh, any delays imposed by "delay" modifiers in
6812 ACLs are no longer actually imposed (and a message to that effect is
6815 22. If a "gecos" field in a passwd entry contained escaped characters, in
6816 particular, if it contained a \" sequence, Exim got it wrong when building
6817 a From: or a Sender: header from that name. A second bug also caused
6818 incorrect handling when an unquoted " was present following a character
6819 that needed quoting.
6821 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
6822 was not being matched caselessly.
6824 24. Arranged for all hyphens in the exim.8 source to be escaped with
6827 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
6828 properly. Recipient callout cache records were still being keyed to include
6829 the sender, even when use_sender was set false. This led to far more
6830 callouts that were necessary. The sender is no longer included in the key
6831 when use_sender is false.
6833 26. Added "control = submission" modifier to ACLs.
6835 27. Added the ${base62d: operator to decode base 62 numbers.
6837 28. dnsdb lookups can now access SRV records.
6839 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
6840 the configuration file.
6842 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
6843 (-v) mode. This makes the output for a verbose queue run more intelligible.
6845 31. Added a use_postmaster feature to recipient callouts.
6847 32. Added the $body_zerocount variable, containing the number of binary zero
6848 bytes in the message body.
6850 33. The time of last modification of the "new" subdirectory is now used as the
6851 "mailbox time last read" when there is a quota error for a maildir
6854 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
6856 35. Added +ignore_unknown as a special item in host lists.
6858 36. Code for decoding IPv6 addresses in host lists is now included, even if
6859 IPv6 support is not being compiled. This fixes a bug in which an IPv6
6860 address was recognized as an IP address, but was then not correctly decoded
6861 into binary, causing unexpected and incorrect effects when compared with
6868 1. Very minor rewording of debugging text in manualroute to say "list of
6869 hosts" instead of "hostlist".
6871 2. If verify=header_syntax was set, and a header line with an unqualified
6872 address (no domain) and a large number of spaces between the end of the
6873 name and the colon was received, the reception process suffered a buffer
6874 overflow, and (when I tested it) crashed. This was caused by some obsolete
6875 code that should have been removed. The fix is to remove it!
6877 3. When running in the test harness, delay a bit after writing a bounce
6878 message to get a bit more predictability in the log output.
6880 4. Added a call to search_tidyup() just before forking a reception process. In
6881 theory, someone could use a lookup in the expansion of smtp_accept_max_
6882 per_host which, without the tidyup, could leave open a database connection.
6884 5. Added the variables $recipient_data and $sender_data which get set from a
6885 lookup success in an ACL "recipients" or "senders" condition, or a router
6886 "senders" option, similar to $domain_data and $local_part_data.
6888 6. Moved the writing of debug_print from before to after the "senders" test
6891 7. Change 4.31/66 (moving the time when the Received: is generated) caused
6892 problems for message scanning, either using a data ACL, or using
6893 local_scan() because the Received: header was not generated till after they
6894 were called (in order to set the time as the time of reception completion).
6895 I have revised the way this works. The header is now generated after the
6896 body is received, but before the ACL or local_scan() are called. After they
6897 are run, the timestamp in the header is updated.
6903 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
6904 before starting a queue runner without re-exec. This happened only when
6905 deliver_drop_privilege was set or when the Exim user was set to root. The
6906 effect of the bug was that timeouts during subsequent deliveries caused
6907 crashes instead of being properly handled. The handler is now left at its
6908 default (and expected) setting.
6910 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
6911 message, again when deliver_drop_privilege is set or Exim is run as root.
6912 The bug described in (1) was not present in this case, but the tidying up
6913 of the other signals was missing. I have made the two cases consistent.
6915 3. The ignore_target_hosts setting on a manualroute router was being ignored
6916 for hosts that were looked up using the /MX notation.
6918 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
6921 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
6922 operated on the sender address. After changing the $sender_address to <>
6923 for the sender address verify, Exim was re-instated it as the original
6924 (before rewriting) address, but remembering that it had rewritten it, so it
6925 wasn't rewriting it again. This bug also had the effect of breaking the
6926 sender address verification caching when the sender address was rewritten.
6928 6. The ignore_target_hosts option was being ignored by the ipliteral router.
6929 This has been changed so that if the ip literal address matches
6930 ignore_target_hosts, the router declines.
6932 7. Added expansion conditions match_domain, match_address, and match_local_
6933 part (NOT match_host).
6935 8. The placeholder for the Received: header didn't have a length field set.
6937 9. Added code to Exim itself and to exim_lock to test for a specific race
6938 condition that could lead to file corruption when using MBX delivery. The
6939 issue is with the lockfile that is created in /tmp. If this file is removed
6940 after a process has opened it but before that process has acquired a lock,
6941 there is the potential for a second process to recreate the file and also
6942 acquire a lock. This could lead to two Exim processes writing to the file
6943 at the same time. The added code performs the same test as UW imapd; it
6944 checks after acquiring the lock that its file descriptor still refers to
6945 the same named file.
6947 10. The buffer for building added header lines was of fixed size, 8192 bytes.
6948 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
6951 11. Added the smtp_active_hostname option. If used, this will typically be made
6952 to depend on the incoming interface address. Because $interface_address is
6953 not set up until the daemon has forked a reception process, error responses
6954 that can happen earlier (such as "too many connections") no longer contain
6957 12. If an expansion in a condition on a "warn" statement fails because a lookup
6958 defers, the "warn" statement is abandoned, and the next ACL statement is
6959 processed. Previously this caused the whole ACL to be aborted.
6961 13. Added the iplsearch lookup type.
6963 14. Added ident_timeout as a log selector.
6965 15. Added tls_certificate_verified as a log selector.
6967 16. Added a global option tls_require_ciphers (compare the smtp transport
6968 option of the same name). This controls incoming TLS connections.
6970 17. I finally figured out how to make tls_require_ciphers do a similar thing
6971 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
6972 before starting the TLS session.
6974 18. Tabs are now shown as \t in -bP output.
6976 19. If the log selector return_path_on_delivery was set, Exim crashed when
6977 bouncing a message because it had too many Received: header lines.
6979 20. If two routers both had headers_remove settings, and the first one included
6980 a superfluous trailing colon, the final name in the first list and the
6981 first name in the second list were incorrectly joined into one item (with a
6982 colon in the middle).
6988 1. Added -C and -D options to the exinext utility, mainly to make it easier
6989 to include in the automated testing, but these could be helpful when
6990 multiple configurations are in use.
6992 2. The exinext utility was not formatting the output nicely when there was
6993 an alternate port involved in the retry record key, nor when there was a
6994 message id as well (for retries that were specific to a specific message
6995 and a specific host). It was also confused by IPv6 addresses, because of
6996 the additional colons they contain. I have fixed the IPv4 problem, and
6997 patched it up to do a reasonable job for IPv6.
6999 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
7000 delivery, the log line now contains "pipelined" if PIPELINING was used.
7002 4. An SMTP transport process used to panic and die if the bind() call to set
7003 an explicit outgoing interface failed. This has been changed; it is now
7004 treated in the same way as a connect() failure.
7006 5. A reference to $sender_host_name in the part of a conditional expansion
7007 that was being skipped was still causing a DNS lookup. This no longer
7010 6. The def: expansion condition was not recognizing references to header lines
7011 that used bh_ and bheader_.
7013 7. Added the _cache feature to named lists.
7015 8. The code for checking quota_filecount in the appendfile transport was
7016 allowing one more file than it should have been.
7018 9. For compatibility with Sendmail, the command line option
7026 and sets the incoming protocol and host name (for trusted callers). The
7027 host name and its colon can be omitted when only the protocol is to be set.
7028 Note the Exim already has two private options, -pd and -ps, that refer to
7029 embedded Perl. It is therefore impossible to set a protocol value of "d" or
7030 "s", but I don't think that's a major issue.
7032 10. A number of refactoring changes to the code, none of which should affect
7035 (a) The number of logging options was getting close to filling up the
7036 32-bit word that was used as a bit map. I have split them into two classes:
7037 those that are passed in the argument to log_write(), and those that are
7038 only ever tested independently outside of that function. These are now in
7039 separate 32-bit words, so there is plenty of room for expansion again.
7040 There is no change in the user interface or the logging behaviour.
7042 (b) When building, for example, log lines, the code previously used a
7043 macro that called string_cat() twice, in order to add two strings. This is
7044 not really sufficiently general. Furthermore, there was one instance where
7045 it was actually wrong because one of the argument was used twice, and in
7046 one call a function was used. (As it happened, calling the function twice
7047 did not affect the overall behaviour.) The macro has been replaced by a
7048 function that can join an arbitrary number of extra strings onto a growing
7051 (c) The code for expansion conditions now uses a table and a binary chop
7052 instead of a serial search (which was left over from when there were very
7053 few conditions). Also, it now recognizes conditions like "pam" even when
7054 the relevant support is not compiled in: a suitably worded error message is
7055 given if an attempt is made to use such a condition.
7057 11. Added ${time_interval:xxxxx}.
7059 12. A bug was causing one of the ddress fields not to be passed back correctly
7060 from remote delivery subprocesses. The field in question was not being
7061 subsequently used, so this caused to problems in practice.
7063 13. Added new log selectors queue_time and deliver_time.
7065 14. Might have fixed a bug in maildirsizefile handling that threw up
7066 "unexpected character" debug warnings, and recalculated the data
7067 unnecessarily. In any case, I expanded the warning message to give more
7070 15. Added the message "Restricted characters in address" to the statements in
7071 the default ACL that block characters like @ and % in local parts.
7073 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
7074 Three changes have been made:
7076 (a) There was a serious bug; a negative response to MAIL caused the whole
7077 recipient domain to be cached as invalid, thereby blocking all messages
7078 to all local parts at the same domain, from all senders. This bug has
7079 been fixed. The domain is no longer cached after a negative response to
7080 MAIL if the sender used is not empty.
7082 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
7085 (c) A new callout option, "use_sender" has been added for people who want
7086 the modified behaviour.
7092 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
7095 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
7096 indeed breaks things for older releases.
7098 3. Added additional logging to the case where there is a problem reading data
7099 from a filter that is running in a subprocess using a pipe, in order to
7100 try to track down a specific problem.
7102 4. Testing facility fudge: when running in the test harness and attempting
7103 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
7104 getting "No route to host". Convert this to a timeout.
7106 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
7109 6. Some OS don't have socklen_t but use size_t instead. This affects the
7110 fifth argument of getsockopt() amongst other things. This is now
7111 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
7112 can be set for individual OS. I have set it for SunOS5, OSF1, and
7113 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
7114 some earlier ones do not.
7116 7. Change 4.30/15 was not doing the test caselessly.
7118 8. The standard form for an IPv6 address literal was being rejected by address
7119 parsing in, for example, MAIL and RCPT commands. An example of this kind of
7120 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
7121 this, as well as the form without the "IPv6" on the front (but only when
7122 address literals are enabled, of course).
7124 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
7126 10. Exim crashed if a message with an empty sender address specified by -f
7127 encountered a router with an errors_to setting. This could be provoked only
7128 by a command such as
7132 where an empty string was supplied; "<>" did not hit this bug.
7134 11. Installed PCRE release 4.5.
7136 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
7137 remained set. It is now erased.
7139 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
7140 times from message ids (which are base 36 rather than the normal 62).
7142 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
7143 were being counted as actual protocol errors, and logged if the log
7144 selector +smtp_protocol_error was set. One cannot be perfect in this test,
7145 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
7146 and DATA following a set of rejected RCPTs do not count as protocol errors.
7147 In other words, Exim assumes they were pipelined, though this may not
7148 actually be the case. Of course, in all cases the client gets an
7149 appropriate error code.
7151 15. If a lookup fails in an ACL condition, a message about the failure may
7152 be available; it is used if testing the ACL cannot continue, because most
7153 such messages specify what the cause of the deferral is. However, some
7154 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
7155 that caused an old message to be retained and used if a later statement
7156 caused a defer, replacing the real cause of the deferral.
7158 16. If an IP address had so many PTR records that the DNS lookup buffer
7159 was not large enough to hold them, Exim could crash while trying to process
7160 the truncated data. It now detects and logs this case.
7162 17. Further to 4.21/58, another change has been made: if (and only if) the
7163 first line of a message (the first header line) ends with CRLF, a bare LF
7164 in a subsequent header line has a space inserted after it, so as not to
7165 terminate the header.
7167 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
7168 unnecessarily, used atoi() instead of strtol(), and didn't check the
7169 termination when getting file sizes from file names by regex.
7171 19. Completely re-implemented the support for maildirsize files, in the light
7172 of a number of problems with the previous contributed implementation
7173 (4.30/29). In particular:
7175 . If the quota is zero, the maildirsize file is maintained, but no quota is
7178 . If the maildir directory does not exist, it is created before any attempt
7179 to write a maildirsize file.
7181 . The quota value in the file is just a cache; if the quota is changed in
7182 the transport, the new value overrides.
7184 . A regular expression is available for excluding directories from the
7187 20. The autoreply transport checks the characters in options that define the
7188 message's headers; it allows continued headers, but it was checking with
7189 isspace() after an embedded newline instead of explicitly looking for a
7192 21. If all the "regular" hosts to which an address was routed had passed their
7193 expiry times, and had not reached their retry times, the address was
7194 bounced, even if fallback hosts were defined. Now Exim should go on to try
7197 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
7198 equivalent code in the SMTP transport. Some hosts send humungous responses
7199 to HELO/EHLO, more than 1024 it seems.
7201 23. Refactoring: code in filter.c used (void *) for "any old type" but this
7202 gives compiler warnings in some environments. I've now done it "properly",
7205 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
7206 (because of problems with the built-in one) was declared to return uschar *
7207 instead of char *, causing compiler failure.
7209 25. Fixed a file descriptor leak when processing alias/forward files.
7211 26. Fixed a minor format string issue in dbfn.c.
7213 27. Typo in exim.c: ("dmbnz" for "dbmnz").
7215 28. If a filter file refered to $h_xxx or $message_headers, and the headers
7216 contained RFC 2047 "words", Exim's memory could, under certain conditions,
7219 29. When a sender address is verified, it is cached, to save repeating the test
7220 when there is more than one recipient in a message. However, when the
7221 verification involves a callout, it is possible for different callout
7222 options to be set for different recipients. It is too complicated to keep
7223 track of this in the cache, so now Exim always runs a verification when a
7224 callout is required, relying on the callout cache for the optimization.
7225 The overhead is duplication of the address routing, but this should not be
7228 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
7229 to be verified with callout=postmaster, and the main callout worked but the
7230 postmaster check failed, the verification correctly failed. However, if a
7231 subsequent RCPT command asked for sender verification *without* the
7232 postmaster check, incorrect caching caused this verification also to fail,
7235 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
7236 it was not caching the DNS options (qualify_single, search_parents) that
7237 were used when the lookup failed. A subsequent lookup with different
7238 options therefore always gave the same answer, though there were cases
7239 where it should not have. (Example: a "domains = !$mx_any" option on a
7240 dnslookup router: the "domains" option is always processed without any
7241 widening, but the router might have qualify_single set.) Now Exim uses the
7242 cached value only when the same options are set.
7244 32. Added John Jetmore's "exipick" utility to the distribution.
7246 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
7247 than a timeout (e.g. a certificate is required, and is not provided), an
7248 Exim server now closes the connection immediately. Previously it waited for
7249 the client to close - but if the client is SSL, it seems that they each
7250 wait for each other, leading to a delay before one of them times out.
7252 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
7253 maintained 0.8.x compatibility because I don't think many are using it, and
7254 it is clearly obsolete.
7256 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
7259 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
7260 client certificate was expired. A simple patch fixes this, though I don't
7261 understand the full logic of why the verify callback is called multiple
7264 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
7265 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
7266 which causes problems with some clients (such as the Certicom SSL Plus
7267 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
7268 disables the coutermeasure allowing Eudora to connect."
7270 38. Exim was not checking that a write() to a log file succeeded. This could
7271 lead to Bad Things if a log got too big, in particular if it hit a file
7272 size limit. Exim now panics and dies if it cannot write to a log file, just
7273 as it does if it cannot open a log file.
7275 39. Modified OS/Makefile-Linux so that it now contains
7277 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
7279 The two -D definitions ensure that Exim is compiled with large file
7280 support, which makes it possible to handle log files that are bigger than
7283 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
7284 instance) a domain was checked against a named list that involved a lookup,
7285 causing $domain_data to be set, then another domain was checked against the
7286 same list, then the first domain was re-checked, the value of $domain_data
7287 after the final check could be wrong. In particular, if the second check
7288 failed, it could be set empty. This bug probably also applied to
7291 41. The strip_trailing_dot option was not being applied to the address given
7292 with the -f command-line option.
7294 42. The code for reading a message's header from the spool was incrementing
7295 $received_count, but never initializing it. This meant that the value was
7296 incorrect (doubled) while delivering a message in the same process in which
7297 it was received. In the most common configuration of Exim, this never
7298 happens - a fresh exec is done - but it can happen when
7299 deliver_drop_privilege is set.
7301 43. When Exim logs an SMTP synchronization error - client data sent too soon -
7302 it now includes up to 150 characters of the unexpected data in the log
7305 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
7306 and building data strings. The size of both of these buffers was 10 000
7307 bytes - far larger than anybody would *ever* want, thought I. Needless to
7308 say, somebody hit the limit. I have increased the maximum line length to
7309 20 000 and the maximum data length of concatenated lines to 100 000. I have
7310 also fixed two bugs, because there was no checking on these buffers. Tsk,
7311 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
7312 buffer is too small.
7314 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
7315 lsearch lookups. Now it does.
7317 46. When parsing a route_list item in a manualroute router, a fixed-length
7318 buffer was used for the list of hosts. I made this 1024 bytes long,
7319 thinking that nobody would ever have a list of hosts that long. Wrong.
7320 Somebody had a whole pile of complicated expansion conditions, and the
7321 string was silently truncated, leading to an expansion error. It turns out
7322 that it is easier to change to an unlimited length (owing to other changes
7323 that have happened since this code was originally written) than to build
7324 structure for giving a limitation error. The length of the item that
7325 expands into the list of hosts is now unlimited.
7327 47. The lsearch lookup could not handle data where the length of text line was
7328 more than 4095 characters. Such lines were truncated, leading to shortened
7329 data being returned. It should now handle lines of any length.
7331 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
7332 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
7335 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
7336 replacing strings such as BIN_DIRECTORY in a source file: the replacement
7337 no longer happens in comment lines. A list of replacements is now placed
7338 at the head of all of the source files, except those whose only change is
7339 to replace PERL_COMMAND in the very first #! line.
7341 50. Replaced the slow insertion sort in queue.c, for sorting the list of
7342 messages on the queue, with a bottom-up merge sort, using code contributed
7343 by Michael Haardt. This should make operations like -bp somewhat faster on
7344 large queues. It won't affect queue runners, except when queue_run_in_order
7347 51. Installed eximstats 1.31 in the distribution.
7349 52. Added support for SRV lookups to the dnslookup router.
7351 53. If an ACL referred to $message_body or $message_body_end, the value was not
7352 reset for any messages that followed in the same SMTP session.
7354 54. The store-handling optimization for building very long strings was not
7355 differentiating between the different store pools. I don't think this
7356 actually made any difference in practice, but I've tidied it.
7358 55. While running the routers to verify a sender address, $sender_address
7359 was still set to the sender address. This is wrong, because when routing to
7360 send a bounce to the sender, it would be empty. Therefore, I have changed
7361 it so that, while verifying a sender address, $sender_address is set to <>.
7362 (There is no change to what happens when verifying a recipient address.)
7364 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
7365 target A or AAAA records (if not already returned) without resetting the
7366 qualify_single or search_parents options of the DNS resolver. These are
7367 inappropriate in this case because the targets of MX and SRV records must
7368 be FQDNs. A broken DNS record could cause trouble if it happened to have a
7369 target that, when qualified, matched something in the local domain. These
7370 two options are now turned off when doing these lookups.
7372 57. It seems that at least some releases of Reiserfs (which does not have the
7373 concept of a fixed number of inodes) returns zero and not -1 for the
7374 number of available inodes. This interacted badly with check_spool_inodes,
7375 which assumed that -1 was the "no such thing" setting. What I have done is
7376 to check that the total number of inodes is greater than zero before doing
7377 the test of how many are available.
7379 58. When a "warn" ACL statement has a log_message modifier, the message is
7380 remembered, and not repeated. This is to avoid a lot of repetition when a
7381 message has many recipients that cause the same warning to be written.
7382 However, Exim was preserving the list of already written lines for an
7383 entire SMTP session, which doesn't seem right. The memory is now reset if a
7384 new message is started.
7386 59. The "rewrite" debugging flag was not showing the result of rewriting in the
7387 debugging output unless log_rewrite was also set.
7389 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
7390 of (int)(handle) when we know that handle contains (void *)(-1).
7392 61. The Exim daemon panic-logs an error return when it closes the incoming
7393 connection. However "connection reset by peer" seems to be common, and
7394 isn't really an error worthy of noting specially, so that particular error
7397 62. When Exim is trying to find all the local interfaces, it used to panic and
7398 die if the ioctl to get the interface flags failed. However, it seems that
7399 on at least one OS (Solaris 9) it is possible to have an interface that is
7400 included in the list of interfaces, but for which you get a failure error
7401 for this call. This happens when the interface is not "plumbed" into a
7402 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
7403 failure of the "get flags" call assumes that the interface is down.
7405 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
7406 makes life easier for people who are doing arithmetic on fields extracted
7407 from dates, where you often get leading zeros that should not be
7408 interpreted as octal.
7410 64. Added qualify_domain to the redirect router, to override the global
7413 65. If a pathologically long header line contained very many addresses (the
7414 report of this problem mentioned 10 000) and each of them was rewritten,
7415 Exim could use up a very large amount of memory. (It kept on making new
7416 copies of the header line as it rewrote, and never released the old ones.)
7417 At the expense of a bit more processing, the header rewriting function has
7418 been changed so that it no longer eats memory in this way.
7420 66. The generation of the Received: header has been moved from the time that a
7421 message starts to be received, to the time that it finishes. The timestamp
7422 in the Received: header should now be very close to that of the <= log
7423 line. There are two side-effects of this change:
7425 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
7426 logged header lines no longer include the local Received: line, because
7427 it has not yet been created. The same applies to a copy of the message
7428 that is returned to a non-SMTP sender when a message is rejected.
7430 (b) When a filter file is tested using -bf, no additional Received: header
7431 is added to the test message. After some thought, I decided that this
7434 This change does not affect the value of $received_for. It is still set
7435 after address rewriting, but before local_scan() is called.
7437 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
7439 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
7440 gave an unhelpful panic error message, and a defer error. I have managed to
7441 change this behaviour so that it now rejects any supplied certificate,
7442 which seems right, as the list of acceptable certificates is empty.
7444 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
7445 gave an unhelpful defer error. I have not managed to make this reject any
7446 supplied certificates, but the error message it gives is "no certificate
7447 supplied", which is not helpful.
7449 70. exigrep's output now also includes lines that are not associated with any
7450 message, but which match the given pattern. Implemented by a patch from
7451 Martin Sluka, which also tidied up the Perl a bit.
7453 71. Recipient callout verification, like sender verification, was using <> in
7454 the MAIL FROM command. This isn't really the right thing, since the actual
7455 sender may affect whether the remote host accepts the recipient or not. I
7456 have changed it to use the actual sender in the callout; this means that
7457 the cache record is now keyed on a recipient/sender pair, not just the
7458 recipient address. There doesn't seem to be a real danger of callout loops,
7459 since a callout by the remote host to check the sender would use <>.
7460 [SEE ABOVE: changed after hitting problems.]
7462 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
7463 temporary errors. However, in the case of such a code being given after
7464 the end of a data transmission (i.e. after ".") Exim was failing to write
7465 a retry record for the message. (Yes, there was some broken host that was
7466 actually sending 8xx at this point.)
7468 73. An unknown lookup type in a host list could cause Exim to panic-die when
7469 the list was checked. (An example that provoked this was putting <; in the
7470 middle of a list instead of at the start.) If this happened during a DATA
7471 ACL check, a -D file could be left lying around. This kind of configuration
7472 error no longer causes Exim to die; instead it causes a defer error. The
7473 incident is still logged to the main and panic logs.
7475 74. Buglet left over from Exim 3 conversion. The message "too many messages
7476 in one connection" was written to the rejectlog but not the mainlog, except
7477 when address rewriting (yes!) was being logged.
7479 75. Added write_rejectlog option.
7481 76. When a system filter was run not as root (that is, when system_filter_user
7482 was set), the values of the $n variables were not being returned to the
7483 main process; thus, they were not subsequently available in the $sn
7486 77. Added +return_path_on_delivery log selector.
7488 78. A connection timeout was being treated differently from recipients deferred
7489 when testing hosts_max_try with a message that was older than the host's
7490 retry timeout. (The host should not be counted, thus allowing all hosts to
7491 be tried at least once before bouncing.) This may have been the cause of an
7492 occasionally reported bug whereby a message would remain on the queue
7493 longer than the retry timeout, but would be bounced if a delivery was
7494 forced. I say "may" because I never totally pinned down the problem;
7495 setting up timeout/retry tests is difficult. See also the next item.
7497 79. The ultimate address timeout was not being applied to errors that involved
7498 a combination of host plus message (for example, a timeout on a MAIL
7499 command). When an address resolved to a number of possible hosts, and they
7500 were not all tried for each delivery (e.g. because of hosts_max_try), a
7501 message could remain on the queue longer than the retry timeout.
7503 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
7506 81. Fixed an obscure SMTP outgoing bug which required at least the following
7507 conditions: (a) there was another message waiting for the same server;
7508 (b) the server returned 5xx to all RCPT commands in the first message so
7509 that the message was not completed; (c) the server dropped the connection
7510 or gave a negative response to the RSET that Exim sends to abort the
7511 transaction. The observed case was a dropped connection after DATA that had
7512 been sent in pipelining mode. That is, the server had advertised PIPELINING
7513 but was not implementing it correctly. The effect of the bug was incorrect
7514 behaviour, such as trying another host, and this could lead to a crash.
7520 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
7521 and daemon.c were passed as pointers to ints; they should have been
7522 pointers to socklen_t variables (which are typically unsigned ints).
7524 2. Some signed/unsigned type warnings in the os.c file for Linux have been
7527 3. Fixed a really odd bug that affected only the testing scheme; patching a
7528 certain fixed string in the binary changed the value of another string that
7529 happened to be identical to the end of the original first string.
7531 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
7532 name", it returns that address as the IP address. On some operating
7533 systems (e.g. Solaris), it also passes back the IP address string as the
7534 "host name". However, on others (e.g. Linux), it passes back an empty
7535 string. Exim wasn't checking for this, and was changing the host name to an
7536 empty string, assuming it had been canonicalized.
7538 5. Although rare, it is permitted to have more than one PTR record for a given
7539 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
7540 all the names associated with an address, because they do in Solaris.
7541 However, it seems that they do not in Linux for data that comes from the
7542 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
7543 I found this out when I moved to a new Linux workstation and tried to run
7544 the Exim test suite.
7546 To get round this problem I have changed the code so that it now does its
7547 own call to the DNS to look up PTR records when searching for a host name.
7548 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
7549 addresses that are only in /etc/hosts are still found.
7551 This behaviour is, however, controlled by an option called host_lookup_
7552 order, which defaults to "bydns:byaddr". If people want to use the other
7553 order, or indeed, just use one or the other means of lookup, they can
7554 specify it in this variable.
7556 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
7557 some operating systems, this comes back from gethostbyaddr() as an empty
7558 string, and this is what Exim used to test for. However, it seems that in
7559 other systems, "." is yielded. Exim now tests for this case too.
7561 7. The values of check_spool_space and check_log_space are now held internally
7562 as a number of kilobytes instead of an absolute number of bytes. If a
7563 numbers is specified without 'K' or 'M', it is rounded up to the nearest
7564 kilobyte. This means that much larger values can be stored.
7566 8. Exim monitor: an attempt to get the action menu when not actually pointing
7567 at a message produces an empty menu entitled "No message selected". This
7568 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
7569 no entries in it ("Shell widget menu has zero width and/or height"). So I
7570 have added a single, blank menu entry in this case.
7572 9. Added ${quote_local_part.
7574 10. MIME decoding is now applied to the contents of Subject: header lines when
7577 11. Now that a reference to $sender_host_address automatically causes a reverse
7578 lookup to occur if necessary (4.13/18), there is no need to arrange for a
7579 host lookup before query-style lookups in lists that might use this
7580 variable. This has therefore been abolished, and the "net-" prefix is no
7581 longer necessary for query-style lookups.
7583 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
7584 have been a typo for LFLAGS, so it has been changed.
7586 13. The install script calls Exim with "-C /dev/null" in order to find the
7587 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
7588 to be output. However, since Exim outputs its version number before the
7589 error, it didn't break the script. It just looked ugly. I fixed this by
7590 always allowing "-C /dev/null" if the caller is root.
7592 14. Ignore overlarge ACL variable number when reading spool file - insurance
7593 against a later release with more variables having written the file.
7595 15. The standard form for an IPv6 address literal was being rejected by EHLO.
7596 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
7597 this, as well as the form without the "IPv6" on the front.
7599 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
7600 OS/Makefile-Darwin file.
7602 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
7603 only to LDAP libraries that do not have LDAP_OPT_DEREF.
7605 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
7606 variable. However, some OS use ints for inodes. I've added cast to long int
7607 to get rid of the compiler warning.
7609 19. I had forgotten to lock out "/../" in configuration file names when
7610 ALT_CONFIG_PREFIX was set.
7612 20. Routers used for verification do not need to specify transports. However,
7613 if such a router generated a host list, and callout was configured, Exim
7614 crashed, because it could not find a port number from the (non-existent)
7615 transport. It now assumes port 25 in this circumstance.
7617 21. Added the -t option to exigrep.
7619 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
7620 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
7621 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
7622 them is set, LOOKUP_LSEARCH is forced.
7624 23. "exim -bV" now outputs a list of lookups that are included in the binary.
7626 24. Added sender and host information to the "rejected by local_scan()" log
7627 line; previously there was no indication of these.
7629 25. Added .include_if_exists.
7631 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
7632 Linux. It turns out that not all file systems support this. Apparently some
7633 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
7634 do it.) To cope with this, the error EINVAL, which means that sync-ing is
7635 not supported on the file descriptor, is now ignored when Exim is trying to
7636 sync a directory. This applies only to Linux.
7638 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
7640 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
7641 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
7644 29. Added support for maildirsize files from supplied patch (modified a bit).
7646 30. The use of :fail: followed by an empty string could lead Exim to respond to
7647 sender verification failures with (e.g.):
7649 550 Verification failed for <xxx>
7650 550 Sender verify failed
7652 where the first response line was missing the '-' that indicates it is not
7653 the final line of the response.
7655 31. The loop for finding the name of the user that called Exim had a hardwired
7656 limit of 10; it now uses the value of finduser_retries, which is used for
7657 all other user lookups.
7659 32. Added $received_count variable, available in data and not_smtp ACLs, and at
7662 33. Exim was neglecting to zero errno before one call of strtol() when
7663 expanding a string and expecting an integer value. On some systems this
7664 resulted in spurious "integer overflow" errors. Also, it was casting the
7665 result into an int without checking.
7667 34. Testing for a connection timeout using "timeout_connect" in the retry rules
7668 did not work. The code looks as if it has *never* worked, though it appears
7669 to have been documented since at least release 1.62. I have made it work.
7671 35. The "timeout_DNS" error in retry rules, also documented since at least
7672 1.62, also never worked. As it isn't clear exactly what this means, and
7673 clearly it isn't a major issue, I have abolished the feature by treating it
7674 as "timeout", and writing a warning to the main and panic logs.
7676 36. The display of retry rules for -brt wasn't always showing the error code
7679 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
7680 timeout_connect_A, timeout_connect_MX.
7682 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
7683 to the empty sender.
7685 39. The daemon was not analysing the content of -oX till after it had closed
7686 stderr and disconnected from the controlling terminal. This meant that any
7687 syntax errors were only noted on the panic log, and the return code from
7688 the command was 0. By re-arranging the code a little, I've made the
7689 decoding happen first, so such errors now appear on stderr, and the return
7690 code is 1. However, the actual setting up of the sockets still happens in
7691 the disconnected process, so errors there are still only recorded on the
7694 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
7695 connections (as happens on some IP stacks) was logged at start up time as
7696 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
7697 it from "IPv6 and IPv4", which means that two separate sockets are being
7700 41. The debug output for gethostbyname2() or getipnodebyname() failures now
7701 says whether AF_INET or AF_INET6 was passed as an argument.
7703 42. Exiwhat output was messed up when time zones were included in log
7706 43. Exiwhat now gives more information about the daemon's listening ports,
7707 and whether -tls-on-connect was used.
7709 44. The "port" option of the smtp transport is now expanded.
7711 45. A "message" modifier in a "warn" statement in a non-message ACL was being
7712 silently ignored. Now an error message is written to the main and panic
7715 46. There's a new ACL modifier called "logwrite" which writes to a log file
7716 as soon as it is encountered.
7718 47. Added $local_user_uid and $local_user_gid at routing time.
7720 48. Exim crashed when trying to verify a sender address that was being
7723 49. Exim was recognizing only a space character after ".include". It now also
7724 recognizes a tab character.
7726 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
7727 extracting the relevant information from the specification. The man page no
7728 longer contains scrambled data for the -d option, and I've added a section
7729 at the front about calling Exim under different names.
7731 51. Added "extra_headers" argument to the "mail" command in filter files.
7733 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
7736 53. Installed eximstats 1.29.
7738 54. Added transport_filter_timeout as a generic transport option.
7740 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
7741 Cc: header lines. This was required by RFC 822, but it not required by RFC
7744 56. Exim used to add From:, Date:, and Message-Id: header lines to any
7745 incoming messages that did not have them. Now it does so only if the
7746 message originates locally, that is, if there is no associated remote host
7747 address. When Resent- header lines are present, this applies to the Resent-
7748 lines rather than the non-Resent- lines.
7750 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
7751 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
7753 58. Messages for configuration errors now include the name of the main
7754 configuration file - useful now that there may be more than one file in a
7755 list (.included file names were always shown).
7757 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
7758 for those rare installations that do not start the daemon as root or run it
7759 setuid root. I've cut out the call to initgroups() if the daemon is not
7762 60. The Exim user and group can now be bound into the binary as text strings
7763 that are looked up at the start of Exim's processing.
7765 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
7767 62. Added $mailstore_basename variable.
7769 63. Installed patch to sieve.c from Michael Haardt.
7771 64. When Exim failed to open the panic log after failing to open the main log,
7772 the original message it was trying to log was written to stderr and debug
7773 output, but if they were not available (the usual case in production), it
7774 was lost. Now it is written to syslog before the two lines that record the
7775 failures to open the logs.
7777 65. Users' Exim filters run in subprocesses under the user's uid. It is
7778 possible for a "deliver" command or an alias in a "personal" command to
7779 provoke an address rewrite. If logging of address rewriting is configured,
7780 this fails because the process is not running as root or exim. There may be
7781 a better way of dealing with this, but for the moment (because 4.30 needs
7782 to be released), I have disabled address rewrite logging when running a
7783 filter in a non-root, non-exim process.
7789 1. The buildconfig auxiliary program wasn't quoting the value set for
7790 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
7791 not defined. This bug was masked in 4.22 by the effect that was fixed in
7794 2. Some messages that were rejected after a message id was allocated were
7795 shown as "incomplete" by exigrep. It no longer does this for messages that
7796 are rejected by local_scan() or the DATA or non-SMTP ACLs.
7798 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
7799 have allow_domain_literals set, the ID did not get logged in the <= line.
7800 Domain literals are now always recognized in Message-ID: header lines.
7802 4. The first argument for a ${extract expansion item is the key name or field
7803 number. Leading and trailing spaces in this item were not being ignored,
7804 causing some misleading effects.
7806 5. When deliver_drop_privilege was set, single queue runner processes started
7807 manually (i.e. by the command "exim -q") or by the daemon (which uses the
7808 same command in the process it spins off) were not dropping privilege.
7810 6. When the daemon running as "exim" started a queue runner, it always
7811 re-executed Exim in the spun-off process. This is a waste of effort when
7812 deliver_drop_privilege is set. The new process now just calls the
7813 queue-runner function directly.
7819 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
7822 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
7823 ignored. Though the use of -oP was forcing the writing of a pid file, it
7824 was always written to the default place.
7826 3. If the message "no IP address found for host xxxx" is generated during
7827 incoming verification, it is now followed by identification of the incoming
7828 connection (so you can more easily find what provoked it).
7830 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
7832 5. Added some features to "harden" Exim a bit more against certain attacks:
7834 (a) There is now a build-time option called FIXED_NEVER_USERS that can
7835 be put in Local/Makefile. This is like the never_users runtime option,
7836 but it cannot be overridden. The default setting is "root".
7838 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
7839 prefix string with which any file named in a -C command line option
7842 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
7843 is retained for -C and -D only if the caller of Exim is root. Without
7844 it, the exim user may also use -C and -D and retain privilege.
7846 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
7847 command line option is disabled.
7849 6. Macro names set by the -D option must start with an upper case letter, just
7850 like macro names defined in the configuration file.
7852 7. Added "dereference=" facility to LDAP.
7854 8. Two instances of the typo "uknown" in the source files are fixed.
7856 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
7857 the Configure-Makefile script screwed up while processing it.
7859 10. Incorporated PCRE 4.4.
7861 11. The SMTP synchronization check was not operating right at the start of an
7862 SMTP session. For example, it could not catch a HELO sent before the client
7863 waited for the greeting. There is now a check for outstanding input at the
7864 point when the greeting is written. Because of the duplex, asynchronous
7865 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
7866 way, but not yet received, when the check is performed.
7868 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
7869 on TCP/IP sockets, because this apparently causes some broken clients to
7872 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
7873 unchanged) from the Cygwin maintainer.
7875 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
7876 was supported instead of testing for mbx. Effectively a typo.
7878 15. The spa authenticator server code was not checking that the input it
7879 received was valid base64.
7881 16. The debug output line for the "set" modifier in ACLs was not showing the
7882 name of the variable that was being set.
7884 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
7886 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
7887 lookup on its own. Something else was needed to trigger the lookup. For
7888 example, a match in host_lookup or the need for a host name in a host list.
7889 Now, if $sender_host_name is referenced and the host name has not yet been
7890 looked up, a lookup is performed. If the lookup fails, the variable remains
7891 empty, and $host_lookup_failed is set to "1".
7893 19. Added "eqi" as a case-independent comparison operator.
7895 20. The saslauthd authentication condition could segfault if neither service
7896 nor realm was specified.
7898 21. If an overflowing value such as "2048M" was set for message_size_limit, the
7899 error message that was logged was misleading, and incoming SMTP
7900 connections were dropped. The message is now more accurate, and temporary
7901 errors are given to SMTP connections.
7903 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
7904 (except RSET) with a 421 error, until QUIT is received. However, it was
7905 failing to send a response to QUIT.
7907 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
7908 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
7909 ACL is now run after the helo_try_verify_hosts code.
7911 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
7912 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
7913 case-independent, so other case variants are also recognized). Apparently
7914 some systems use these upper case variants.
7916 25. If more than two messages were waiting for the same host, and a transport
7917 filter was specified for the transport, Exim sent two messages over the
7918 same TCP/IP connection, and then failed with "socket operation on non-
7919 socket" when it tried to send the third.
7921 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
7923 27. The extern definition of crypt16() in expand.c was not being excluded when
7924 the OS had its own crypt16() function.
7926 28. Added bounce_return_body as a new option, and bounce_return_size_limit
7927 as a preferred synonym for return_size_limit, both as an option and as an
7930 29. Added LIBS=-liconv to OS/Makefile-OSF1.
7932 30. Changed the default configuration ACL to relax the local part checking rule
7933 for addresses that are not in any local domains. For these addresses,
7934 slashes and pipe symbols are allowed within local parts, but the sequence
7935 /../ is explicitly forbidden.
7937 31. SPA server authentication was not clearing the challenge buffer before
7940 32. log_message in a "warn" ACL statement was writing to the reject log as
7941 well as to the main log, which contradicts the documentation and doesn't
7942 seem right (because no rejection is happening). So I have stopped it.
7944 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
7945 However, I am unable to do any testing of this.
7947 34. Fixed an infelicity in the appendfile transport. When checking directories
7948 for a mailbox, to see if any needed to be created, it was accidentally
7949 using path names with one or more superfluous leading slashes; tracing
7950 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
7952 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
7953 contents are added to the log line that is written for every discarded
7954 recipient. (Previously a log_message setting was ignored.)
7956 36. The ${quote: operator now quotes the string if it is empty.
7958 37. The install script runs exim in order to find its version number. If for
7959 some reason other than non-existence or emptiness, which it checks, it
7960 could not run './exim', it was installing it with an empty version number,
7961 i.e. as "exim-". This error state is now caught, and the installation is
7964 38. An argument was missing from the function that creates an error message
7965 when Exim fails to connect to the socket for saslauthd authentication.
7966 This could cause Exim to crash, or give a corrupted message.
7968 39. Added isip, isip4, and isip6 to ${if conditions.
7970 40. The ACL variables $acl_xx are now saved with the message, and can be
7971 accessed later in routers, transports, and filters.
7973 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
7974 strings in the file are not string-expanded.
7976 42. If a MAIL command specified a SIZE value that was too large to fit into an
7977 int variable, the check against message_size_limit failed. Such values are
7978 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
7979 day this will have to be increased, but I don't think I want to be around
7980 when emails are that large.
7987 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
7988 iconv() is not standard in FreeBSD.
7990 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
7991 IPv6 enabled. The observed symptom was a segmentation fault on return from
7992 the function os_common_find_running_interfaces() in src/os.c.
7994 3. In the check_special_case() function in daemon.c I had used "errno" as an
7995 argument name, which causes warnings on some systems. This was basically a
7996 typo, since it was named "eno" in the comments!
7998 4. The code that waits for the clock to tick (at a resolution of some fraction
7999 of a second) so as to ensure message-id uniqueness was always waiting for
8000 at least one whole tick, when it could have waited for less. [This is
8001 almost certainly not relevant at current processor speeds, where it is
8002 unlikely to ever wait at all. But we try to future-proof.]
8004 5. The function that sleeps for a time interval that includes fractions of a
8005 second contained a race. It did not block SIGALRM between setting the
8006 timer, and suspending (a couple of lines later). If the interval was short
8007 and the sigsuspend() was delayed until after it had expired, the suspension
8008 never ended. On busy systems this could lead to processes getting stuck for
8011 6. Some uncommon configurations may cause a lookup to happen in a queue runner
8012 process, before it forks any delivery processes. The open lookup caching
8013 mechanism meant that the open file or database connection was passed into
8014 the delivery process. The problem was that delivery processes always tidy
8015 up cached lookup data. This could cause a problem for the next delivery
8016 process started by the queue runner, because the external queue runner
8017 process does not know about the closure. So the next delivery process
8018 still has data in the lookup cache. In the case of a file lookup, there was
8019 no problem because closing a file descriptor in a subprocess doesn't affect
8020 the parent. However, if the lookup was caching a connection to a database,
8021 the connection was closed, and the second delivery process was likely to
8022 see errors such as "PGSQL: query failed: server closed the connection
8023 unexpectedly". The problem has been fixed by closing all cached lookups
8024 in a queue runner before running a delivery process.
8026 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
8027 seem to have the "const" qualifier which it has on other OS. I've
8030 8. Change 4.21/2 was too strict. It is only if there are two authenticators
8031 *of the same type* (client or server) with the same public name that an
8032 error should be diagnosed.
8034 9. When Exim looked up a host name for an IP address, but failed to find the
8035 original IP address when looking up the host name (a safety check), it
8036 output the message "<ip address> does not match any IP for NULL", which was
8037 confusing, to say the least. The bug was that the host name should have
8038 appeared instead of "NULL".
8040 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
8041 user that is built into the binary, and the -C or -D options is used, root
8042 privilege is dropped before the configuration file is read. In addition,
8043 logging is switched to stderr instead of the normal log files. If the
8044 configuration then re-defines the Exim user, the unprivileged environment
8045 is probably not what is expected, so Exim logs a panic warning message (but
8048 However, if deliver_drop_privilege is set, the unprivileged state may well
8049 be exactly what is intended, so the warning has been cut out in that case,
8050 and Exim is allowed to try to write to its normal log files.
8056 1. smtp_return_error_details was not giving details for temporary sender
8057 or receiver verification errors.
8059 2. Diagnose a configuration error if two authenticators have the same public
8062 3. Exim used not to create the message log file for a message until the first
8063 delivery attempt. This could be confusing when incoming messages were held
8064 for policy or load reasons. The message log file is now created at the time
8065 the message is received, and an initial "Received" line is written to it.
8067 4. The automatically generated man page for command line options had a minor
8068 bug that caused no ill effects; however, a more serious problem was that
8069 the procedure for building the man page automatically didn't always
8070 operate. Consequently, release 4.20 contains an out-of-date version. This
8071 shouldn't happen again.
8073 5. When building Exim with embedded Perl support, the script that builds the
8074 Makefile was calling 'perl' to find its compile-time parameters, ignoring
8075 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
8077 6. The freeze_tell option was not being used for messages that were frozen on
8078 arrival, either by an ACL or by local_scan().
8080 7. Added the smtp_incomplete_transaction log selector.
8082 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
8083 was accepting AUTH without a new EHLO.
8085 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
8086 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
8089 10. Logging of TCP/IP connections (when configured) now happens in the main
8090 daemon process instead of the child process, so that the TCP/IP connection
8091 count is more accurate (but it can never be perfect).
8093 11. The use of "drop" in a nested ACL was not being handled correctly in the
8094 outer ACL. Now, if condition failure induced by the nested "drop" causes
8095 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
8096 or "require"), the connection is dropped.
8098 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
8099 that yield "discard" can now be used with an "accept" or a "discard" verb,
8100 but an error is generated for any others (because I can't see a useful way
8101 to define what should happen).
8103 13. When an ACL is read dynamically from a file (or anywhere else), the lines
8104 are now processed in the same way as lines in the Exim configuration file.
8105 In particular, continuation lines are supported.
8107 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
8109 15. Added -ti meaning -t -i.
8111 16. Check for letters, digits, hyphens, and dots in the names of dnslist
8112 domains, and warn by logging if others are found.
8114 17. At least on BSD, alignment is not guaranteed for the array of ifreq's
8115 returned from GIFCONF when Exim is trying to find the list of interfaces on
8116 a host. The code in os.c has been modified to copy each ifreq to an aligned
8117 structure in all cases.
8119 Also, in some cases, the returned ifreq's were being copied to a 'struct
8120 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
8121 means the last couple of bytes of an IPv6 address could be chopped if the
8122 ifreq contained only a normal sockaddr (14 bytes storage).
8124 18. Named domain lists were not supported in the hosts_treat_as_local option.
8125 An entry such as +xxxx was not recognized, and was treated as a literal
8128 19. Ensure that header lines added by a DATA ACL are included in the reject log
8129 if the ACL subsequently rejects the message.
8131 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
8132 MD5 (which is deprecated).
8134 21. When testing a filter file using -bf, Exim was writing a message when it
8135 took the sender from a "From " line in the message, but it was not doing so
8136 when it took $return_path from a Return-Path: header line. It now does.
8138 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
8139 with a valid header line field name (a series of printing characters
8140 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
8142 23. Changed "disc" in the source to "disk" to conform to the documentation and
8143 the book and for uniformity.
8145 24. Ignore Sendmail's -Ooption=value command line item.
8147 25. When execve() failed while trying to run a command in a pipe transport,
8148 Exim was returning EX_UNAVAILABLE (69) from the subprocess. However, this
8149 could be confused with a return value of 69 from the command itself. This
8150 has been changed to 127, the value the shell returns if it is asked to run
8151 a non-existent command. The wording for the related log line suggests a
8152 non-existent command as the problem.
8154 26. If received_header_text expands to an empty string, do not add a Received:
8155 header line to the message. (Well, it adds a token one on the spool, but
8156 marks it "old" so that it doesn't get used or transmitted.)
8158 27. Installed eximstats 1.28 (addition of -nt option).
8160 28. There was no check for failure on the call to getsockname() in the daemon
8161 code. This can fail if there is a shortage of resources on the system, with
8162 ENOMEM, for example. A temporary error is now given on failure.
8164 29. Contrary to the C standard, it seems that in some environments, the
8165 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
8166 program. Exim now does this explicitly; it affects the formatting of
8167 timestamps using strftime().
8169 30. If exiqsumm was given junk data, it threw up some uninitialized variable
8170 complaints. I've now initialized all the variables, to avoid this.
8172 32. Header lines added by a system filter were not being "seen" during
8173 transport-time rewrites.
8175 33. The info_callback() function passed to OpenSSL is set up with type void
8176 (*)(SSL *, int, int), as described somewhere. However, when calling the
8177 function (actually a macro) that sets it up, the type void(*)() is
8178 expected. I've put in a cast to prevent warnings from picky compilers.
8180 34. If a DNS black list lookup found a CNAME record, but there were no A
8181 records associated with the domain it pointed at, Exim crashed.
8183 35. If a DNS black list lookup returned more than one A record, Exim ignored
8184 all but the first. It now scans all returned addresses if a particular IP
8185 value is being sought. In this situation, the contents of the
8186 $dnslist_value variable are a list of all the addresses, separated by a
8189 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
8190 to accept a host name and all its aliases if the forward lookup for any of
8191 them yielded the IP address of the incoming connection. Now it accepts only
8192 those names whose forward lookup yields the correct IP address. Any other
8193 names are discarded. This closes a loophole whereby a rogue DNS
8194 administrator could create reverse DNS records to break through a
8195 wildcarded host restriction in an ACL.
8197 37. If a user filter or a system filter that ran in a subprocess used any of
8198 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
8199 the wrong values were passed to the pipe command ($thisaddress had the
8200 value of $0, $0 had the value of $1, etc). This bug was introduced by
8201 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
8203 38. Improved the line breaking for long SMTP error messages from ACLs.
8204 Previously, if there was no break point between 40 and 75 characters, Exim
8205 left the rest of the message alone. Two changes have been made: (a) I've
8206 reduced the minimum length to 35 characters; (b) if it can't find a break
8207 point between 35 and 75 characters, it looks ahead and uses the first one
8208 that it finds. This may give the occasional overlong line, but at least the
8209 remaining text gets split now.
8211 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
8212 file descriptors might be low, and that setting 1000 would always raise it.
8213 It turns out that in some environments, the limit is already over 1000 and
8214 that lowering it causes trouble. So now Exim takes care not to decrease it.
8216 40. When delivering a message, the value of $return_path is set to $sender_
8217 address at the start of routing (routers may change the value). By an
8218 oversight, this default was not being set up when an address was tested by
8219 -bt or -bv, which affected the outcome if any router or filter referred to
8222 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
8223 log only when "message" or "log_message" are set. However, if one of the
8224 conditions was an address verification, or a call to a nested ACL, the
8225 messages generated by the underlying test were being passed through. This
8226 no longer happens. The underlying message is available in $acl_verify_
8227 message for both "message" and "log_message" expansions, so it can be
8228 passed through if needed.
8230 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
8231 new expansion $bh_ to give the encoded byte string without charset
8232 translation. Translation happens only if iconv() is available; HAVE_ICONV
8233 indicates this at build time. HEADERS_CHARSET gives the charset to
8234 translate to; headers_charset can change it in the configuration, and
8235 "headers charset" can change it in an individual filter file.
8237 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
8238 that creates RFC 2047 encoded "words" labels them as that charset instead
8239 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
8240 expansion operator; (ii) when Exim creates a From: line for a local
8241 message; (iii) when a header line is rewritten to include a "phrase" part.
8243 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
8244 buggy, causing it to skip the first lines of messages whose message ID
8245 ended in 'D'. This would not have bitten before Exim release 4.14, because
8246 message IDs were unlikely to end in 'D' before then. The effect was to have
8247 incorrect size information for certain domains.
8249 45. #include "config.h" was missing at the start of the crypt16.c module. This
8250 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
8253 46. If there was a timeout during a "random" callout check, Exim treated it as
8254 a failure of the random address, and carried on sending RSET and the real
8255 address. If the delay was just some slowness somewhere, the response to the
8256 original RCPT would be taken as a response to RSET and so on, causing
8257 mayhem of various kinds.
8259 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
8260 when I implemented it. It didn't allow for the fact that some option values
8261 may legitimately be negative (e.g. size_addition), and it didn't even do
8262 the right test for positive values.
8264 48. Domain names in DNS records are case-independent. Exim always looks them up
8265 in lower case. Some resolvers return domain names in exactly the case they
8266 appear in the zone file, that is, they may contain uppercase letters. Not
8267 all resolvers do this - some return always lower case. Exim was treating a
8268 change of case by a resolver as a change of domain, similar to a widening
8269 of a domain abbreviation. This triggered its re-routing code and so it was
8270 trying to route what was effectively the same domain again. This normally
8271 caused routing to fail (because the router wouldn't handle the domain
8272 twice). Now Exim checks for this case specially, and just changes the
8273 casing of the domain that it ultimately uses when it transmits the message
8276 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
8279 50. If a filter generated a file delivery with a non-absolute name (possible if
8280 no home directory exists for the router), the forbid_file option was not
8283 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
8284 the existing equality matching.
8286 52. Exim was using ints instead of ino_t variables in some places where it was
8287 dealing with inode numbers.
8289 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
8290 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
8291 called TMPDIR, and if it finds it is different, it changes its value.
8293 54. The smtp_printf() function is now made available to local_scan() so
8294 additional output lines can be written before returning. There is also an
8295 smtp_fflush() function to enable the detection of a dropped connection.
8296 The variables smtp_input and smtp_batched_input are exported to
8299 55. Changed the default runtime configuration: the message "Unknown user"
8300 has been removed from the ACL, and instead placed on the localuser router,
8301 using the cannot_route_message feature. This means that any verification
8302 failures that generate their own messages won't get overridden. Similarly,
8303 the "Unrouteable address" message that was in the ACL for unverifiable
8304 relay addresses has also been removed.
8306 56. Added hosts_avoid_esmtp to the smtp transport.
8308 57. The exicyclog script was not checking for the esoteric option
8309 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
8310 will work only if exicyclog is run under the appropriate euid.
8312 58. Following a discussion on the list, the rules by which Exim recognises line
8313 endings on incoming messages have been changed. The -dropcr and drop_cr
8314 options are now no-ops, retained only for backwards compatibility. The
8315 following line terminators are recognized: LF CRLF CR. However, special
8316 processing applies to CR:
8318 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
8319 nor a local message in the state where . is a terminator.
8321 (ii) If a bare CR is encountered in a header line, an extra space is added
8322 after the line terminator so as not to end the header. The reasoning
8323 behind this is that bare CRs in header lines are most likely either
8324 to be mistakes, or people trying to play silly games.
8326 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
8327 was being incorrectly given as 18 bytes larger than it should have been.
8328 This is a VOB (very old bug).
8330 60. This may never have affected anything current, but just in case it has:
8331 When the local host is found other than at the start of a list of hosts,
8332 the local host, those with the same MX, and any that follow, are discarded.
8333 When the list in question was part of a longer list of hosts, the following
8334 hosts (not currently being processed) were also being discarded. This no
8335 longer happens. I'm not sure if this situation could ever has previously
8338 61. Added the "/MX" feature to lists of hosts in the manualroute and query
8341 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
8342 header. This is something that is recommended in a new Internet Draft, and
8343 is something that is documented as being done by Sendmail. There are two
8344 possible values. For messages generated by the autoreply transport, Exim
8347 Auto-Submitted: auto-replied
8349 whereas for all other generated messages (e.g. bounces) it adds
8351 Auto-Submitted: auto-generated
8353 63. The "personal" condition in filters now includes a test for the
8354 Auto-Submitted: header. If it contains the string "auto-" the message it
8355 not considered personal.
8357 64. Added rcpt_include_affixes as a generic transport option.
8359 65. Added queue_only_override (default true).
8361 66. Added the syslog_duplication option.
8363 67. If what should have been the first header line of a message consisted of
8364 a space followed by a colon, Exim was mis-interpreting it as a header line.
8365 It isn't of course - it is syntactically invalid and should therefore be
8366 treated as the start of the message body. The misbehaviour could have
8367 caused a number of strange effects, including loss of data in subsequent
8368 header lines, and spool format errors.
8370 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
8371 client host had authenticated. This control can now be exercised by an ACL
8372 for more flexibility.
8374 69. By default, callouts do not happen when testing with -bh. There is now a
8375 variant, -bhc, which does actually run the callout code, including
8376 consulting and updating the callout cache.
8378 70. Added support for saslauthd authentication, courtesy of Alexander
8381 71. If statvfs() failed on the spool or log directories while checking their
8382 size for availability, Exim confusingly gave the error "space shortage".
8383 Furthermore, in debugging mode it crashed with a floating point exception.
8384 These checks are done if check_{spool,log}_{space,inodes} are set, and when
8385 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
8386 serious problem, Exim now writes to the main and panic logs when this
8387 happens, with details of the failure. It then refuses to accept the
8388 incoming message, giving the message "spool directory problem" or "log
8389 directory problem" with a 421 code for SMTP messages.
8391 72. When Exim is about to re-exec itself, it ensures that the file descriptors
8392 0, 1, and 2 exist, because some OS complain for execs without them (see
8393 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
8394 descriptors. However, the code omitted to check that the open succeeded,
8395 causing mysterious errors if for some reason the permissions on /dev/null
8396 got screwed. Now Exim writes a message to the main and panic logs, and
8397 bombs out if it can't open /dev/null.
8399 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
8400 interact so that it is all more flexible. It is supposed to remain
8401 backwards compatible. Also added extra_local_interfaces.
8403 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
8404 to bomb out with an assertion failure - to the client this appears as a
8405 connection drop. This problem occurs in the part of the code that was taken
8406 from the Samba project. Fortunately, the assertion is in a very simple
8407 function, so I have fixed this by reproducing the function inline in the
8408 one place where it is called, and arranging for authentication to fail
8409 instead of killing the process with assert().
8411 75. The SPA client code was not working when the server requested OEM rather
8412 than Unicode encoding.
8414 76. Added code to make require_files with a specific uid setting more usable in
8415 the case where statting the file as root fails - usually a non-root-mounted
8416 NFS file system. When this happens and the failure is EACCES, Exim now
8417 forks a subprocess and does the per-uid checking as the relevant uid.
8419 77. Added process_log_path.
8421 78. If log_file_path was not explicitly set, a setting of check_log_space or
8422 check_log_inodes was ignored.
8424 79. If a space check for the spool or log partitions fails, the incident is now
8425 logged. Of course, in the latter case the data may get lost...
8427 80. Added the %p formatting code to string_format() so that it can be used to
8428 print addresses in debug_print(). Adjusted all the address printing in the
8429 debugging in store.c to use %p rather than %d.
8431 81. There was a concern that a line of code in smtp_in.c could overflow a
8432 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
8433 initially expressed, the concern was not well-founded, because trailing
8434 spaces are removed early. However, if the trailing spaces were followed by
8435 a NULL, they did not get removed, so the overflow was possible. Two fixes
8438 (a) I re-wrote the offending code in a cleaner fashion.
8439 (b) If an incoming SMTP command contains a NULL character, it is rejected
8442 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
8443 runs initgroups(), so that if the Exim user is in any additional groups,
8444 they will be used during message reception.
8450 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / exim.git / blob