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.
20 JH/01 Move the wait-for-next-tick (needed for unique messmage IDs) from
21 after reception to before a subsequent reception. This should
22 mean slightly faster delivery, and also confirmation of reception
25 JH/02 Move from using the pcre library to pcre2. The former is no longer
26 being developed or supported (by the original developer).
28 JH/03 Constification work in the filters module required a major version
29 bump for the local-scan API. Specifically, the "headers_charset"
30 global which is visible via the API is now const and may therefore
31 not be modified by local-scan code.
33 JH/04 Fix ClamAV TCP use under FreeBSD. Previously the OS-specific shim for
34 sendfile() didi not account for the way the ClamAV driver code called it.
36 JH/05 Bug 2819: speed up command-line messages being read in. Previously a
37 time check was being done for every character; replace that with one
40 JH/06 Bug 2815: Fix ALPN sent by server under OpenSSL. Previously the string
41 sent was prefixed with a length byte.
43 JH/07 Change the SMTP feature name for pipelining connect to be compliant with
44 RFC 5321. Previously Dovecot (at least) would log errors during
47 JH/08 Remove stripping of the binaries from the FreeBSD build. This was added
48 in 4.61 without a reason logged. Binaries will be bigger, which might
49 matter on diskspace-constrained systems, but debug is easier.
51 JH/09 Fix macro-definition during "-be" expansion testing. The move to
52 write-protected store for macros had not accounted for these runtime
53 additions; fix by removing this protection for "-be" mode.
55 JH/10 Convert all uses of select() to poll(). FreeBSD 12.2 was found to be
56 handing out large-numbered file descriptors, violating the usual Unix
57 assumption (and required by Posix) that the lowest possible number will be
58 allocated by the kernel when a new one is needed. In the daemon, and any
59 child procesees, values higher than 1024 (being bigger than FD_SETSIZE)
60 are not useable for FD_SET() [and hence select()] and overwrite the stack.
61 Assorted crashes happen.
63 JH/11 Fix use of $sender_host_name in daemon process. When used in certain
64 main-section options or in a connect ACL, the value from the first ever
65 connection was never replaced for subsequent connections. Found by
68 JH/12 Bug 2838: Fix for i32lp64 hard-align platforms. Found for SPARC Linux,
69 though only once PCRE2 was introduced: the memory accounting used under
70 debug offset allocations by an int, giving a hard trap in early startup.
71 Change to using a size_t. Debug and fix by John Paul Adrian Glaubitz.
73 JH/13 Bug 2845: Fix handling of tls_require_ciphers for OpenSSL when a value
74 with underbars is given. The write-protection of configuration introduced
75 in 4.95 trapped when normalisation was applied to an option not needing
78 JH/14 Bug 1895: TLS: Deprecate RFC 5114 Diffie-Hellman parameters.
80 JH/15 Fix a resource leak in *BSD. An off-by-one error resulted in the daemon
81 failing to close the certificates directory, every hour or any time it
84 JH/16 Debugging initiated by an ACL control now continues through into routing
85 and transport processes. Previously debugging stopped any time Exim
86 re-execs, or for processing a queued message.
88 JH/17 The "expand" debug selector now gives more detail, specifically on the
89 result of expansion operators and items.
91 JH/18 Bug 2751: Fix include_directory in redirect routers. Previously a
92 bad comparison between the option value and the name of the file to
93 be included was done, and a mismatch was wrongly identified.
94 4.88 to 4.95 are affected.
96 JH/19 Support for Berkeley DB versions 1 and 2 is withdrawn.
98 JH/20 When built with NDBM for hints DB's check for nonexistence of a name
99 supplied as the db file-pair basename. Previously, if a directory
100 path was given, for example via the autoreply "once" option, the DB
101 file.pag and file.dir files would be created in that directory's
104 JH/21 Remove the "allow_insecure_tainted_data" main config option and the
105 "taint" log_selector. These were previously deprecated.
107 JH/22 Fix static address-list lookups to properly return the matched item.
108 Previously only the domain part was returned.
110 JH/23 Bug 2864: FreeBSD: fix transport hang after 4xx/5xx response. Previously
111 the call into OpenSSL to send a TLS Close was being repeated; this
112 resulted in the library waiting for the peer's Close. If that was never
113 sent we waited forever. Fix by tracking send calls.
115 JH/24 The ${run} expansion item now expands its command string elements after
116 splitting. Previously it was before; the new ordering makes handling
117 zero-length arguments simpler. The old ordering can be obtained by
118 appending a new option "preexpand", after a comma, to the "run".
120 JH/25 Taint-check exec arguments for transport-initiated external processes.
121 Previously, tainted values could be used. This affects "pipe", "lmtp" and
122 "queryprogram" transport, transport-filter, and ETRN commands.
123 The ${run} expansion is also affected: in "preexpand" mode no part of
124 the command line may be tainted, in default mode the executable name
127 JH/26 Fix CHUNKING on a continued-transport. Previously the usabliility of
128 the the facility was not passed across execs, and only the first message
129 passed over a connection could use BDAT; any further ones using DATA.
131 JH/27 Support the PIPECONNECT facility in the smtp transport when the helo_data
132 uses $sending_ip_address and an interface is specified.
133 Previously any use of the local address in the EHLO name disabled
134 PIPECONNECT, the common case being to use the rDNS of it.
136 JH/28 OpenSSL: fix transport-required OCSP stapling verification under session
137 resumption. Previously verify failed because no certificate status is
138 passed on the wire for the restarted session. Fix by using the recorded
139 ocsp status of the stored session for the new connection.
141 JH/29 TLS resumption: the key for session lookup in the client now includes
142 more info that a server could potentially use in configuring a TLS
143 session, avoiding oferring mismatching sessions to such a server.
144 Previously only the server IP was used.
146 JH/30 Fix string_copyn() for limit greater than actual string length.
147 Previously the copied amount was the limit, which could result in a
148 overlapping memcpy for newly allocated destination soon after a
149 source string shorter than the limit. Found/investigated by KM.
151 JH/31 Bug 2886: GnuTLS: Do not free the cached creds on transport connection
152 close; it may be needed for a subsequent connection. This caused a
153 SEGV on primary-MX defer. Found/investigated by Gedalya & Andreas.
155 JH/32 Fix CHUNKING for a second message on a connection when the first was
156 rejected. Previously we did not reset the chunking-offered state, and
157 erroneously rejected the BDAT command. Investigation help from
160 JH/33 Fis ${srs_encode ...} to handle an empty sender address, now returning
161 an empty address. Previously the expansion returned an error.
163 HS/01 Bug 2855: Handle a v4mapped sender address given us by a frontending
164 proxy. Previously these were misparsed, leading to paniclog entries.
170 JH/01 Bug 1329: Fix format of Maildir-format filenames to match other mail-
171 related applications. Previously an "H" was used where available info
172 says that "M" should be, so change to match.
174 JH/02 Bug 2587: Fix pam expansion condition. Tainted values are commonly used
175 as arguments, so an implementation trying to copy these into a local
176 buffer was taking a taint-enforcement trap. Fix by using dynamically
177 created buffers. Similar fix for radius expansion condition.
179 JH/03 Bug 2586: Fix listcount expansion operator. Using tainted arguments is
180 reasonable, eg. to count headers. Fix by using dynamically created
181 buffers rather than a local. Do similar fixes for ACL actions "dcc",
182 "log_reject_target", "malware" and "spam"; the arguments are expanded
183 so could be handling tainted values.
185 JH/04 Bug 2590: Fix -bi (newaliases). A previous code rearrangement had
186 broken the (no-op) support for this sendmail command. Restore it
187 to doing nothing, silently, and returning good status.
189 JH/05 Bug 2593: Fix "vacation" in Exim filter. Previously, when a "once"
190 record path was given (or the default used) without a leading directory
191 path, an error occurred on trying to open it. Use the transport's working
194 JH/06 Bug 2594: Change the name used for certificate name checks in the smtp
195 transport. Previously it was the name on the DNS A-record; use instead
196 the head of the CNAME chain leading there (if there is one). This seems
197 to align better with RFC 6125.
199 JH/07 Bug 2597: Fix a resource leak. Using a lookup in obtaining a value for
200 smtp_accept_max_per_host allocated resources which were not released
201 when the limit was exceeded. This eventually crashed the daemon. Fix
202 by adding a release action in that path.
204 JH/08 Bug 2598: Fix verify ACL condition. The options for the condition are
205 expanded; previously using tainted values was rejected. Fix by using
206 dynamically-created buffers.
208 JH/09 Relax restrictions on ACL verify condition needing access to message
209 headers. Previously they were only permitted in data and non-smtp ACLs;
210 permit also mime, dkim, prdr quit and notquit. Applies to header-syntax,
211 not_blind, header_sender and header_names_ascii verification.
213 JH/10 Bug 2603: Fix coding of string copying to only evaluate arguments once.
214 Previously a macro used one argument twice; when called with the
215 argument as an expression having side-effects, incorrect operation
216 resulted. Use an inlineable function.
218 JH/11 Bug 2604: Fix request to cutthrough-deliver when a connection is already
219 held open for a verify callout. Previously this wan not accounted for
220 and a corrupt onward SMTP conversation resulted.
222 JH/12 Bug 2607: Fix the ${srs_encode } expansion to handle quoted local_parts.
223 Previously they were embedded naively in the constructed address; when
224 needed, strip the quoting and quote the entire local_part.
225 Also make the inbound_srs expansion condition handle quoting.
227 JH/13 Fix dsearch "subdir" filter to ignore ".". Previously only ".." was
228 excluded, not matching the documentation.
230 JH/14 Bug 2606: Fix a segfault in sqlite lookups. When no, or a bad, filename
231 was given for the sqlite_dbfile a trap resulted.
233 JH/15 Bug 2620: Fix "spam" ACL condition. Previously, tainted values for the
234 "name" argument resulted in a trap. There is no reason to disallow such;
235 this was a coding error.
237 JH/16 Bug 2615: Fix pause during message reception, on systems that have been
238 suspended/resumed. The Linux CLOCK_MONOTONIC does not account for time
239 spent suspended, ignoring the POSIX definition. Previously we assumed
240 it did and a constant offset from real time could be used as a correction.
241 Change to using the same clock source for the start-of-message and the
242 post-message next-tick-wait. Also change to using CLOCK_BOOTTIME if it
243 exists, just to get a clock slightly more aligned to reality.
245 JH/17 Bug 2295: Fix DKIM signing to always semicolon-terminate. Although the
246 RFC says it is optional some validators care. The missing char was not
247 intended but triggered by a line-wrap alignment. Discovery and fix by
248 Guillaume Outters, hacked on by JH.
250 JH/18 Bug 2617: Fix a taint trap in parse_fix_phrase(). Previously when the
251 name being quoted was tainted a trap would be taken. Fix by using
252 dynamically created buffers. The routine could have been called by a
253 rewrite with the "h" flag, by using the "-F" command-line option, or
254 by using a "name=" option on a control=submission ACL modifier.
256 JH/19 SPF: change the Authentication-Results expansion component to give
257 smtp.helo when the sender domain is empty. Previously it gave
260 JH/20 Bug 2631: ACL dnslist conditions now ignore and log any lookups returns
261 not in 127.0.0.0/8 to help in spotting list domains taken over by a
262 domain-parking registrar.
264 JH/21 Bug 2630: Fix eol-replacement string for the ${readsocket } expansion.
265 Previously when a whitespace character was specified it was not inserted
266 after removing the newline.
268 JH/22 Bug 2265: Force SNI usage for smtp transport DANE'd connections, to be
269 the domain part of the recipient address. This overrides any tls_sni
270 option set, which was previously used.
272 JH/23 Logging: with the +tls_sni log_selector, do not wrap the received SNI
275 JH/24 Bug 2634: Fix a taint trap seen on NetBSD: the testing coded for
276 is_tainted() had an off-by-one error in the overenthusiastic direction.
277 Find and fix by Gavan. Although NetBSD is not a supported platform for
278 4.94 this bug could affect other platforms.
280 PP/01 Fix default prime selection to be consistent.
281 One path used ike23 still, instead of exim.dev.20160529.3; now both
282 execution flows will use the same DH primes (currently
283 exim.dev.20160529.3).
285 JH/25 OpenSSL: Fix back-compatibility behaviour surrounding tls_certificates
286 option in smtp transport, to match the documentation. Previously
287 verification was not being done in some cases where it should have been.
289 JH/26 Bug 2646: fix a memory usage issue in ldap lookups. Previously, when more
290 than one server was defined and depending on the platform memory layout
291 details, an internal consistency trap could be hit while walking the list
294 JH/27 Bug 2648: fix the passing of an authenticator public-name through spool
295 files. The value is used by the authresults expansion item. Previously
296 if this was used in a router or transport, a crash could result.
298 JH/28 Fix spurious logging of select error. Some platforms, notably FreeBSD,
299 have a sufficient incidence of EINTR returns from select that an
300 interaction with other operations done by the main daemon loop exposed
301 a bug in the error-handling. This was benign apart from the log
304 JH/29 Bug 2675: add outgoing-interface I= element to deferred "==" log lines,
305 for consistency with delivered "=>" and failed "**" lines. While we're
306 there, handle PRX and TFO.
308 JH/30 Bug 2677: fix matching of long addresses. Since 4.93 a limit of 256 was
309 applied. This resulted, if any header-line rewrite rules were configured,
310 in a panic-log triggerable by sending a message with a long address in
311 a header. Fix by increasing the arbitrary limit to larger than a single
312 (dewrapped) 5322 header line maximum size.
314 JH/31 The ESMTP option name advertised for the SUPPORT_EARLY_PIPE build option
315 is changed from X_PIPE_CONNECT to PIPE_CONNECT. This is in line with
316 RFC 6648 which deprecates X- options in protocols as a general practice.
317 Changeover between the implementations is handled by the mechanisms
320 JH/32 Bug 2599: fix delay of delivery to a local address where there is also
321 a remote which uses callout/hold. Previously the local was queued.
323 JH/33 Fix a taint trap in the ${listextract } expansion when the source data
326 JH/34 Fix the placement of a multiple-message delivery marker in the delivery
327 log line. The asterisk is now consistently appended to the remote IP
328 (and port, if given), and will also be provided on defer and fail log
329 lines. Previously it could be placed on the local IP if that was being
330 logged, and was only provided on delivery lines.
332 JH/35 Bug 2343: Harden exim_tidydb against corrupt wait- files.
334 JH/36 Bug 2687: Fix interpretation of multiple ^ chars in a plaintext
335 authenticator client_send option. Previously the next char, after a pair
336 was collapsed, was taken verbatim (so ^^^foo became ^^foo; ^^^^foo became
337 ^^\x00foo). Fixed to get ^\x00foo and ^^foo respectively to match the
338 documentation. There is still no way to get a leading ^ immediately
339 after a NUL (ie. for the password of a PLAIN method authenticator.
341 JH/37 Enforce the expected size, for fixed-size records read from hints-DB
342 files. For bad sizes read, delete the record and whine to paniclog.
344 JH/38 When logging an AUTH failure, as server, do not include sensitive
345 information. Previously, the credentials would be included if given
346 as part of the AUTH command line and an ACL denied authentication.
348 JH/39 Bug 2691: fix $local_part_data. When the matching list element
349 referred to a file, bad data was returned. This likely also affected
352 JH/40 The gsasl authenticator now supports caching of the salted password
353 generated by the client-side implementation. This required the addition
354 of a new variable: $auth4.
356 JH/41 Fix daemon SIGHUP on FreeBSD. Previously, a named socket for IPC was
357 left undeleted; the attempt to re-create it then failed - resulting in
358 the usual "SIGHUP tp have daemon reload configuration" to not work.
359 This affected any platform not supporting "abstract" Unix-domain
360 sockets (i.e. not Linux).
362 JH/42 Bug 2693: Harden against a peer which reneges on a 452 "too many
363 recipients" response to RCPT in a later response, with a 250. The
364 previous coding assumed this would not happen, and under PIPELINING
365 would result in both lost and duplicate recipients for a message.
367 JH/43 Bug 2694: Fix weighted distribution of work to multiple spamd servers.
368 Previously the weighting was incorrectly applied. Similar fix for socks
369 proxies. Found and fixed by Heiko Schlichting.
371 JH/44 Bug 2701: Fix list-expansion of dns_ipv4_lookup. Previously, it did
372 not handle sub-lists included using the +namedlist syntax. While
373 investigating, the same found for dns_trust_aa, dns_again_means_nonexist,
374 dnssec_require_domains, dnssec_request_domains, srv_fail_domains,
377 JH/45 Use a (new) separate store pool-pair for DKIM verify working data.
378 Previously the permanent pool was used, so the sore could not be freed.
379 This meant a connection with many messages would use continually-growing
382 JH/46 Use an exponentially-increasing block size when malloc'ing store. Do it
383 per-pool so as not to waste too much space. Previously a constant size
384 was used which resulted in O(n^2) behaviour; now we get O(n log n) making
385 DOS attacks harder. The cost is wasted memory use in the larger blocks.
387 JH/47 Use explicit alloc/free for DNS lookup workspace. This permits using the
388 same space repeatedly, and a smaller process footprint.
390 JH/48 Use a less bogus-looking filename for a temporary used for DH-parameters
391 for GnuTLS. Previously the name started "%s" which, while not a bug,
392 looked as if if might be one.
394 JH/49 Bug 2710: when using SOCKS for additional messages after the first (a
395 "continued connection") make the $proxy_* variables available. Previously
396 the information was not passed across the exec() call for subsequent
397 transport executions. This also mean that the log lines for the
398 messages can show the proxy information.
400 JH/50 Bug 2672: QT elements in log lines, unless disabled, now exclude the
401 receive time. With modern systems the difference is significant.
402 The historical behaviour can be restored by disabling (a new) log_selector
403 "queue_time_exclusive".
405 JH/51 Taint-check ACL line. Previously, only filenames (for out-of-line ACL
406 content) were specifically tested for. Now, also cover expansions
407 resulting in ACL names and inline ACL content.
409 JH/52 Fix ${ip6norm:} operator. Previously, any trailing line text was dropped,
410 making it unusable in complex expressions.
412 JH/53 Bug 2743: fix immediate-delivery via named queue. Previously this would
413 fail with a taint-check on the spoolfile name, and leave the message
416 HS/01 Enforce absolute PID file path name.
418 HS/02 Handle SIGINT as we handle SIGTERM: terminate the Exim process.
420 PP/01 Add a too-many-bad-recipients guard to the default config's RCPT ACL.
422 PP/02 Bug 2643: Correct TLS DH constants.
423 A missing NUL termination in our code-generation tool had led to some
424 incorrect Diffie-Hellman constants in the Exim source.
425 Reported by kylon94, code-gen tool fix by Simon Arlott.
427 PP/03 Impose security length checks on various command-line options.
428 Fixes CVE-2020-SPRSS reported by Qualys.
430 PP/04 Fix Linux security issue CVE-2020-SLCWD and guard against PATH_MAX
431 better. Reported by Qualys.
433 PP/05 Fix security issue CVE-2020-PFPSN and guard against cmdline invoker
434 providing a particularly obnoxious sender full name.
437 PP/06 Fix CVE-2020-28016 (PFPZA): Heap out-of-bounds write in parse_fix_phrase()
439 PP/07 Refuse to allocate too little memory, block negative/zero allocations.
442 PP/08 Change default for recipients_max from unlimited to 50,000.
444 PP/09 Fix security issue with too many recipients on a message (to remove a
445 known security problem if someone does set recipients_max to unlimited,
446 or if local additions add to the recipient list).
447 Fixes CVE-2020-RCPTL reported by Qualys.
449 PP/10 Fix security issue in SMTP verb option parsing
450 Fixes CVE-2020-EXOPT reported by Qualys.
452 PP/11 Fix security issue in BDAT state confusion.
453 Ensure we reset known-good where we know we need to not be reading BDAT
454 data, as a general case fix, and move the places where we switch to BDAT
455 mode until after various protocol state checks.
456 Fixes CVE-2020-BDATA reported by Qualys.
458 HS/03 Die on "/../" in msglog file names
460 QS/01 Creation of (database) files in $spool_dir: only uid=0 or the uid of
461 the Exim runtime user are allowed to create files.
463 QS/02 PID file creation/deletion: only possible if uid=0 or uid is the Exim
466 QS/03 When reading the output from interpreted forward files we do not
467 pass the pipe between the parent and the interpreting process to
468 executed child processes (if any).
470 QS/04 Always die if requested from internal logging, even is logging is
473 JH/54 DMARC: recent versions of the OpenDMARC library appear to have broken
474 the API; compilation noo longer completes with DMARC support included.
475 This affects 1.4.1-1 on Fedora 33 (1.3.2-3 is functional); and has
476 been reported on other platforms.
478 JH/55 TLS: as server, reject connections with ALPN indicating non-smtp use.
480 JH/56 Make the majority of info read from config files readonly, for defence-in-
481 depth against exploits. Suggestion by Qualys.
482 Not supported on Solaris 10.
484 JH/57 Fix control=fakreject for a custom message containing tainted data.
485 Previously this resulted in a log complaint, due to a re-expansion present
486 since fakereject was originally introduced.
488 JH/58 GnuTLS: Fix certextract expansion. If a second modifier after a tag
489 modifier was given, a loop resulted.
491 JH/59 DKIM: Fix small-message verification under TLS with chunking. If a
492 pipelined SMTP command followed the BDAT LAST then it would be
493 incorrectly treated as part of the message body, causing a verification
496 JH/60 Bug 2805: Fix logging of domain-literals in Message_ID: headers. They
497 require looser validation rules than those for 821-level addresses,
498 which only permit IP addresses.
504 JH/01 Avoid costly startup code when not strictly needed. This reduces time
505 for some exim process initialisations. It does mean that the logging
506 of TLS configuration problems is only done for the daemon startup.
508 JH/02 Early-pipelining support code is now included unless disabled in Makefile.
510 JH/03 DKIM verification defaults no long accept sha1 hashes, to conform to
511 RFC 8301. They can still be enabled, using the dkim_verify_hashes main
514 JH/04 Support CHUNKING from an smtp transport using a transport_filter, when
515 DKIM signing is being done. Previously a transport_filter would always
516 disable CHUNKING, falling back to traditional DATA.
518 JH/05 Regard command-line recipients as tainted.
520 JH/06 Bug 340: Remove the daemon pid file on exit, when due to SIGTERM.
522 JH/07 Bug 2489: Fix crash in the "pam" expansion condition. It seems that the
523 PAM library frees one of the arguments given to it, despite the
524 documentation. Therefore a plain malloc must be used.
526 JH/08 Bug 2491: Use tainted buffers for the transport smtp context. Previously
527 on-stack buffers were used, resulting in a taint trap when DSN information
528 copied from a received message was written into the buffer.
530 JH/09 Bug 2493: Harden ARC verify against Outlook, whick has been seen to mix
531 the ordering of its ARC headers. This caused a crash.
533 JH/10 Bug 2492: Use tainted memory for retry record when needed. Previously when
534 a new record was being constructed with information from the peer, a trap
537 JH/11 Bug 2494: Unset the default for dmarc_tld_file. Previously a naiive
538 installation would get error messages from DMARC verify, when it hit the
539 nonexistent file indicated by the default. Distros wanting DMARC enabled
540 should both provide the file and set the option.
541 Also enforce no DMARC verification for command-line sourced messages.
543 JH/12 Fix an uninitialised flag in early-pipelining. Previously connections
544 could, depending on the platform, hang at the STARTTLS response.
546 JH/13 Bug 2498: Reset a counter used for ARC verify before handling another
547 message on a connection. Previously if one message had ARC headers and
548 the following one did not, a crash could result when adding an
549 Authentication-Results: header.
551 JH/14 Bug 2500: Rewind some of the common-coding in string handling between the
552 Exim main code and Exim-related utities. The introduction of taint
553 tracking also did many adjustments to string handling. Since then, eximon
554 frequently terminated with an assert failure.
556 JH/15 When PIPELINING, synch after every hundred or so RCPT commands sent and
557 check for 452 responses. This slightly helps the inefficieny of doing
558 a large alias-expansion into a recipient-limited target. The max_rcpt
559 transport option still applies (and at the current default, will override
560 the new feature). The check is done for either cause of synch, and forces
561 a fast-retry of all 452'd recipients using a new MAIL FROM on the same
562 connection. The new facility is not tunable at this time.
564 JH/16 Fix the variables set by the gsasl authenticator. Previously a pointer to
565 library live data was being used, so the results became garbage. Make
566 copies while it is still usable.
568 JH/17 Logging: when the deliver_time selector ise set, include the DT= field
569 on delivery deferred (==) and failed (**) lines (if a delivery was
570 attemtped). Previously it was only on completion (=>) lines.
572 JH/18 Authentication: the gsasl driver not provides the $authN variables in time
573 for the expansion of the server_scram_iter and server_scram_salt options.
575 WB/01 SPF: DNS lookups for the obsolete SPF RR type done by the libspf2 library
576 are now specifically given a NO_DATA response without hitting the system
577 resolver. The library goes on to do the now-standard TXT lookup.
578 Use of dnsdb lookups is not affected.
580 JH/19 Bug 2507: Modules: on handling a dynamic-module (lookups) open failure,
581 only retrieve the errormessage once. Previously two calls to dlerror()
582 were used, and the second one (for mainlog/paniclog) retrieved null
585 JH/20 Taint checking: disallow use of tainted data for
586 - the appendfile transport file and directory options
587 - the pipe transport command
588 - the autoreply transport file, log and once options
589 - file names used by the redirect router (including filter files)
591 - paths used by single-key lookups
592 Previously this was permitted.
594 JH/21 Bug 2501: Fix init call in the heimdal authenticator. Previously it
595 adjusted the size of a major service buffer; this failed because the
596 buffer was in use at the time. Change to a compile-time increase in the
597 buffer size, when this authenticator is compiled into exim.
599 JH/22 Taint-checking: move to safe-mode taint checking on all platforms. The
600 previous fast-mode was untenable in the face of glibs using mmap to
601 support larger malloc requests.
603 PP/01 Update the openssl_options possible values through OpenSSL 1.1.1c.
604 New values supported, if defined on system where compiled:
605 allow_no_dhe_kex, cryptopro_tlsext_bug, enable_middlebox_compat,
606 no_anti_replay, no_encrypt_then_mac, prioritize_chacha, tlsext_padding
608 JH/23 Performance improvement in the initial phase of a two-pass queue run. By
609 running a limited number of proceses in parallel, a benefit is gained. The
610 amount varies with the platform hardware and load. The use of the option
611 queue_run_in_order means we cannot do this, as ordering becomes
614 JH/24 Bug 2524: fix the cyrus_sasl auth driver gssapi usage. A previous fix
615 had introduced a string-copy (for ensuring NUL-termination) which was not
616 appropriate for that case, which can include embedded NUL bytes in the
617 block of data. Investigation showed the copy to actually be needless, the
618 data being length-specified.
620 JH/25 Fix use of concurrent TLS connections under GnuTLS. When a callout was
621 done during a receiving connection, and both used TLS, global info was
622 used rather than per-connection info for tracking the state of data
623 queued for transmission. This could result in a connection hang.
625 JH/26 Fix use of the SIZE parameter on MAIL commands, on continued connections.
626 Previously, when delivering serveral messages down a single connection
627 only the first would provide a SIZE. This was due to the size information
628 not being properly tracked.
630 JH/27 Bug 2530: When operating in a timezone with sub-minute offset, such as
631 TAI (at 37 seconds currently), pretend to be in UTC for time-related
632 expansion and logging. Previously, spurious values such as a future
633 minute could be seen.
635 JH/28 Bug 2533: Fix expansion of ${tr } item. When called in some situations
636 it could crash from a null-deref. This could also affect the
637 ${addresses: } operator and ${readsock } item.
639 JH/29 Bug 2537: Fix $mime_part_count. When a single connection had a non-mime
640 message following a mime one, the variable was not reset.
642 JH/30 When an pipelined-connect fails at the first response, assume incorrect
643 cached capability (perhaps the peer reneged?) and immediately retry in
646 JH/31 Fix spurious detection of timeout while writing to transport filter.
648 JH/32 Bug 2541: Fix segfault on bad cmdline -f (sender) argument. Previously
649 an attempt to copy the string was made before checking it.
651 JH/33 Fix the dsearch lookup to return an untainted result. Previously the
652 taint of the lookup key was maintained; we now regard the presence in the
653 filesystem as sufficient validation.
655 JH/34 Fix the readsocket expansion to not segfault when an empty "options"
656 argument is supplied.
658 JH/35 The dsearch lookup now requires that the directory is an absolute path.
659 Previously this was not checked, and nonempty relative paths made an
660 access under Exim's current working directory.
662 JH/36 Bug 2554: Fix msg:defer event for the hosts_max_try_hardlimit case.
663 Previously no event was raised.
665 JH/37 Bug 2552: Fix the check on spool space during reception to use the SIZE
666 parameter supplied by the sender MAIL FROM command. Previously it was
667 ignored, and only the check_spool_space option value for the required
670 JH/38 Fix $dkim_key_length. This should, after a DKIM verification, present
671 the size of the signing public-key. Previously it was instead giving
672 the size of the signature hash.
674 JH/39 DKIM verification: the RFC 8301 restriction on sizes of RSA keys is now
675 the default. See the (new) dkim_verify_min_keysizes option.
677 JH/40 Fix a memory-handling bug: when a connection carried multiple messages
678 and an ACL use a lookup for checking either the local_part or domain,
679 stale data could be accessed. Ensure that variable references are
680 dropped between messages.
682 JH/41 Bug 2571: Fix SPA authenticator. Running as a server, an offset supplied
683 by the client was not checked as pointing within response data before
684 being used. A malicious client could thus cause an out-of-bounds read and
685 possibly gain authentication. Fix by adding the check.
687 JH/42 Internationalisation: change the default for downconversion in the smtp
688 transport to be "if needed". Previously it was "as previously set" for
689 the message, which usually meant "if needed" for message-submission but
690 "no" for everything else. However, MTAs have been seen using SMTPUTF8
691 even when the envelope addresses did not need it, resulting in forwarding
692 failures to non-supporting MTAs. A downconvert in such cases will be
693 a no-op on the addresses, merely dropping the use of SMTPUTF8 by the
694 transport. The change does mean that addresses needing conversion will
695 be converted when previously a delivery failure would occur.
697 JH/43 Fix possible long line in DSN. Previously when a very long SMTP error
698 response was received it would be used unchecked in a fail-DSN, violating
699 standards on line-length limits. Truncate if needed.
701 HS/01 Remove parameters of the link to www.open-spf.org. The linked form
702 doesn't work. (Additionally add a new main config option to configure the
709 JH/01 OpenSSL: With debug enabled output keying information sufficient, server
710 side, to decode a TLS 1.3 packet capture.
712 JH/02 OpenSSL: Suppress the sending of (stateful) TLS1.3 session tickets.
713 Previously the default library behaviour applied, sending two, each in
716 JH/03 Debug output for ACL now gives the config file name and line number for
719 JH/04 The default received_header_text now uses the RFC 8314 tls cipher clause.
721 JH/05 DKIM: ensure that dkim_domain elements are lowercased before use.
723 JH/06 Fix buggy handling of autoreply bounce_return_size_limit, and a possible
724 buffer overrun for (non-chunking) other transports.
726 JH/07 GnuTLS: Our use of late (post-handshake) certificate verification, under
727 TLS1.3, means that a server rejecting a client certificate is not visible
728 to the client until the first read of encrypted data (typically the
729 response to EHLO). Add detection for that case and treat it as a failed
730 TLS connection attempt, so that the normal retry-in-clear can work (if
731 suitably configured).
733 JB/01 Bug 2375: fix expansions of 822 addresses having comments in local-part
734 and/or domain. Found and fixed by Jason Betts.
736 JH/08 Add hardening against SRV & TLSA lookups the hit CNAMEs (a nonvalid
737 configuration). If a CNAME target was not a wellformed name pattern, a
740 JH/09 Logging: Fix initial listening-on line for multiple ports for an IP when
741 the OS reports them interleaved with other addresses.
743 JH/10 OpenSSL: Fix aggregation of messages. Previously, when PIPELINING was
744 used both for input and for a verify callout, both encrypted, SMTP
745 responses being sent by the server could be lost. This resulted in
746 dropped connections and sometimes bounces generated by a peer sending
749 JH/11 Harden plaintext authenticator against a badly misconfigured client-send
750 string. Previously it was possible to cause undefined behaviour in a
751 library routine (usually a crash). Found by "zerons".
753 JH/12 Bug 2384: fix "-bP smtp_receive_timeout". Previously it returned no
756 JH/13 Bug 2386: Fix builds with Dane under LibreSSL 2.9.0 onward. Some old
757 API was removed, so update to use the newer ones.
759 JH/14 Bug 1891: Close the log file if receiving a non-smtp message, without
760 any timeout set, is taking a long time. Previously we would hang on to a
761 rotated logfile "forever" if the input was arriving with long gaps
762 (a previous attempt to fix addressed lack, for a long time, of initial
765 HS/01 Bug 2390: Use message_id for tempfile creation to avoid races in a
766 shared (NFS) environment. The length of the tempfile name is now
767 4 + 16 ("hdr.$message_exim_id") which might break on file
768 systems which restrict the file name length to lower values.
771 HS/02 Bug 2390: Use message_id for tempfile creation to avoid races in a
772 shared (NFS) environment.
774 HS/03 Bug 2392: exigrep does case sensitive *option* processing (as it
775 did for all versions <4.90). Notably -M, -m, --invert, -I may be
778 JH/15 Use unsigned when creating bitmasks in macros, to avoid build errors
779 on some platforms for bit 31.
781 JH/16 GnuTLS: rework ciphersuite strings under recent library versions. Thanks
782 to changes apparently associated with TLS1.3 handling some of the APIs
783 previously used were either nonfunctional or inappropriate. Strings
784 like TLS1.3:ECDHE_SECP256R1__RSA_PSS_RSAE_SHA256__AES_256_GCM__AEAD:256
785 and TLS1.2:ECDHE_SECP256R1__RSA_SHA256__AES_128_CBC__SHA256:128 replace
786 the previous TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256 .
787 This affects log line X= elements, the $tls_{in,out}_cipher variables,
788 and the use of specific cipher names in the encrypted= ACL condition.
790 JH/17 OpenSSL: the default openssl_options now disables ssl_v3.
792 JH/18 GnuTLS: fix $tls_out_ocsp under hosts_request_ocsp. Previously the
793 verification result was not updated unless hosts_require_ocsp applied.
795 JH/19 Bug 2398: fix listing of a named-queue. Previously, even with the option
796 queue_list_requires_admin set to false, non-admin users were denied the
799 JH/20 Bug 2389: fix server advertising of usable certificates, under GnuTLS in
800 directory-of-certs mode. Previously they were advertised despite the
803 JH/21 The smtp transport option "hosts_noproxy_tls" is now unset by default.
804 A single TCP connection by a client will now hold a TLS connection open
805 for multiple message deliveries, by default. Previously the default was to
808 JH/22 The smtp transport option "hosts_try_dane" now enables all hosts by
809 default. If built with the facility, DANE will be used. The facility
810 SUPPORT_DANE is now enabled in the prototype build Makefile "EDITME".
812 JH/23 The build default is now for TLS to be included; the SUPPORT_TLS define
813 is replaced with DISABLE_TLS. Either USE_GNUTLS or (the new) USE_OPENSSL
814 must be defined and you must still, unless you define DISABLE_TLS, manage
815 the the include-dir and library-file requirements that go with that
816 choice. Non-TLS builds are still supported.
818 JH/24 Fix duplicated logging of peer name/address, on a transport connection-
821 JH/25 The smtp transport option "hosts_try_fastopen" now enables all hosts by
822 default. If the platform supports and has the facility enabled, it will
823 be requested on all coneections.
825 JH/26 The PIPE_CONNECT facility is promoted from experimental status and is now
826 controlled by the build-time option SUPPORT_PIPE_CONNECT.
828 PP/01 Unbreak heimdal_gssapi, broken in 4.92.
830 JH/27 Bug 2404: Use the main-section configuration option "dsn_from" for
831 success-DSN messages. Previously the From: header was always the default
832 one for these; the option was ignored.
834 JH/28 Fix the timeout on smtp response to apply to the whole response.
835 Previously it was reset for every read, so a teergrubing peer sending
836 single bytes within the time limit could extend the connection for a
837 long time. Credit to Qualsys Security Advisory Team for the discovery.
839 JH/29 Fix DSN Final-Recipient: field. Previously it was the post-routing
840 delivery address, which leaked information of the results of local
841 forwarding. Change to the original envelope recipient address, per
844 JH/30 Bug 2411: Fix DSN generation when RFC 3461 failure notification is
845 requested. Previously not bounce was generated and a log entry of
846 error ignored was made.
848 JH/31 Avoid re-expansion in ${sort } expansion. (CVE-2019-13917)
850 JH/32 Introduce a general tainting mechanism for values read from the input
851 channel, and values derived from them. Refuse to expand any tainted
852 values, to catch one form of exploit.
854 JH/33 Bug 2413: Fix dkim_strict option. Previously the expansion result
855 was unused and the unexpanded text used for the test. Found and
856 fixed by Ruben Jenster.
858 JH/34 Fix crash after TLS shutdown. When the TCP/SMTP channel was left open,
859 an attempt to use a TLS library read routine dereffed a nul pointer,
862 JH/35 Bug 2409: filter out-of-spec chars from callout response before using
863 them in our smtp response.
865 JH/36 Have the general router option retry_use_local_part default to true when
866 any of the restrictive preconditions are set (to anything). Previously it
867 was only for check_local user. The change removes one item of manual
868 configuration which is required for proper retries when a remote router
869 handles a subset of addresses for a domain.
871 JH/37 Appendfile: when evaluating quota use (non-quota_size_regex) take the file
872 link count into consideration.
874 HS/04 Fix handling of very log lines in -H files. If a -<key> <value> line
875 caused the extension of big_buffer, the following lines were ignored.
877 JH/38 Bug 1395: Teach the DNS negative-cache about TTL value from the SOA in
878 accordance with RFC 2308. Previously there was no expiry, so a longlived
879 receive process (eg. due to ACL delays) versus a short SOA value could
882 HS/05 Handle trailing backslash gracefully. (CVE-2019-15846)
884 JH/39 Promote DMARC support to mainline.
886 JH/40 Bug 2452: Add a References: header to DSNs.
888 JH/41 With GnuTLS 3.6.0 (and later) do not attempt to manage Diffie-Hellman
889 parameters. The relevant library call is documented as "Deprecated: This
890 function is unnecessary and discouraged on GnuTLS 3.6.0 or later. Since
891 3.6.0, DH parameters are negotiated following RFC7919."
893 HS/06 Change the default of dnssec_request_domains to "*"
895 JH/42 Bug 2545: Fix CHUNKING for all RCPT commands rejected. Previously we
896 carried on and emitted a BDAT command, even when PIPELINING was not
899 JH/43 Bug 2465: Fix taint-handling in dsearch lookup. Previously a nontainted
900 buffer was used for the filename, resulting in a trap when tainted
901 arguments (eg. $domain) were used.
903 JH/44 With OpenSSL 1.1.1 (onwards) disable renegotiation for TLS1.2 and below;
904 recommended to avoid a possible server-load attack. The feature can be
905 re-enabled via the openssl_options main cofiguration option.
907 JH/45 local_scan API: documented the current smtp_printf() call. This changed
908 for version 4.90 - adding a "more data" boolean to the arguments.
909 Bumped the ABI version number also, this having been missed previously;
910 release versions 4.90 to 4.92.3 inclusive were effectively broken in
911 respect of usage of smtp_printf() by either local_scan code or libraries
912 accessed via the ${dlfunc } expansion item. Both will need coding
913 adjustment for any calls to smtp_printf() to match the new function
914 signature; a FALSE value for the new argument is always safe.
916 JH/46 FreeBSD: fix use of the sendfile() syscall. The shim was not updating
917 the file-offset (which the Linux syscall does, and exim expects); this
918 resulted in an indefinite loop.
920 JH/47 ARC: fix crash in signing, triggered when a configuration error failed
921 to do ARC verification. The Authentication-Results: header line added
922 by the configuration then had no ARC item.
924 JH/48 Bug 2784: fix shutdown=no in the ${readsocket) expansion item. Previously
925 an incorrect mode was used for reading the result, resulting in it being
932 JH/01 Remove code calling the customisable local_scan function, unless a new
933 definition "HAVE_LOCAL_SCAN=yes" is present in the Local/Makefile.
935 JH/02 Bug 1007: Avoid doing logging from signal-handlers, as that can result in
936 non-signal-safe functions being used.
938 JH/03 Bug 2269: When presented with a received message having a stupidly large
939 number of DKIM-Signature headers, disable DKIM verification to avoid
940 a resource-consumption attack. The limit is set at twenty.
942 JH/04 Add variables $arc_domains, $arc_oldest_pass for ARC verify. Fix the
943 report of oldest_pass in ${authres } in consequence, and separate out
944 some descriptions of reasons for verification fail.
946 JH/05 Bug 2273: Cutthrough delivery left a window where the received messsage
947 files in the spool were present and unlocked. A queue-runner could spot
948 them, resulting in a duplicate delivery. Fix that by doing the unlock
949 after the unlink. Investigation by Tim Stewart. Take the opportunity to
950 add more error-checking on spoolfile handling while that code is being
953 PP/01 Refuse to open a spool data file (*-D) if it's a symlink.
954 No known attacks, no CVE, this is defensive hardening.
956 JH/06 Bug 2275: The MIME ACL unlocked the received message files early, and
957 a queue-runner could start a delivery while other operations were ongoing.
958 Cutthrough delivery was a common victim, resulting in duplicate delivery.
959 Found and investigated by Tim Stewart. Fix by using the open message data
960 file handle rather than opening another, and not locally closing it (which
961 releases a lock) for that case, while creating the temporary .eml format
962 file for the MIME ACL. Also applies to "regex" and "spam" ACL conditions.
964 JH/07 Bug 177: Make a random-recipient callout success visible in ACL, by setting
965 $sender_verify_failure/$recipient_verify_failure to "random".
967 JH/08 When generating a selfsigned cert, use serial number 1 since zero is not
970 JH/09 Bug 2274: Fix logging of cmdline args when starting in an unlinked cwd.
971 Previously this would segfault.
973 JH/10 Fix ARC signing for case when DKIM signing failed. Previously this would
976 JH/11 Bug 2264: Exim now only follows CNAME chains one step by default. We'd
977 like zero, since the resolver should be doing this for us, But we need one
978 as a CNAME but no MX presence gets the CNAME returned; we need to check
979 that doesn't point to an MX to declare it "no MX returned" rather than
980 "error, loop". A new main option is added so the older capability of
981 following some limited number of chain links is maintained.
983 JH/12 Add client-ip info to non-pass iprev ${authres } lines.
985 JH/13 For receent Openssl versions (1.1 onward) use modern generic protocol
986 methods. These should support TLS 1.3; they arrived with TLS 1.3 and the
987 now-deprecated earlier definitions used only specified the range up to TLS
988 1.2 (in the older-version library docs).
990 JH/14 Bug 2284: Fix DKIM signing for body lines starting with a pair of dots.
992 JH/15 Rework TLS client-side context management. Stop using a global, and
993 explicitly pass a context around. This enables future use of TLS for
994 connections to service-daemons (eg. malware scanning) while a client smtp
995 connection is using TLS; with cutthrough connections this is quite likely.
997 JH/16 Fix ARC verification to do AS checks in reverse order.
999 JH/17 Support a "tls" option on the ${readsocket } expansion item.
1001 JH/18 Bug 2287: Fix the protocol name (eg utf8esmtp) for multiple messages
1002 using the SMTPUTF8 option on their MAIL FROM commands, in one connection.
1003 Previously the "utf8" would be re-prepended for every additional message.
1005 JH/19 Reject MAIL FROM commands with SMTPUTF8 when the facility was not advertised.
1006 Previously thery were accepted, resulting in issues when attempting to
1007 forward messages to a non-supporting MTA.
1009 PP/02 Let -n work with printing macros too, not just options.
1011 JH/20 Bug 2296: Fix cutthrough for >1 address redirection. Previously only
1012 one parent address was copied, and bogus data was used at delivery-logging
1013 time. Either a crash (after delivery) or bogus log data could result.
1014 Discovery and analysis by Tim Stewart.
1016 PP/03 Make ${utf8clean:} expansion operator detect incomplete final character.
1017 Previously if the string ended mid-character, we did not insert the
1018 promised '?' replacement.
1020 PP/04 Documentation: current string operators work on bytes, not codepoints.
1022 JH/21 Change as many as possible of the global flags into one-bit bitfields; these
1023 should pack well giving a smaller memory footprint so better caching and
1024 therefore performance. Group the declarations where this can't be done so
1025 that the byte-sized flag variables are not interspersed among pointer
1026 variables, giving a better chance of good packing by the compiler.
1028 JH/22 Bug 1896: Fix the envelope from for DMARC forensic reports to be possibly
1029 non-null, to avoid issues with sites running BATV. Previously reports were
1030 sent with an empty envelope sender so looked like bounces.
1032 JH/23 Bug 2318: Fix the noerror command within filters. It wasn't working.
1033 The ignore_error flag wasn't being returned from the filter subprocess so
1034 was not set for later routers. Investigation and fix by Matthias Kurz.
1036 JH/24 Bug 2310: Raise a msg:fail:internal event for each undelivered recipient,
1037 and a msg:complete for the whole, when a message is manually removed using
1038 -Mrm. Developement by Matthias Kurz, hacked on by JH.
1040 JH/25 Avoid fixed-size buffers for pathnames in DB access. This required using
1041 a "Gnu special" function, asprintf() in the DB utility binary builds; I
1042 hope that is portable enough.
1044 JH/26 Bug 2311: Fix DANE-TA verification under GnuTLS. Previously it was also
1045 requiring a known-CA anchor certificate; make it now rely entirely on the
1046 TLSA as an anchor. Checking the name on the leaf cert against the name
1047 on the A-record for the host is still done for TA (but not for EE mode).
1049 JH/27 Fix logging of proxy address. Previously, a pointless "PRX=[]:0" would be
1050 included in delivery lines for non-proxied connections, when compiled with
1051 SUPPORT_SOCKS and running with proxy logging enabled.
1053 JH/28 Bug 2314: Fire msg:fail:delivery event even when error is being ignored.
1054 Developement by Matthias Kurz, tweaked by JH. While in that bit of code,
1055 move the existing event to fire before the normal logging of message
1056 failure so that custom logging is bracketed by normal logging.
1058 JH/29 Bug 2322: A "fail" command in a non-system filter (file) now fires the
1059 msg:fail:internal event. Developement by Matthias Kurz.
1061 JH/30 Bug 2329: Increase buffer size used for dns lookup from 2k, which was
1062 far too small for todays use of crypto signatures stored there. Go all
1063 the way to the max DNS message size of 64kB, even though this might be
1064 overmuch for IOT constrained device use.
1066 JH/31 Fix a bad use of a copy function, which could be used to pointlessly
1067 copy a string over itself. The library routine is documented as not
1068 supporting overlapping copies, and on MacOS it actually raised a SIGABRT.
1070 JH/32 For main options check_spool_space and check_inode_space, where the
1071 platform supports 64b integers, support more than the previous 2^31 kB
1072 (i.e. more than 2 TB). Accept E, P and T multipliers in addition to
1073 the previous G, M, k.
1075 JH/33 Bug 2338: Fix the cyrus-sasl authenticator to fill in the
1076 $authenticated_fail_id variable on authentication failure. Previously
1079 JH/34 Increase RSA keysize of autogen selfsign cert from 1024 to 2048. RHEL 8.0
1080 OpenSSL didn't want to use such a weak key. Do for GnuTLS also, and for
1081 more-modern GnuTLS move from GNUTLS_SEC_PARAM_LOW to
1082 GNUTLS_SEC_PARAM_MEDIUM.
1084 JH/35 OpenSSL: fail the handshake when SNI processing hits a problem, server
1085 side. Previously we would continue as if no SNI had been received.
1087 JH/36 Harden the handling of string-lists. When a list consisted of a sole
1088 "<" character, which should be a list-separator specification, we walked
1089 off past the nul-terimation.
1091 JH/37 Bug 2341: Send "message delayed" warning MDNs (restricted to external
1092 causes) even when the retry time is not yet met. Previously they were
1093 not, meaning that when (say) an account was over-quota and temp-rejecting,
1094 and multiple senders' messages were queued, only one sender would get
1095 notified on each configured delay_warning cycle.
1097 JH/38 Bug 2351: Log failures to extract envelope addresses from message headers.
1099 JH/39 OpenSSL: clear the error stack after an SSL_accept(). With anon-auth
1100 cipher-suites, an error can be left on the stack even for a succeeding
1101 accept; this results in impossible error messages when a later operation
1104 AM/01 Bug 2359: GnuTLS: repeat lowlevel read and write operations while they
1105 return error codes indicating retry. Under TLS1.3 this becomes required.
1107 JH/40 Fix the feature-cache refresh for EXPERIMENTAL_PIPE_CONNECT. Previously
1108 it only wrote the new authenticators, resulting in a lack of tracking of
1109 peer changes of ESMTP extensions until the next cache flush.
1111 JH/41 Fix the loop reading a message header line to check for integer overflow,
1112 and more-often against header_maxsize. Previously a crafted message could
1113 induce a crash of the recive process; now the message is cleanly rejected.
1115 JH/42 Bug 2366: Fix the behaviour of the dkim_verify_signers option. It had
1116 been totally disabled for all of 4.91. Discovery and fix by "Mad Alex".
1122 GF/01 DEFER rather than ERROR on redis cluster MOVED response.
1123 When redis_servers is set to a list of > 1 element, and the Redis servers
1124 in that list are in cluster configuration, convert the REDIS_REPLY_ERROR
1125 case of MOVED into a DEFER case instead, thus moving the query onto the
1126 next server in the list. For a cluster of N elements, all N servers must
1127 be defined in redis_servers.
1129 GF/02 Catch and remove uninitialized value warning in exiqsumm
1130 Check for existence of @ARGV before looking at $ARGV[0]
1132 JH/01 Replace the store_release() internal interface with store_newblock(),
1133 which internalises the check required to safely use the old one, plus
1134 the allocate and data copy operations duplicated in both (!) of the
1135 extant use locations.
1137 JH/02 Disallow '/' characters in queue names specified for the "queue=" ACL
1138 modifier. This matches the restriction on the commandline.
1140 JH/03 Fix pgsql lookup for multiple result-tuples with a single column.
1141 Previously only the last row was returned.
1143 JH/04 Bug 2217: Tighten up the parsing of DKIM signature headers. Previously
1144 we assumed that tags in the header were well-formed, and parsed the
1145 element content after inspecting only the first char of the tag.
1146 Assumptions at that stage could crash the receive process on malformed
1149 JH/05 Bug 2215: Fix crash associated with dnsdb lookup done from DKIM ACL.
1150 While running the DKIM ACL we operate on the Permanent memory pool so that
1151 variables created with "set" persist to the DATA ACL. Also (at any time)
1152 DNS lookups that fail create cache records using the Permanent pool. But
1153 expansions release any allocations made on the current pool - so a dnsdb
1154 lookup expansion done in the DKIM ACL releases the memory used for the
1155 DNS negative-cache, and bad things result. Solution is to switch to the
1156 Main pool for expansions.
1157 While we're in that code, add checks on the DNS cache during store_reset,
1158 active in the testsuite.
1159 Problem spotted, and debugging aided, by Wolfgang Breyha.
1161 JH/06 Fix issue with continued-connections when the DNS shifts unreliably.
1162 When none of the hosts presented to a transport match an already-open
1163 connection, close it and proceed with the list. Previously we would
1164 queue the message. Spotted by Lena with Yahoo, probably involving
1167 JH/07 Bug 2214: Fix SMTP responses resulting from non-accept result of MIME ACL.
1168 Previously a spurious "250 OK id=" response was appended to the proper
1171 JH/08 The "support for" informational output now, which built with Content
1172 Scanning support, has a line for the malware scanner interfaces compiled
1173 in. Interface can be individually included or not at build time.
1175 JH/09 The "aveserver", "kavdaemon" and "mksd" interfaces are now not included
1176 by the template makefile "src/EDITME". The "STREAM" support for an older
1177 ClamAV interface method is removed.
1179 JH/10 Bug 2223: Fix mysql lookup returns for the no-data case (when the number of
1180 rows affected is given instead).
1182 JH/11 The runtime Berkeley DB library version is now additionally output by
1183 "exim -d -bV". Previously only the compile-time version was shown.
1185 JH/12 Bug 2230: Fix cutthrough routing for nonfirst messages in an initiating
1186 SMTP connection. Previously, when one had more recipients than the
1187 first, an abortive onward connection was made. Move to full support for
1188 multiple onward connections in sequence, handling cutthrough connection
1189 for all multi-message initiating connections.
1191 JH/13 Bug 2229: Fix cutthrough routing for nonstandard port numbers defined by
1192 routers. Previously, a multi-recipient message would fail to match the
1193 onward-connection opened for the first recipient, and cause its closure.
1195 JH/14 Bug 2174: A timeout on connect for a callout was also erroneously seen as
1196 a timeout on read on a GnuTLS initiating connection, resulting in the
1197 initiating connection being dropped. This mattered most when the callout
1198 was marked defer_ok. Fix to keep the two timeout-detection methods
1201 JH/15 Relax results from ACL control request to enable cutthrough, in
1202 unsupported situations, from error to silently (except under debug)
1203 ignoring. This covers use with PRDR, frozen messages, queue-only and
1206 HS/01 Fix Buffer overflow in base64d() (CVE-2018-6789)
1208 JH/16 Fix bug in DKIM verify: a buffer overflow could corrupt the malloc
1209 metadata, resulting in a crash in free().
1211 PP/01 Fix broken Heimdal GSSAPI authenticator integration.
1212 Broken in f2ed27cf5, missing an equals sign for specified-initialisers.
1213 Broken also in d185889f4, with init system revamp.
1215 JH/17 Bug 2113: Fix conversation closedown with the Avast malware scanner.
1216 Previously we abruptly closed the connection after reading a malware-
1217 found indication; now we go on to read the "scan ok" response line,
1220 JH/18 Bug 2239: Enforce non-usability of control=utf8_downconvert in the mail
1221 ACL. Previously, a crash would result.
1223 JH/19 Speed up macro lookups during configuration file read, by skipping non-
1224 macro text after a replacement (previously it was only once per line) and
1225 by skipping builtin macros when searching for an uppercase lead character.
1227 JH/20 DANE support moved from Experimental to mainline. The Makefile control
1228 for the build is renamed.
1230 JH/21 Fix memory leak during multi-message connections using STARTTLS. A buffer
1231 was allocated for every new TLS startup, meaning one per message. Fix
1232 by only allocating once (OpenSSL) or freeing on TLS-close (GnuTLS).
1234 JH/22 Bug 2236: When a DKIM verification result is overridden by ACL, DMARC
1235 reported the original. Fix to report (as far as possible) the ACL
1236 result replacing the original.
1238 JH/23 Fix memory leak during multi-message connections using STARTTLS under
1239 OpenSSL. Certificate information is loaded for every new TLS startup,
1240 and the resources needed to be freed.
1242 JH/24 Bug 2242: Fix exim_dbmbuild to permit directoryless filenames.
1244 JH/25 Fix utf8_downconvert propagation through a redirect router. Previously it
1247 JH/26 Bug 2253: For logging delivery lines under PRDR, append the overall
1248 DATA response info to the (existing) per-recipient response info for
1249 the "C=" log element. It can have useful tracking info from the
1250 destination system. Patch from Simon Arlott.
1252 JH/27 Bug 2251: Fix ldap lookups that return a single attribute having zero-
1253 length value. Previously this would segfault.
1255 HS/02 Support Avast multiline protoocol, this allows passing flags to
1256 newer versions of the scanner.
1258 JH/28 Ensure that variables possibly set during message acceptance are marked
1259 dead before release of memory in the daemon loop. This stops complaints
1260 about them when the debug_store option is enabled. Discovered specifically
1261 for sender_rate_period, but applies to a whole set of variables.
1262 Do the same for the queue-runner and queue-list loops, for variables set
1263 from spool message files. Do the same for the SMTP per-message loop, for
1264 certain variables indirectly set in ACL operations.
1266 JH/29 Bug 2250: Fix a longstanding bug in heavily-pipelined SMTP input (such
1267 as a multi-recipient message from a mailinglist manager). The coding had
1268 an arbitrary cutoff number of characters while checking for more input;
1269 enforced by writing a NUL into the buffer. This corrupted long / fast
1270 input. The problem was exposed more widely when more pipelineing of SMTP
1271 responses was introduced, and one Exim system was feeding another.
1272 The symptom is log complaints of SMTP syntax error (NUL chars) on the
1273 receiving system, and refused recipients seen by the sending system
1274 (propating to people being dropped from mailing lists).
1275 Discovered and pinpointed by David Carter.
1277 JH/30 The (EXPERIMENTAL_DMARC) variable $dmarc_ar_header is withdrawn, being
1278 replaced by the ${authresults } expansion.
1280 JH/31 Bug 2257: Fix pipe transport to not use a socket-only syscall.
1282 HS/03 Set a handler for SIGTERM and call exit(3) if running as PID 1. This
1283 allows proper process termination in container environments.
1285 JH/32 Bug 2258: Fix spool_wireformat in combination with LMTP transport.
1286 Previously the "final dot" had a newline after it; ensure it is CR,LF.
1288 JH/33 SPF: remove support for the "spf" ACL condition outcome values "err_temp"
1289 and "err_perm", deprecated since 4.83 when the RFC-defined words
1290 "temperror" and "permerror" were introduced.
1292 JH/34 Re-introduce enforcement of no cutthrough delivery on transports having
1293 transport-filters or DKIM-signing. The restriction was lost in the
1294 consolidation of verify-callout and delivery SMTP handling.
1295 Extend the restriction to also cover ARC-signing.
1297 JH/35 Cutthrough: for a final-dot response timeout (and nonunderstood responses)
1298 in defer=pass mode supply a 450 to the initiator. Previously the message
1301 PP/02 DANE: add dane_require_tls_ciphers SMTP Transport option; if unset,
1302 tls_require_ciphers is used as before.
1304 HS/03 Malware Avast: Better match the Avast multiline protocol. Add
1305 "pass_unscanned". Only tmpfails from the scanner are written to
1306 the paniclog, as they may require admin intervention (permission
1307 denied, license issues). Other scanner errors (like decompression
1308 bombs) do not cause a paniclog entry.
1310 JH/36 Fix reinitialisation of DKIM logging variable between messages.
1311 Previously it was possible to log spurious information in receive log
1314 JH/37 Bug 2255: Revert the disable of the OpenSSL session caching. This
1315 triggered odd behaviour from Outlook Express clients.
1317 PP/03 Add util/renew-opendmarc-tlds.sh script for safe renewal of public
1320 JH/38 DKIM: accept Ed25519 pubkeys in SubjectPublicKeyInfo-wrapped form,
1321 since the IETF WG has not yet settled on that versus the original
1322 "bare" representation.
1324 JH/39 Fix syslog logging for syslog_timestamp=no and log_selector +millisec.
1325 Previously the millisecond value corrupted the output.
1326 Fix also for syslog_pid=no and log_selector +pid, for which the pid
1327 corrupted the output.
1333 JH/01 Rework error string handling in TLS interface so that the caller in
1334 more cases is responsible for logging. This permits library-sourced
1335 string to be attached to addresses during delivery, and collapses
1336 pairs of long lines into single ones.
1338 PP/01 Allow PKG_CONFIG_PATH to be set in Local/Makefile and use it correctly
1339 during configuration. Wildcards are allowed and expanded.
1341 JH/02 Rework error string handling in DKIM to pass more info back to callers.
1342 This permits better logging.
1344 JH/03 Rework the transport continued-connection mechanism: when TLS is active,
1345 do not close it down and have the child transport start it up again on
1346 the passed-on TCP connection. Instead, proxy the child (and any
1347 subsequent ones) for TLS via a unix-domain socket channel. Logging is
1348 affected: the continued delivery log lines do not have any DNSSEC, TLS
1349 Certificate or OCSP information. TLS cipher information is still logged.
1351 JH/04 Shorten the log line for daemon startup by collapsing adjacent sets of
1352 identical IP addresses on different listening ports. Will also affect
1355 PP/02 Bug 2070: uClibc defines __GLIBC__ without providing glibc headers;
1356 add noisy ifdef guards to special-case this sillyness.
1357 Patch from Bernd Kuhls.
1359 JH/05 Tighten up the checking in isip4 (et al): dotted-quad components larger
1360 than 255 are no longer allowed.
1362 JH/06 Default openssl_options to include +no_ticket, to reduce load on peers.
1363 Disable the session-cache too, which might reduce our load. Since we
1364 currrectly use a new context for every connection, both as server and
1365 client, there is no benefit for these.
1366 GnuTLS appears to not support tickets server-side by default (we don't
1367 call gnutls_session_ticket_enable_server()) but client side is enabled
1368 by default on recent versions (3.1.3 +) unless the PFS priority string
1371 PP/03 Add $SOURCE_DATE_EPOCH support for reproducible builds, per spec at
1372 <https://reproducible-builds.org/specs/source-date-epoch/>.
1374 JH/07 Fix smtp transport use of limited max_rcpt under mua_wrapper. Previously
1375 the check for any unsuccessful recipients did not notice the limit, and
1376 erroneously found still-pending ones.
1378 JH/08 Pipeline CHUNKING command and data together, on kernels that support
1379 MSG_MORE. Only in-clear (not on TLS connections).
1381 JH/09 Avoid using a temporary file during transport using dkim. Unless a
1382 transport-filter is involved we can buffer the headers in memory for
1383 creating the signature, and read the spool data file once for the
1384 signature and again for transmission.
1386 JH/10 Enable use of sendfile in Linux builds as default. It was disabled in
1387 4.77 as the kernel support then wasn't solid, having issues in 64bit
1388 mode. Now, it's been long enough. Add support for FreeBSD also.
1390 JH/11 Bug 2104: Fix continued use of a transport connection with TLS. In the
1391 case where the routing stage had gathered several addresses to send to
1392 a host before calling the transport for the first, we previously failed
1393 to close down TLS in the old transport process before passing the TCP
1394 connection to the new process. The new one sent a STARTTLS command
1395 which naturally failed, giving a failed delivery and bloating the retry
1396 database. Investigation and fix prototype from Wolfgang Breyha.
1398 JH/12 Fix check on SMTP command input synchronisation. Previously there were
1399 false-negatives in the check that the sender had not preempted a response
1400 or prompt from Exim (running as a server), due to that code's lack of
1401 awareness of the SMTP input buffering.
1403 PP/04 Add commandline_checks_require_admin option.
1404 Exim drops privileges sanely, various checks such as -be aren't a
1405 security problem, as long as you trust local users with access to their
1406 own account. When invoked by services which pass untrusted data to
1407 Exim, this might be an issue. Set this option in main configuration
1408 AND make fixes to the calling application, such as using `--` to stop
1411 JH/13 Do pipelining under TLS. Previously, although safe, no advantage was
1412 taken. Now take care to pack both (client) MAIL,RCPT,DATA, and (server)
1413 responses to those, into a single TLS record each way (this usually means
1414 a single packet). As a side issue, smtp_enforce_sync now works on TLS
1417 PP/05 OpenSSL/1.1: use DH_bits() for more accurate DH param sizes. This
1418 affects you only if you're dancing at the edge of the param size limits.
1419 If you are, and this message makes sense to you, then: raise the
1420 configured limit or use OpenSSL 1.1. Nothing we can do for older
1423 JH/14 For the "sock" variant of the malware scanner interface, accept an empty
1424 cmdline element to get the documented default one. Previously it was
1427 JH/15 Fix a crash in the smtp transport caused when two hosts in succession
1428 are unsuable for non-message-specific reasons - eg. connection timeout,
1429 banner-time rejection.
1431 JH/16 Fix logging of delivery remote port, when specified by router, under
1434 PP/06 Repair manualroute's ability to take options in any order, even if one
1435 is the name of a transport.
1438 HS/01 Cleanup, prevent repeated use of -p/-oMr (CVE-2017-1000369)
1440 JH/17 Change the list-building routines interface to use the expanding-string
1441 triplet model, for better allocation and copying behaviour.
1443 JH/18 Prebuild the data-structure for "builtin" macros, for faster startup.
1444 Previously it was constructed the first time a possibly-matching string
1445 was met in the configuration file input during startup; now it is done
1448 JH/19 Bug 2141: Use the full-complex API for Berkeley DB rather than the legacy-
1449 compatible one, to avoid the (poorly documented) possibility of a config
1450 file in the working directory redirecting the DB files, possibly correpting
1451 some existing file. CVE-2017-10140 assigned for BDB.
1453 JH/20 Bug 2147: Do not defer for a verify-with-callout-and-random which is not
1454 cache-hot. Previously, although the result was properly cached, the
1455 initial verify call returned a defer.
1457 JH/21 Bug 2151: Avoid using SIZE on the MAIL for a callout verify, on any but
1458 the main verify for receipient in uncached-mode.
1460 JH/22 Retire historical build files to an "unsupported" subdir. These are
1461 defined as "ones for which we have no current evidence of testing".
1463 JH/23 DKIM: enforce the DNS pubkey record "h" permitted-hashes optional field,
1464 if present. Previously it was ignored.
1466 JH/24 Start using specified-initialisers in C structure init coding. This is
1467 a C99 feature (it's 2017, so now considered safe).
1469 JH/25 Use one-bit bitfields for flags in the "addr" data structure. Previously
1470 if was a fixed-sized field and bitmask ops via macros; it is now more
1473 PP/07 GitHub PR 56: Apply MariaDB build fix.
1474 Patch provided by Jaroslav Škarvada.
1476 PP/08 Bug 2161: Fix regression in sieve quoted-printable handling introduced
1477 during Coverity cleanups [4.87 JH/47]
1478 Diagnosis and fix provided by Michael Fischer v. Mollard.
1480 JH/26 Fix DKIM bug: when the pseudoheader generated for signing was exactly
1481 the right size to place the terminating semicolon on its own folded
1482 line, the header hash was calculated to an incorrect value thanks to
1483 the (relaxed) space the fold became.
1485 HS/02 Fix Bug 2130: large writes from the transport subprocess were chunked
1486 and confused the parent.
1488 JH/27 Fix SOCKS bug: an unitialized pointer was deref'd by the transport process
1489 which could crash as a result. This could lead to undeliverable messages.
1491 JH/28 Logging: "next input sent too soon" now shows where input was truncated
1494 JH/29 Fix queue_run_in_order to ignore the PID portion of the message ID. This
1495 matters on fast-turnover and PID-randomising systems, which were getting
1496 out-of-order delivery.
1498 JH/30 Fix a logging bug on aarch64: an unsafe routine was previously used for
1499 a possibly-overlapping copy. The symptom was that "Remote host closed
1500 connection in response to HELO" was logged instead of the actual 4xx
1503 JH/31 Fix CHUNKING code to properly flush the unwanted chunk after an error.
1504 Previously only that bufferd was discarded, resulting in SYMTP command
1507 JH/32 DKIM: when a message has multiple signatures matching an identity given
1508 in dkim_verify_signers, run the dkim acl once for each. Previously only
1509 one run was done. Bug 2189.
1511 JH/33 Downgrade an unfound-list name (usually a typo in the config file) from
1512 "panic the current process" to "deliberately defer". The panic log is
1513 still written with the problem list name; the mail and reject logs now
1514 get a temp-reject line for the message that was being handled, saying
1515 something like "domains check lookup or other defer". The SMTP 451
1516 message is still "Temporary local problem".
1518 JH/34 Bug 2199: Fix a use-after-free while reading smtp input for header lines.
1519 A crafted sequence of BDAT commands could result in in-use memory beeing
1520 freed. CVE-2017-16943.
1522 HS/03 Bug 2201: Fix checking for leading-dot on a line during headers reading
1523 from SMTP input. Previously it was always done; now only done for DATA
1524 and not BDAT commands. CVE-2017-16944.
1526 JH/35 Bug 2201: Flush received data in BDAT mode after detecting an error fatal
1527 to the message (such as an overlong header line). Previously this was
1528 not done and we did not exit BDAT mode. Followon from the previous item
1529 though a different problem.
1535 JH/01 Bug 1922: Support IDNA2008. This has slightly different conversion rules
1536 than -2003 did; needs libidn2 in addition to libidn.
1538 JH/02 The path option on a pipe transport is now expanded before use.
1540 PP/01 GitHub PR 50: Do not call ldap_start_tls_s on ldapi:// connections.
1541 Patch provided by "Björn", documentation fix added too.
1543 JH/03 Bug 2003: fix Proxy Protocol v2 handling: the address size field was
1544 missing a wire-to-host endian conversion.
1546 JH/04 Bug 2004: fix CHUNKING in non-PIPELINEING mode. Chunk data following
1547 close after a BDAT command line could be taken as a following command,
1548 giving a synch failure. Fix by only checking for synch immediately
1549 before acknowledging the chunk.
1551 PP/02 GitHub PR 52: many spelling fixes, which include fixing parsing of
1552 no_require_dnssec option and creation of _HAVE_TRANSPORT_APPEND_MAILDIR
1553 macro. Patches provided by Josh Soref.
1555 JH/05 Have the EHLO response advertise VRFY, if there is a vrfy ACL defined.
1556 Previously we did not; the RFC seems ambiguous and VRFY is not listed
1557 by IANA as a service extension. However, John Klensin suggests that we
1560 JH/06 Bug 2017: Fix DKIM verification in -bh test mode. The data feed into
1561 the dkim code may be unix-mode line endings rather than smtp wire-format
1562 CRLF, so prepend a CR to any bare LF.
1564 JH/07 Rationalise the coding for callout smtp conversations and transport ones.
1565 As a side-benfit, callouts can now use PIPELINING hence fewer round-trips.
1567 JH/08 Bug 2016: Fix DKIM verification vs. CHUNKING. Any BDAT commands after
1568 the first were themselves being wrongly included in the feed into dkim
1569 processing; with most chunk sizes in use this resulted in an incorrect
1570 body hash calculated value.
1572 JH/09 Bug 2014: permit inclusion of a DKIM-Signature header in a received
1573 DKIM signature block, for verification. Although advised against by
1574 standards it is specifically not ruled illegal.
1576 JH/10 Bug 2025: Fix reception of (quoted) local-parts with embedded spaces.
1578 JH/11 Bug 2029: Fix crash in DKIM verification when a message signature block is
1579 missing a body hash (the bh= tag).
1581 JH/12 Bug 2018: Re-order Proxy Protocol startup versus TLS-on-connect startup.
1582 It seems that HAProxy sends the Proxy Protocol information in clear and
1583 only then does a TLS startup, so do the same.
1585 JH/13 Bug 2027: Avoid attempting to use TCP Fast Open for non-transport client
1586 TCP connections (such as for Spamd) unless the daemon successfully set
1587 Fast Open mode on its listening sockets. This fixes breakage seen on
1588 too-old kernels or those not configured for Fast Open, at the cost of
1589 requiring both directions being enabled for TFO, and TFO never being used
1590 by non-daemon-related Exim processes.
1592 JH/14 Bug 2000: Reject messages recieved with CHUNKING but with malformed line
1593 endings, at least on the first header line. Try to canonify any that get
1594 past that check, despite the cost.
1596 JH/15 Angle-bracket nesting (an error inserted by broken sendmails) levels are
1597 now limited to an arbitrary five deep, while parsing addresses with the
1598 strip_excess_angle_brackets option enabled.
1600 PP/03 Bug 2018: For Proxy Protocol and TLS-on-connect, do not over-read and
1601 instead leave the unprompted TLS handshake in socket buffer for the
1602 TLS library to consume.
1604 PP/04 Bug 2018: Also handle Proxy Protocol v2 safely.
1606 PP/05 FreeBSD compat: handle that Ports no longer create /usr/bin/perl
1608 JH/16 Drop variables when they go out of scope. Memory management drops a whole
1609 region in one operation, for speed, and this leaves assigned pointers
1610 dangling. Add checks run only under the testsuite which checks all
1611 variables at a store-reset and panics on a dangling pointer; add code
1612 explicitly nulling out all the variables discovered. Fixes one known
1613 bug: a transport crash, where a dangling pointer for $sending_ip_address
1614 originally assigned in a verify callout, is re-used.
1616 PP/06 Drop '.' from @INC in various Perl scripts.
1618 PP/07 Switch FreeBSD iconv to always use the base-system libc functions.
1620 PP/08 Reduce a number of compilation warnings under clang; building with
1621 CC=clang CFLAGS+=-Wno-dangling-else -Wno-logical-op-parentheses
1622 should be warning-free.
1624 JH/17 Fix inbound CHUNKING when DKIM disabled at runtime.
1626 HS/01 Fix portability problems introduced by PP/08 for platforms where
1627 realloc(NULL) is not equivalent to malloc() [SunOS et al].
1629 HS/02 Bug 1974: Fix missing line terminator on the last received BDAT
1630 chunk. This allows us to accept broken chunked messages. We need a more
1631 general solution here.
1633 PP/09 Wrote util/chunking_fixqueue_finalnewlines.pl to help recover
1634 already-broken messages in the queue.
1636 JH/18 Bug 2061: Fix ${extract } corrupting an enclosing ${reduce } $value.
1638 JH/19 Fix reference counting bug in routing-generated-address tracking.
1644 JH/01 Use SIZE on MAIL FROM in a cutthrough connection, if the destination
1645 supports it and a size is available (ie. the sending peer gave us one).
1647 JH/02 The obsolete acl condition "demime" is removed (finally, after ten
1648 years of being deprecated). The replacements are the ACLs
1649 acl_smtp_mime and acl_not_smtp_mime.
1651 JH/03 Upgrade security requirements imposed for hosts_try_dane: previously
1652 a downgraded non-dane trust-anchor for the TLS connection (CA-style)
1653 or even an in-clear connection were permitted. Now, if the host lookup
1654 was dnssec and dane was requested then the host is only used if the
1655 TLSA lookup succeeds and is dnssec. Further hosts (eg. lower priority
1656 MXs) will be tried (for hosts_try_dane though not for hosts_require_dane)
1657 if one fails this test.
1658 This means that a poorly-configured remote DNS will make it incommunicado;
1659 but it protects against a DNS-interception attack on it.
1661 JH/04 Bug 1810: make continued-use of an open smtp transport connection
1662 non-noisy when a race steals the message being considered.
1664 JH/05 If main configuration option tls_certificate is unset, generate a
1665 self-signed certificate for inbound TLS connections.
1667 JH/06 Bug 165: hide more cases of password exposure - this time in expansions
1668 in rewrites and routers.
1670 JH/07 Retire gnutls_require_mac et.al. These were nonfunctional since 4.80
1671 and logged a warning sing 4.83; now they are a configuration file error.
1673 JH/08 Bug 1836: Fix crash in VRFY handling when handed an unqualified name
1674 (lacking @domain). Apply the same qualification processing as RCPT.
1676 JH/09 Bug 1804: Avoid writing msglog files when in -bh or -bhc mode.
1678 JH/10 Support ${sha256:} applied to a string (as well as the previous
1681 JH/11 Cutthrough: avoid using the callout hints db on a verify callout when
1682 a cutthrough deliver is pending, as we always want to make a connection.
1683 This also avoids re-routing the message when later placing the cutthrough
1684 connection after a verify cache hit.
1685 Do not update it with the verify result either.
1687 JH/12 Cutthrough: disable when verify option success_on_redirect is used, and
1688 when routing results in more than one destination address.
1690 JH/13 Cutthrough: expand transport dkim_domain option when testing for dkim
1691 signing (which inhibits the cutthrough capability). Previously only
1692 the presence of an option was tested; now an expansion evaluating as
1693 empty is permissible (obviously it should depend only on data available
1694 when the cutthrough connection is made).
1696 JH/14 Fix logging of errors under PIPELINING. Previously the log line giving
1697 the relevant preceding SMTP command did not note the pipelining mode.
1699 JH/15 Fix counting of empty lines in $body_linecount and $message_linecount.
1700 Previously they were not counted.
1702 JH/16 DANE: treat a TLSA lookup response having all non-TLSA RRs, the same
1703 as one having no matching records. Previously we deferred the message
1704 that needed the lookup.
1706 JH/17 Fakereject: previously logged as a normal message arrival "<="; now
1707 distinguished as "(=".
1709 JH/18 Bug 1867: make the fail_defer_domains option on a dnslookup router work
1710 for missing MX records. Previously it only worked for missing A records.
1712 JH/19 Bug 1850: support Radius libraries that return REJECT_RC.
1714 JH/20 Bug 1872: Ensure that acl_smtp_notquit is run when the connection drops
1715 after the data-go-ahead and data-ack. Patch from Jason Betts.
1717 JH/21 Bug 1846: Send DMARC forensic reports for reject and quarantine results,
1718 even for a "none" policy. Patch from Tony Meyer.
1720 JH/22 Fix continued use of a connection for further deliveries. If a port was
1721 specified by a router, it must also match for the delivery to be
1724 JH/23 Bug 1874: fix continued use of a connection for further deliveries.
1725 When one of the recipients of a message was unsuitable for the connection
1726 (has no matching addresses), we lost track of needing to mark it
1727 deferred. As a result mail would be lost.
1729 JH/24 Bug 1832: Log EHLO response on getting conn-close response for HELO.
1731 JH/25 Decoding ACL controls is now done using a binary search; the source code
1732 takes up less space and should be simpler to maintain. Merge the ACL
1733 condition decode tables also, with similar effect.
1735 JH/26 Fix problem with one_time used on a redirect router which returned the
1736 parent address unchanged. A retry would see the parent address marked as
1737 delivered, so not attempt the (identical) child. As a result mail would
1740 JH/27 Fix a possible security hole, wherein a process operating with the Exim
1741 UID can gain a root shell. Credit to http://www.halfdog.net/ for
1742 discovery and writeup. Ubuntu bug 1580454; no bug raised against Exim
1745 JH/28 Enable {spool,log} filesystem space and inode checks as default.
1746 Main config options check_{log,spool}_{inodes,space} are now
1747 100 inodes, 10MB unless set otherwise in the configuration.
1749 JH/29 Fix the connection_reject log selector to apply to the connect ACL.
1750 Previously it only applied to the main-section connection policy
1753 JH/30 Bug 1897: fix callouts connection fallback from TLS to cleartext.
1755 PP/01 Changed default Diffie-Hellman parameters to be Exim-specific, created
1756 by me. Added RFC7919 DH primes as an alternative.
1758 PP/02 Unbreak build via pkg-config with new hash support when crypto headers
1759 are not in the system include path.
1761 JH/31 Fix longstanding bug with aborted TLS server connection handling. Under
1762 GnuTLS, when a session startup failed (eg because the client disconnected)
1763 Exim did stdio operations after fclose. This was exposed by a recent
1764 change which nulled out the file handle after the fclose.
1766 JH/32 Bug 1909: Fix OCSP proof verification for cases where the proof is
1767 signed directly by the cert-signing cert, rather than an intermediate
1768 OCSP-signing cert. This is the model used by LetsEncrypt.
1770 JH/33 Bug 1914: Ensure socket is nonblocking before draining after SMTP QUIT.
1772 HS/01 Fix leak in verify callout under GnuTLS, about 3MB per recipient on
1773 an incoming connection.
1775 HS/02 Bug 1802: Do not half-close the connection after sending a request
1778 HS/03 Use "auto" as the default EC curve parameter. For OpenSSL < 1.0.2
1779 fallback to "prime256v1".
1781 JH/34 SECURITY: Use proper copy of DATA command in error message.
1782 Could leak key material. Remotely exploitable. CVE-2016-9963.
1788 JH/01 Bug 1664: Disable OCSP for GnuTLS library versions at/before 3.3.16
1789 and 3.4.4 - once the server is enabled to respond to an OCSP request
1790 it does even when not requested, resulting in a stapling non-aware
1791 client dropping the TLS connection.
1793 TF/01 Code cleanup: Overhaul the debug_selector and log_selector machinery to
1794 support variable-length bit vectors. No functional change.
1796 TF/02 Improve the consistency of logging incoming and outgoing interfaces.
1797 The I= interface field on outgoing lines is now after the H= remote
1798 host field, same as incoming lines. There is a separate
1799 outgoing_interface log selector which allows you to disable the
1802 JH/02 Bug 728: Close logfiles after a daemon-process "exceptional" log write.
1803 If not running log_selector +smtp_connection the mainlog would be held
1804 open indefinitely after a "too many connections" event, including to a
1805 deleted file after a log rotate. Leave the per net connection logging
1806 leaving it open for efficiency as that will be quickly detected by the
1807 check on the next write.
1809 HS/01 Bug 1671: Fix post transport crash.
1810 Processing the wait-<transport> messages could crash the delivery
1811 process if the message IDs didn't exist for some reason. When
1812 using 'split_spool_directory=yes' the construction of the spool
1813 file name failed already, exposing the same netto behaviour.
1815 JH/03 Bug 425: Capture substrings in $regex1, $regex2 etc from regex &
1816 mime_regex ACL conditions.
1818 JH/04 Bug 1686: When compiled with EXPERIMENTAL_DSN_INFO: Add extra information
1819 to DSN fail messages (bounces): remote IP, remote greeting, remote response
1820 to HELO, local diagnostic string.
1822 JH/05 Downgrade message for a TLS-certificate-based authentication fail from
1823 log line to debug. Even when configured with a tls authenticator many
1824 client connections are expected to not authenticate in this way, so
1825 an authenticate fail is not an error.
1827 HS/02 Add the Exim version string to the process info. This way exiwhat
1828 gives some more detail about the running daemon.
1830 JH/06 Bug 1395: time-limit caching of DNS lookups, to the TTL value. This may
1831 matter for fast-change records such as DNSBLs.
1833 JH/07 Bug 1678: Always record an interface option value, if set, as part of a
1834 retry record, even if constant. There may be multiple transports with
1835 different interface settings and the retry behaviour needs to be kept
1838 JH/08 Bug 1586: exiqgrep now refuses to run if there are unexpected arguments.
1840 JH/09 Bug 1700: ignore space & tab embedded in base64 during decode.
1842 JH/10 Bug 840: fix log_defer_output option of pipe transport
1844 JH/11 Bug 830: use same host for all RCPTS of a message, even under
1845 hosts_randomize. This matters a lot when combined with mua_wrapper.
1847 JH/12 Bug 1706: percent and underbar characters are no longer escaped by the
1848 ${quote_pgsql:<string>} operator.
1850 JH/13 Bug 1708: avoid misaligned access in cached lookup.
1852 JH/14 Change header file name for freeradius-client. Relevant if compiling
1853 with Radius support; from the Gentoo tree and checked under Fedora.
1855 JH/15 Bug 1712: Introduce $prdr_requested flag variable
1857 JH/16 Bug 1714: Permit an empty string as expansion result for transport
1858 option transport_filter, meaning no filtering.
1860 JH/17 Bug 1713: Fix non-PDKIM_DEBUG build. Patch from Jasen Betts.
1862 JH/18 Bug 1709: When built with TLS support, the tls_advertise_hosts option now
1863 defaults to "*" (all hosts). The variable is now available when not built
1864 with TLS, default unset, mainly to enable keeping the testsuite sane.
1865 If a server certificate is not supplied (via tls_certificate) an error is
1866 logged, and clients will find TLS connections fail on startup. Presumably
1867 they will retry in-clear.
1868 Packagers of Exim are strongly encouraged to create a server certificate
1869 at installation time.
1871 HS/03 Add -bP config_file as a synonym for -bP configure_file, for consistency
1872 with the $config_file variable.
1874 JH/19 Two additional event types: msg:rcpt:defer and msg:rcpt:host:defer. Both
1875 in transport context, after the attempt, and per-recipient. The latter type
1876 is per host attempted. The event data is the error message, and the errno
1877 information encodes the lookup type (A vs. MX) used for the (first) host,
1878 and the trailing two digits of the smtp 4xx response.
1880 GF/01 Bug 1715: Fix for race condition in exicyclog, where exim could attempt
1881 to write to mainlog (or rejectlog, paniclog) in the window between file
1882 creation and permissions/ownership being changed. Particularly affects
1883 installations where exicyclog is run as root, rather than exim user;
1884 result is that the running daemon panics and dies.
1886 JH/20 Bug 1701: For MySQL lookups, support MySQL config file option group names.
1888 JH/21 Bug 1720: Add support for priority groups and weighted-random proxy
1889 selection for the EXPERIMENTAL_SOCKS feature, via new per-proxy options
1890 "pri" and "weight". Note that the previous implicit priority given by the
1891 list order is no longer honoured.
1893 JH/22 Bugs 963, 1721: Fix some corner cases in message body canonicalization
1894 for DKIM processing.
1896 JH/23 Move SOCKS5 support from Experimental to mainline, enabled for a build
1897 by defining SUPPORT_SOCKS.
1899 JH/26 Move PROXY support from Experimental to mainline, enabled for a build
1900 by defining SUPPORT_PROXY. Note that the proxy_required_hosts option
1901 is renamed to hosts_proxy, and the proxy_{host,target}_{address,port}.
1902 variables are renamed to proxy_{local,external}_{address,port}.
1904 JH/27 Move Internationalisation support from Experimental to mainline, enabled
1905 for a build by defining SUPPORT_I18N
1907 JH/28 Bug 1745: Fix redis lookups to handle (quoted) spaces embedded in parts
1908 of the query string, and make ${quote_redis:} do that quoting.
1910 JH/29 Move Events support from Experimental to mainline, enabled by default
1911 and removable for a build by defining DISABLE_EVENT.
1913 JH/30 Updated DANE implementation code to current from Viktor Dukhovni.
1915 JH/31 Fix bug with hosts_connection_nolog and named-lists which were wrongly
1916 cached by the daemon.
1918 JH/32 Move Redis support from Experimental to mainline, enabled for a build
1919 by defining LOOKUP_REDIS. The libhiredis library is required.
1921 JH/33 Bug 1748: Permit ACL dnslists= condition in non-smtp ACLs if explicit
1922 keys are given for lookup.
1924 JH/34 Bug 1192: replace the embedded copy of PolarSSL RSA routines in the DKIM
1925 support, by using OpenSSL or GnuTLS library ones. This means DKIM is
1926 only supported when built with TLS support. The PolarSSL SHA routines
1927 are still used when the TLS library is too old for convenient support.
1929 JH/35 Require SINGLE_DH_USE by default in OpenSSL (main config option
1930 openssl_options), for security. OpenSSL forces this from version 1.1.0
1931 server-side so match that on older versions.
1933 JH/36 Bug 1778: longstanding bug in memory use by the ${run } expansion: A fresh
1934 allocation for $value could be released as the expansion processing
1935 concluded, but leaving the global pointer active for it.
1937 JH/37 Bug 1769: Permit a VRFY ACL to override the default 252 response,
1938 and to use the domains and local_parts ACL conditions.
1940 JH/38 Fix cutthrough bug with body lines having a single dot. The dot was
1941 incorrectly not doubled on cutthrough transmission, hence seen as a
1942 body-termination at the receiving system - resulting in truncated mails.
1943 Commonly the sender saw a TCP-level error, and retransmitted the message
1944 via the normal store-and-forward channel. This could result in duplicates
1945 received - but deduplicating mailstores were liable to retain only the
1946 initial truncated version.
1948 JH/39 Bug 1781: Fix use of DKIM private-keys having trailing '=' in the base-64.
1950 JH/40 Fix crash in queryprogram router when compiled with EXPERIMENTAL_SRS.
1952 JH/41 Bug 1792: Fix selection of headers to sign for DKIM: bottom-up. While
1953 we're in there, support oversigning also; bug 1309.
1955 JH/42 Bug 1796: Fix error logged on a malware scanner connection failure.
1957 HS/04 Add support for keep_environment and add_environment options.
1959 JH/43 Tidy coding issues detected by gcc --fsanitize=undefined. Some remain;
1960 either intentional arithmetic overflow during PRNG, or testing config-
1963 JH/44 Bug 1800: The combination of a -bhc commandline option and cutthrough
1964 delivery resulted in actual delivery. Cancel cutthrough before DATA
1967 JH/45 Fix cutthrough, when connection not opened by verify and target hard-
1968 rejects a recipient: pass the reject to the originator.
1970 JH/46 Multiple issues raised by Coverity. Some were obvious or plausible bugs.
1971 Many were false-positives and ignorable, but it's worth fixing the
1974 JH/47 Fix build on HP-UX and older Solaris, which need (un)setenv now also
1975 for the new environment-manipulation done at startup. Move the routines
1976 from being local to tls.c to being global via the os.c file.
1978 JH/48 Bug 1807: Fix ${extract } for the numeric/3-string case. While preparsing
1979 an extract embedded as result-arg for a map, the first arg for extract
1980 is unavailable so we cannot tell if this is a numbered or keyed
1981 extraction. Accept either.
1987 JH/01 Bug 1545: The smtp transport option "retry_include_ip_address" is now
1990 JH/02 The smtp transport option "multi_domain" is now expanded.
1992 JH/03 The smtp transport now requests PRDR by default, if the server offers
1995 JH/04 Certificate name checking on server certificates, when exim is a client,
1996 is now done by default. The transport option tls_verify_cert_hostnames
1997 can be used to disable this per-host. The build option
1998 EXPERIMENTAL_CERTNAMES is withdrawn.
2000 JH/05 The value of the tls_verify_certificates smtp transport and main options
2001 default to the word "system" to access the system default CA bundle.
2002 For GnuTLS, only version 3.0.20 or later.
2004 JH/06 Verification of the server certificate for a TLS connection is now tried
2005 (but not required) by default. The verification status is now logged by
2006 default, for both outbound TLS and client-certificate supplying inbound
2009 JH/07 Changed the default rfc1413 lookup settings to disable calls. Few
2012 JH/08 The EXPERIMENTAL_DSN compile option is no longer needed; all Delivery
2013 Status Notification (bounce) messages are now MIME format per RFC 3464.
2014 Support for RFC 3461 DSN options NOTIFY,ENVID,RET,ORCPT can be advertised
2015 under the control of the dsn_advertise_hosts option, and routers may
2016 have a dsn_lasthop option.
2018 JH/09 A timeout of 2 minutes is now applied to all malware scanner types by
2019 default, modifiable by a malware= option. The list separator for
2020 the options can now be changed in the usual way. Bug 68.
2022 JH/10 The smtp_receive_timeout main option is now expanded before use.
2024 JH/11 The incoming_interface log option now also enables logging of the
2025 local interface on delivery outgoing connections.
2027 JH/12 The cutthrough-routing facility now supports multi-recipient mails,
2028 if the interface and destination host and port all match.
2030 JH/13 Bug 344: The verify = reverse_host_lookup ACL condition now accepts a
2033 JH/14 Bug 1573: The spam= ACL condition now additionally supports Rspamd.
2034 Patch from Andrew Lewis.
2036 JH/15 Bug 670: The spamd_address main option (for the spam= ACL condition)
2037 now supports optional time-restrictions, weighting, and priority
2038 modifiers per server. Patch originally by <rommer@active.by>.
2040 JH/16 The spamd_address main option now supports a mixed list of local
2041 and remote servers. Remote servers can be IPv6 addresses, and
2042 specify a port-range.
2044 JH/17 Bug 68: The spamd_address main option now supports an optional
2045 timeout value per server.
2047 JH/18 Bug 1581: Router and transport options headers_add/remove can
2048 now have the list separator specified.
2050 JH/19 Bug 392: spamd_address, and clamd av_scanner, now support retry
2053 JH/20 Bug 1571: Ensure that $tls_in_peerdn is set, when verification fails
2056 JH/21 Support for the A6 type of dns record is withdrawn.
2058 JH/22 Bug 608: The result of a QUIT or not-QUIT toplevel ACL now matters
2059 rather than the verbs used.
2061 JH/23 Bug 1572: Increase limit on SMTP confirmation message copy size
2062 from 255 to 1024 chars.
2064 JH/24 Verification callouts now attempt to use TLS by default.
2066 HS/01 DNSSEC options (dnssec_require_domains, dnssec_request_domains)
2067 are generic router options now. The defaults didn't change.
2069 JH/25 Bug 466: Add RFC2322 support for MIME attachment filenames.
2070 Original patch from Alexander Shikoff, worked over by JH.
2072 HS/02 Bug 1575: exigrep falls back to autodetection of compressed
2073 files if ZCAT_COMMAND is not executable.
2075 JH/26 Bug 1539: Add timeout/retry options on dnsdb lookups.
2077 JH/27 Bug 286: Support SOA lookup in dnsdb lookups.
2079 JH/28 Bug 1588: Do not use the A lookup following an AAAA for setting the FQDN.
2080 Normally benign, it bites when the pair was led to by a CNAME;
2081 modern usage is to not canonicalize the domain to a CNAME target
2082 (and we were inconsistent anyway for A-only vs AAAA+A).
2084 JH/29 Bug 1632: Removed the word "rejected" from line logged for ACL discards.
2086 JH/30 Check the forward DNS lookup for DNSSEC, in addition to the reverse,
2087 when evaluating $sender_host_dnssec.
2089 JH/31 Check the HELO verification lookup for DNSSEC, adding new
2090 $sender_helo_dnssec variable.
2092 JH/32 Bug 1397: Enable ECDHE on OpenSSL, just the NIST P-256 curve.
2094 JH/33 Bug 1346: Note MAIL cmd seen in -bS batch, to avoid smtp_no_mail log.
2096 JH/34 Bug 1648: Fix a memory leak seen with "mailq" and large queues.
2098 JH/35 Bug 1642: Fix support of $spam_ variables at delivery time. Was
2099 documented as working, but never had. Support all but $spam_report.
2101 JH/36 Bug 1659: Guard checking of input smtp commands again pseudo-command
2102 added for tls authenticator.
2104 HS/03 Add perl_taintmode main config option
2110 TL/01 When running the test suite, the README says that variables such as
2111 no_msglog_check are global and can be placed anywhere in a specific
2112 test's script, however it was observed that placement needed to be near
2113 the beginning for it to behave that way. Changed the runtest perl
2114 script to read through the entire script once to detect and set these
2115 variables, reset to the beginning of the script, and then run through
2116 the script parsing/test process like normal.
2118 TL/02 The BSD's have an arc4random API. One of the functions to induce
2119 adding randomness was arc4random_stir(), but it has been removed in
2120 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
2121 function when detected.
2123 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
2124 cause callback expansion.
2126 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
2127 syntax errors in an expansion can be treated as a string instead of
2128 logging or causing an error, due to the internal use of bool_lax
2129 instead of bool when processing it.
2131 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
2132 server certificates when making smtp deliveries.
2134 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
2136 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
2138 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
2140 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
2141 Merged patch from Sebastian Wiedenroth.
2143 JH/05 Fix results-pipe from transport process. Several recipients, combined
2144 with certificate use, exposed issues where response data items split
2145 over buffer boundaries were not parsed properly. This eventually
2146 resulted in duplicates being sent. This issue only became common enough
2147 to notice due to the introduction of connection certificate information,
2148 the item size being so much larger. Found and fixed by Wolfgang Breyha.
2150 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
2151 size buffer was used, resulting in syntax errors when an expansion
2154 JH/07 Add support for directories of certificates when compiled with a GnuTLS
2155 version 3.3.6 or later.
2157 JH/08 Rename the TPDA experimental facility to Event Actions. The #ifdef
2158 is EXPERIMENTAL_EVENT, the main-configuration and transport options
2159 both become "event_action", the variables become $event_name, $event_data
2160 and $event_defer_errno. There is a new variable $verify_mode, usable in
2161 routers, transports and related events. The tls:cert event is now also
2162 raised for inbound connections, if the main configuration event_action
2165 TL/06 In test suite, disable OCSP for old versions of openssl which contained
2166 early OCSP support, but no stapling (appears to be less than 1.0.0).
2168 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
2169 server certificate names available under the smtp transport option
2170 "tls_verify_cert_hostname" now do not permit multi-component wildcard
2173 JH/10 Time-related extraction expansions from certificates now use the main
2174 option "timezone" setting for output formatting, and are consistent
2175 between OpenSSL and GnuTLS compilations. Bug 1541.
2177 JH/11 Fix a crash in mime ACL when meeting a zero-length, quoted or RFC2047-
2178 encoded parameter in the incoming message. Bug 1558.
2180 JH/12 Bug 1527: Autogrow buffer used in reading spool files. Since they now
2181 include certificate info, eximon was claiming there were spoolfile
2184 JH/13 Bug 1521: Fix ldap lookup for single-attr request, multiple-attr return.
2186 JH/14 Log delivery-related information more consistently, using the sequence
2187 "H=<name> [<ip>]" wherever possible.
2189 TL/07 Bug 1547: Omit RFCs from release. Draft and RFCs have licenses which
2190 are problematic for Debian distribution, omit them from the release
2193 JH/15 Updates and fixes to the EXPERIMENTAL_DSN feature.
2195 JH/16 Fix string representation of time values on 64bit time_t architectures.
2198 JH/17 Fix a null-indirection in certextract expansions when a nondefault
2199 output list separator was used.
2204 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
2205 checkers that were complaining about end of non-void function with no
2208 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
2209 This was a regression introduced in 4.83 by another bugfix.
2211 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
2213 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
2214 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
2220 TF/01 Correctly close the server side of TLS when forking for delivery.
2222 When a message was received over SMTP with TLS, Exim failed to clear up
2223 the incoming connection properly after forking off the child process to
2224 deliver the message. In some situations the subsequent outgoing
2225 delivery connection happened to have the same fd number as the incoming
2226 connection previously had. Exim would try to use TLS and fail, logging
2227 a "Bad file descriptor" error.
2229 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
2230 utilities have not been installed.
2232 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
2233 temporary space as the ACL may create new global variables.
2235 TL/01 LDAP support uses per connection or global context settings, depending
2236 upon the detected version of the libraries at build time.
2238 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
2239 to extract and use the src ip:port in logging and expansions as if it
2240 were a direct connection from the outside internet. PPv2 support was
2241 updated based on HAProxy spec change in May 2014.
2243 JH/02 Add ${listextract {number}{list}{success}{fail}}.
2245 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
2246 Properly escape header and check for NULL return.
2248 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
2251 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
2253 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
2254 characters in header names, implemented as a verify condition.
2255 Contributed by Michael Fischer v. Mollard.
2257 TL/05 Rename SPF condition results err_perm and err_temp to standardized
2258 results permerror and temperror. Previous values are deprecated but
2259 still accepted. In a future release, err_perm and err_temp will be
2260 completely removed, which will be a backward incompatibility if the
2261 ACL tests for either of these two old results. Patch contributed by
2262 user bes-internal on the mailing list.
2264 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
2266 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
2267 selectors, in both main and reject logs.
2269 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
2272 JH/07 Add malware type "sock" for talking to simple daemon.
2274 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
2276 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
2277 routers/transports under cutthrough routing.
2279 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
2280 numbers. Touch up "bool" conditional to keep the same definition.
2282 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
2284 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
2286 JH/12 Expand items in router/transport headers_add or headers_remove lists
2287 individually rather than the list as a whole. Bug 1452.
2289 Required for reasonable handling of multiple headers_ options when
2290 they may be empty; requires that headers_remove items with embedded
2291 colons must have them doubled (or the list-separator changed).
2293 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
2294 view the policy declared in the DMARC record. Currently, $dmarc_status
2295 is a combined value of both the record presence and the result of the
2298 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
2300 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
2301 dnslookup router and the smtp transport (applying to the forward
2304 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
2305 of ldap servers used for a specific lookup. Patch provided by Heiko
2308 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
2309 New variable $lookup_dnssec_authenticated for observability.
2311 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
2312 Patch submitted by Lars Timman.
2314 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
2316 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
2317 Requires trusted mode and valid format message id, aborts otherwise.
2318 Patch contributed by Heiko Schlichting.
2320 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
2321 certextract with support for various fields. Bug 1358.
2323 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
2324 is requested by default, modifiable by smtp transport option
2327 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
2328 operate on certificate variables to give certificate fingerprints
2329 Also new ${sha256:cert_variable}.
2331 JH/23 The PRDR feature is moved from being Experimental into the mainline.
2333 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
2334 Christian Aistleitner.
2336 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
2338 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
2339 file. Patch from Wolfgang Breyha.
2341 JH/25 Expand the coverage of the delivery $host and $host_address to
2342 client authenticators run in verify callout. Bug 1476.
2344 JH/26 Port service names are now accepted for tls_on_connect_ports, to
2345 align with daemon_smtp_ports. Bug 72.
2347 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
2348 support and error reporting did not work properly.
2350 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
2351 and is readable. Patch from Andrew Colin Kissa.
2353 TL/14 Enhance documentation of ${run expansion and how it parses the
2354 commandline after expansion, particularly in the case when an
2355 unquoted variable expansion results in an empty value.
2357 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
2359 PP/02 Fix internal collision of T_APL on systems which support RFC3123
2360 by renaming away from it. Addresses GH issue 15, reported by
2363 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
2365 TL/15 SECURITY: prevent double expansion in math comparison functions
2366 (can expand unsanitized data). Not remotely exploitable.
2373 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
2375 PP/02 Make -n do something, by making it not do something.
2376 When combined with -bP, the name of an option is not output.
2378 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
2381 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
2382 $sender_host_name and config options to manage this, and basic check
2385 PP/05 DSCP support for outbound connections and control modifier for inbound.
2387 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
2388 (Only plugin which currently uses this is kerberos4, which nobody should
2389 be using, but we should make it available and other future plugins might
2390 conceivably use it, even though it would break NAT; stuff *should* be
2391 using channel bindings instead).
2393 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
2394 name; added for Sendmail compatibility; requires admin caller.
2395 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
2396 just ignore it); requires trusted caller.
2397 Also parse but ignore: -Ac -Am -X<logfile>
2400 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
2402 TL/02 Add +smtp_confirmation as a default logging option.
2404 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
2405 Patch by Magnus Holmgren from 2007-02-20.
2407 TL/04 Bugzilla 1281 - Spec typo.
2408 Bugzilla 1283 - Spec typo.
2409 Bugzilla 1290 - Spec grammar fixes.
2411 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
2413 TL/06 Add Experimental DMARC support using libopendmarc libraries.
2415 TL/07 Fix an out of order global option causing a segfault. Reported to dev
2416 mailing list by by Dmitry Isaikin.
2418 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
2420 JH/02 Support "G" suffix to numbers in ${if comparisons.
2422 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
2424 NM/01 Bugzilla 1197 - Spec typo
2425 Bugzilla 1196 - Spec examples corrections
2427 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
2429 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
2430 gnutls_enable_pkcs11, but renamed to more accurately indicate its
2433 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
2434 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
2436 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
2437 "acl {{name}{arg}...}", and optional args on acl condition
2440 JH/05 Permit multiple router/transport headers_add/remove lines.
2442 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
2444 JH/07 Avoid using a waiting database for a single-message-only transport.
2445 Performance patch from Paul Fisher. Bugzilla 1262.
2447 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
2450 JH/09 Add $headers_added variable, with content from use of ACL modifier
2451 add_header (but not yet added to the message). Bugzilla 199.
2453 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
2454 Pulled from Bugzilla 817 by Wolfgang Breyha.
2456 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
2458 (nb: this is the same fix as in Exim 4.80.1)
2460 JH/11 Add A= logging on delivery lines, and a client_set_id option on
2463 JH/12 Add optional authenticated_sender logging to A= and a log_selector
2466 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
2468 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
2469 advertise SMTP AUTH mechanism to us, instead of a generic
2470 protocol violation error. Also, make Exim more robust to bad
2471 data from the Dovecot auth socket.
2473 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
2475 When a queue runner is handling a message, Exim first routes the
2476 recipient addresses, during which it prunes them based on the retry
2477 hints database. After that it attempts to deliver the message to
2478 any remaining recipients. It then updates the hints database using
2481 So if a recipient address works intermittently, it can get repeatedly
2482 deferred at routing time. The retry hints record remains fresh so the
2483 address never reaches the final cutoff time.
2485 This is a fairly common occurrence when a user is bumping up against
2486 their storage quota. Exim had some logic in its local delivery code
2487 to deal with this. However it did not apply to per-recipient defers
2488 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
2490 This change adds a proper retry rule check during routing so that the
2491 final cutoff time is checked against the message's age. We only do
2492 this check if there is an address retry record and there is not a
2493 domain retry record; this implies that previous attempts to handle
2494 the address had the retry_use_local_parts option turned on. We use
2495 this as an approximation for the destination being like a local
2496 delivery, as in LMTP.
2498 I suspect this new check makes the old local delivery cutoff check
2499 redundant, but I have not verified this so I left the code in place.
2501 TF/02 Correct gecos expansion when From: is a prefix of the username.
2503 Test 0254 submits a message to Exim with the header
2507 When I ran the test suite under the user fanf2, Exim expanded
2508 the header to contain my full name, whereas it should have added
2509 a Resent-Sender: header. It erroneously treats any prefix of the
2510 username as equal to the username.
2512 This change corrects that bug.
2514 GF/01 DCC debug and logging tidyup
2515 Error conditions log to paniclog rather than rejectlog.
2516 Debug lines prefixed by "DCC: " to remove any ambiguity.
2518 TF/03 Avoid unnecessary rebuilds of lookup-related code.
2520 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
2521 Bug spotted by Jeremy Harris; was flawed since initial commit.
2522 Would have resulted in OCSP responses post-SNI triggering an Exim
2523 NULL dereference and crash.
2525 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
2527 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
2528 Bug detection, analysis and fix by Samuel Thibault.
2529 Bugzilla 1331, Debian bug #698092.
2531 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
2533 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
2534 Server implementation by Todd Lyons, client by JH.
2535 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
2536 config variable "prdr_enable" controls whether the server
2537 advertises the facility. If the client requests PRDR a new
2538 acl_data_smtp_prdr ACL is called once for each recipient, after
2539 the body content is received and before the acl_smtp_data ACL.
2540 The client is controlled by both of: a hosts_try_prdr option
2541 on the smtp transport, and the server advertisement.
2542 Default client logging of deliveries and rejections involving
2543 PRDR are flagged with the string "PRDR".
2545 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
2546 fclose(). Diagnosis by Todd Lyons.
2548 PP/17 Update configure.default to handle IPv6 localhost better.
2549 Patch by Alain Williams (plus minor tweaks).
2552 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
2553 This is now consistent with GnuTLS, and is now documented: the
2554 previous undocumented portable approach to treating the option as
2555 unset was to force an expansion failure. That still works, and
2556 an empty string is now equivalent.
2558 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
2559 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
2560 not performing validation itself.
2562 PP/20 Added force_command boolean option to pipe transport.
2563 Patch from Nick Koston, of cPanel Inc.
2565 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
2568 TF/04 Added udpsend ACL modifier and hexquote expansion operator
2570 PP/21 Fix eximon continuous updating with timestamped log-files.
2571 Broken in a format-string cleanup in 4.80, missed when I repaired the
2572 other false fix of the same issue.
2573 Report and fix from Heiko Schlichting.
2576 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
2577 Report from Prashanth Katuri.
2579 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
2580 It's SecureTransport, so affects any MacOS clients which use the
2581 system-integrated TLS libraries, including email clients.
2583 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
2584 using a MIME ACL for non-SMTP local injection.
2585 Report and assistance in diagnosis by Warren Baker.
2587 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
2589 JH/16 Fix comparisons for 64b. Bugzilla 1385.
2591 TL/09 Add expansion variable $authenticated_fail_id to keep track of
2592 last id that failed so it may be referenced in subsequent ACL's.
2594 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
2597 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
2598 ldap library initialization, allowing self-signed CA's to be
2599 used. Also properly sets require_cert option later in code by
2600 using NULL (global ldap config) instead of ldap handle (per
2601 session). Bug diagnosis and testing by alxgomz.
2603 TL/12 Enhanced documentation in the ratelimit.pl script provided in
2604 the src/util/ subdirectory.
2606 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
2607 renamed to Transport Post Delivery Action by Jeremy Harris, as
2610 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
2611 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
2612 redis_servers = needs to be configured which will be used by the redis
2613 lookup. Patch from Warren Baker, of The Packet Hub.
2615 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
2617 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
2618 hostname or reverse DNS when processing a host list. Used suggestions
2619 from multiple comments on this bug.
2621 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
2623 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
2624 Missed a few lines, added it to make the runtest require no keyboard
2627 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
2628 contains upper case chars. Make router use caseful_local_part.
2630 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
2631 support when GnuTLS has been built with p11-kit.
2637 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
2639 This, or similar/improved, will also be change PP/11 of 4.82.
2645 PP/01 Handle short writes when writing local log-files.
2646 In practice, only affects FreeBSD (8 onwards).
2647 Bugzilla 1053, with thanks to Dmitry Isaikin.
2649 NM/01 Bugzilla 949 - Documentation tweak
2651 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
2654 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
2656 PP/02 Implemented gsasl authenticator.
2658 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
2660 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
2661 `pkg-config foo` for cflags/libs.
2663 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
2664 with rest of GSASL and with heimdal_gssapi.
2666 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
2667 `pkg-config foo` for cflags/libs for the TLS implementation.
2669 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
2670 properties get this fed in as external SSF. A number of robustness
2671 and debugging improvements to the cyrus_sasl authenticator.
2673 PP/08 cyrus_sasl server now expands the server_realm option.
2675 PP/09 Bugzilla 1214 - Log authentication information in reject log.
2676 Patch by Jeremy Harris.
2678 PP/10 Added dbmjz lookup type.
2680 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
2682 PP/12 MAIL args handles TAB as well as SP, for better interop with
2683 non-compliant senders.
2684 Analysis and variant patch by Todd Lyons.
2686 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
2687 Bug report from Lars Müller <lars@samba.org> (via SUSE),
2688 Patch from Dirk Mueller <dmueller@suse.com>
2690 PP/13 tls_peerdn now print-escaped for spool files.
2691 Observed some $tls_peerdn in wild which contained \n, which resulted
2692 in spool file corruption.
2694 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
2695 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
2696 or write after TLS renegotiation, which otherwise led to messages
2699 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
2700 as a tracking header (ie: a signed header comes before the signature).
2701 Patch from Wolfgang Breyha.
2703 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
2704 comma-sep list; embedded commas doubled.
2706 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
2708 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
2710 Report and patch from Dmitry Banschikov.
2712 PP/16 Removed "dont_insert_empty_fragments" from "openssl_options".
2713 Removed SSL_clear() after SSL_new() which led to protocol negotiation
2714 failures. We appear to now support TLS1.1+ with Exim.
2716 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
2717 lets Exim select keys and certificates based upon TLS SNI from client.
2718 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
2719 before an outbound SMTP session. New log_selector, +tls_sni.
2721 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
2722 NULL dereference. Report and patch from Alun Jones.
2724 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
2725 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
2726 Not seeing resolver debug output on NetBSD, but suspect this is a
2727 resolver implementation change.
2729 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
2730 Left warnings. Added "eximon gdb" invocation mode.
2732 PP/21 Defaulting "accept_8bitmime" to true, not false.
2734 PP/22 Added -bw for inetd wait mode support.
2736 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
2737 locate the relevant includes and libraries. Made this the default.
2739 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
2740 Bugzilla 1246, report and most of solution from Tomasz Kusy.
2742 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
2743 This may cause build issues on older platforms.
2745 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
2746 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
2747 gnutls_require_mac & gnutls_require_protocols (no longer supported).
2748 Added SNI support via GnuTLS too.
2749 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
2751 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
2753 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
2754 Applied second patch from Janne, implementing suggestion to default
2755 multiple-strings-in-record handling to match SPF spec.
2757 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
2759 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
2760 read-only, out of scope).
2761 Patch from Wolfgang Breyha, report from Stuart Northfield.
2763 PP/29 Fix three issues highlighted by clang analyser static analysis.
2764 Only crash-plausible issue would require the Cambridge-specific
2765 iplookup router and a misconfiguration.
2766 Report from Marcin Mirosław.
2768 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
2770 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
2771 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
2772 As part of this, removing so much warning spew let me fix some minor
2773 real issues in debug logging.
2775 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
2776 assignment on my part. Fixed.
2778 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
2779 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
2780 Janne Snabb (who went above and beyond: thank you).
2782 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
2783 string otherwise requires a connection and a bunch more work and it's
2784 relatively easy to get wrong. Should also expose TLS library linkage
2787 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
2788 64-bit ${eval} (JH/03).
2790 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
2791 GNU libc to support some of the 64-bit stuff, should not lead to
2792 conflicts. Defined before os.h is pulled in, so if a given platform
2793 needs to override this, it can.
2795 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
2796 protection layer was required, which is not implemented.
2797 Bugzilla 1254, patch from Wolfgang Breyha.
2799 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
2800 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
2801 tls_dhparam take prime identifiers. Also unbreak combination of
2802 OpenSSL+DH_params+TLSSNI.
2804 PP/39 Disable SSLv2 by default in OpenSSL support.
2810 PP/01 Solaris build fix for Oracle's LDAP libraries.
2811 Bugzilla 1109, patch from Stephen Usher.
2813 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
2815 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
2818 TF/02 Fix a couple more cases where we did not log the error message
2819 when unlink() failed. See also change 4.74-TF/03.
2821 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
2822 lock up or crash if it happened to be inside a call to libc when it
2823 got a SIGUSR1 from exiwhat.
2825 The SIGUSR1 handler appends the current process status to the process
2826 log which is later printed by exiwhat. It used to use the general
2827 purpose logging code to do this, but several functions it calls are
2828 not safe for signals.
2830 The new output code in the SIGUSR1 handler is specific to the process
2831 log, and simple enough that it's easy to inspect for signal safety.
2832 Removing some special cases also simplifies the general logging code.
2833 Removing the spurious timestamps from the process log simplifies
2836 TF/04 Improved ratelimit ACL condition.
2838 The /noupdate option has been deprecated in favour of /readonly which
2839 has clearer semantics. The /leaky, /strict, and /readonly update modes
2840 are mutually exclusive. The update mode is no longer included in the
2841 database key; it just determines when the database is updated. (This
2842 means that when you upgrade Exim will forget old rate measurements.)
2844 Exim now checks that the per_* options are used with an update mode that
2845 makes sense for the current ACL. For example, when Exim is processing a
2846 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
2847 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
2848 must specify per_mail/readonly. If you omit the update mode it defaults to
2849 /leaky where that makes sense (as before) or /readonly where required.
2851 The /noupdate option is now undocumented but still supported for
2852 backwards compatibility. It is equivalent to /readonly except that in
2853 ACLs where /readonly is required you may specify /leaky/noupdate or
2854 /strict/noupdate which are treated the same as /readonly.
2856 A useful new feature is the /count= option. This is a generalization
2857 of the per_byte option, so that you can measure the throughput of other
2858 aggregate values. For example, the per_byte option is now equivalent
2859 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
2861 The per_rcpt option has been generalized using the /count= mechanism
2862 (though it's more complicated than the per_byte equivalence). When it is
2863 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
2864 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
2865 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
2866 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
2867 Note that using per_rcpt with a non-readonly update mode in more than
2868 one ACL will cause the recipients to be double-counted. (The per_mail
2869 and per_byte options don't have this problem.)
2871 The handling of very low rates has changed slightly. If the computed rate
2872 is less than the event's count (usually one) then this event is the first
2873 after a long gap. In this case the rate is set to the same as this event's
2874 count, so that the first message of a spam run is counted properly.
2876 The major new feature is a mechanism for counting the rate of unique
2877 events. The new per_addr option counts the number of different
2878 recipients that someone has sent messages to in the last time period. It
2879 behaves like per_rcpt if all the recipient addresses are different, but
2880 duplicate recipient addresses do not increase the measured rate. Like
2881 the /count= option this is a general mechanism, so the per_addr option
2882 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
2883 example, measure the rate that a client uses different sender addresses
2884 with the options per_mail/unique=$sender_address. There are further
2885 details in the main documentation.
2887 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
2889 TF/06 Removed a few PCRE remnants.
2891 TF/07 Automatically extract Exim's version number from tags in the git
2892 repository when doing development or release builds.
2894 PP/02 Raise smtp_cmd_buffer_size to 16kB.
2895 Bugzilla 879. Patch from Paul Fisher.
2897 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
2898 Heavily based on revision 40f9a89a from Simon Arlott's tree.
2901 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
2903 PP/05 Variable $av_failed, true if the AV scanner deferred.
2904 Bugzilla 1078. Patch from John Horne.
2906 PP/06 Stop make process more reliably on build failure.
2907 Bugzilla 1087. Patch from Heiko Schlittermann.
2909 PP/07 Make maildir_use_size_file an _expandable_ boolean.
2910 Bugzilla 1089. Patch from Heiko Schlittermann.
2912 PP/08 Handle ${run} returning more data than OS pipe buffer size.
2913 Bugzilla 1131. Patch from Holger Weiß.
2915 PP/09 Handle IPv6 addresses with SPF.
2916 Bugzilla 860. Patch from Wolfgang Breyha.
2918 PP/10 GnuTLS: support TLS 1.2 & 1.1.
2920 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
2923 PP/11 match_* no longer expand right-hand-side by default.
2924 New compile-time build option, EXPAND_LISTMATCH_RHS.
2925 New expansion conditions, "inlist", "inlisti".
2927 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
2929 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
2931 PP/14 fix log_write() format string regression from TF/03.
2932 Bugzilla 1152. Patch from Dmitry Isaikin.
2938 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
2940 PP/02 Harmonised TLS library version reporting; only show if debugging.
2941 Layout now matches that introduced for other libraries in 4.74 PP/03.
2943 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
2945 PP/04 New "dns_use_edns0" global option.
2947 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
2950 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
2951 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
2953 TK/01 Updated PolarSSL code to 0.14.2.
2954 Bugzilla 1097. Patch from Andreas Metzler.
2956 PP/07 Catch divide-by-zero in ${eval:...}.
2957 Fixes bugzilla 1102.
2959 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
2962 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
2963 format-string attack -- SECURITY: remote arbitrary code execution.
2965 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
2966 time unintentionally subject to list matching rules, letting the header
2967 cause arbitrary Exim lookups (of items which can occur in lists, *not*
2968 arbitrary string expansion). This allowed for information disclosure.
2970 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
2971 INT_MIN/-1 -- value coerced to INT_MAX.
2977 NM/01 Workaround for PCRE version dependency in version reporting
2980 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
2981 This fixes portability to compilers other than gcc, notably
2982 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
2984 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
2985 makefiles for portability to HP-UX and POSIX correctness.
2987 PP/01 Permit LOOKUP_foo enabling on the make command-line.
2988 Also via indented variable definition in the Makefile.
2989 (Debugging by Oliver Heesakkers).
2991 PP/02 Restore caching of spamd results with expanded spamd_address.
2992 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
2994 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
2995 Improves build reliability. Fix from: Frank Elsner
2997 NM/02 Fix wide character breakage in the rfc2047 coding
2998 Fixes bug 1064. Patch from Andrey N. Oktyabrski
3000 NM/03 Allow underscore in dnslist lookups
3001 Fixes bug 1026. Patch from Graeme Fowler
3003 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
3004 Code patches from Adam Ciarcinski of NetBSD.
3006 NM/04 Fixed exiqgrep to cope with mailq missing size issue
3009 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
3010 is logged, to avoid truncation. Patch from John Horne.
3012 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
3013 Patch from Jakob Hirsch.
3015 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
3016 SQL string expansion failure details.
3017 Patch from Andrey Oktyabrski.
3019 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
3020 Patch from Simon Arlott.
3022 PP/09 New lookups functionality failed to compile on old gcc which rejects
3023 extern declarations in function scope.
3024 Patch from Oliver Fleischmann
3026 PP/10 Use sig_atomic_t for flags set from signal handlers.
3027 Check getgroups() return and improve debugging.
3028 Fixed developed for diagnosis in bug 927 (which turned out to be
3031 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
3032 Patch from Mark Zealey.
3034 PP/12 Bugzilla 1056: Improved spamd server selection.
3035 Patch from Mark Zealey.
3037 PP/13 Bugzilla 1086: Deal with maildir quota file races.
3038 Based on patch from Heiko Schlittermann.
3040 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
3041 Patch from Uwe Doering, sign-off by Michael Haardt.
3043 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
3044 variable declaration deep within a block. Bug and patch from
3047 PP/15 lookups-Makefile IRIX compatibility coercion.
3049 PP/16 Make DISABLE_DKIM build knob functional.
3051 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
3052 Patch by Simon Arlott
3054 TF/03 Fix valgrind.h portability to C89 compilers that do not support
3055 variable argument macros. Our copy now differs from upstream.
3061 TF/01 Failure to get a lock on a hints database can have serious
3062 consequences so log it to the panic log.
3064 TF/02 Log LMTP confirmation messages in the same way as SMTP,
3065 controlled using the smtp_confirmation log selector.
3067 TF/03 Include the error message when we fail to unlink a spool file.
3069 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
3070 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
3071 for maintaining out-of-tree patches for some time.
3073 PP/01 Bugzilla 139: Documentation and portability issues.
3074 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
3075 Handle per-OS dynamic-module compilation flags.
3077 PP/02 Let /dev/null have normal permissions.
3078 The 4.73 fixes were a little too stringent and complained about the
3079 permissions on /dev/null. Exempt it from some checks.
3080 Reported by Andreas M. Kirchwitz.
3082 PP/03 Report version information for many libraries, including
3083 Exim version information for dynamically loaded libraries. Created
3084 version.h, now support a version extension string for distributors
3085 who patch heavily. Dynamic module ABI change.
3087 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
3088 privilege escalation vulnerability whereby the Exim run-time user
3089 can cause root to append content of the attacker's choosing to
3092 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
3095 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
3096 If dropping privileges for untrusted macros, we disabled normal logging
3097 on the basis that it would fail; for the Exim run-time user, this is not
3098 the case, and it resulted in successful deliveries going unlogged.
3099 Fixed. Reported by Andreas Metzler.
3105 PP/01 Date: & Message-Id: revert to normally being appended to a message,
3106 only prepend for the Resent-* case. Fixes regression introduced in
3107 Exim 4.70 by NM/22 for Bugzilla 607.
3109 PP/02 Include check_rfc2047_length in configure.default because we're seeing
3110 increasing numbers of administrators be bitten by this.
3112 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
3114 PP/03 Bugzilla 994: added openssl_options main configuration option.
3116 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
3118 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
3120 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
3122 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
3123 without a peer certificate, leading to a segfault because of an
3124 assumption that peers always have certificates. Be a little more
3125 paranoid. Problem reported by Martin Tscholak.
3127 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
3128 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
3129 NB: ClamAV planning to remove STREAM in "middle of 2010".
3130 CL also introduces -bmalware, various -d+acl logging additions and
3131 more caution in buffer sizes.
3133 PP/09 Implemented reverse_ip expansion operator.
3135 PP/10 Bugzilla 937: provide a "debug" ACL control.
3137 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
3139 PP/12 Bugzilla 973: Implement --version.
3141 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
3143 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
3145 PP/15 Bugzilla 816: support multiple condition rules on Routers.
3147 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
3148 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
3149 ignore trailing whitespace.
3151 JJ/02 prevent non-panic DKIM error from being sent to paniclog
3153 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
3156 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
3157 Notification from Dr Andrew Aitchison.
3159 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
3160 ExtendedDetectionInfo response format.
3161 Notification from John Horne.
3163 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
3166 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
3167 XSL and documented dependency on system catalogs, with examples of how
3170 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
3173 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
3174 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
3175 configuration file which is writeable by the Exim user or group.
3177 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
3178 of configuration files to cover files specified with the -C option if
3179 they are going to be used with root privileges, not just the default
3182 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
3183 option (effectively making it always true).
3185 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
3186 files to be used while preserving root privileges.
3188 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
3189 that rogue child processes cannot use them.
3191 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
3192 run-time user, instead of root.
3194 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
3195 Exim run-time user without dropping privileges.
3197 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
3198 result string, instead of calling string_vformat() twice with the same
3201 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
3202 for other users. Others should always drop root privileges if they use
3203 -C on the command line, even for a whitelisted configure file.
3205 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
3207 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
3213 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
3214 $data_path, and $header_path variables; fixed documentation bugs and
3217 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
3218 exipick to access non-standard spools, including the "frozen" queue
3221 NM/01 Bugzilla 965: Support mysql stored procedures.
3222 Patch from Alain Williams
3224 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
3226 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
3227 Patch from Andreas Metzler
3229 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
3230 Patch from Kirill Miazine
3232 NM/05 Bugzilla 671: Added umask to procmail example.
3234 JJ/03 installed exipick 20100323.0, fixing doc bug
3236 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
3237 directory. Notification and patch from Dan Rosenberg.
3239 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
3241 TK/02 Improve log output when DKIM signing operation fails.
3243 MH/01 Treat the transport option dkim_domain as a colon separated
3244 list, not as a single string, and sign the message with each element,
3245 omitting multiple occurences of the same signer.
3247 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
3248 Bugzilla 985, 986. Patch by Simon Arlott
3250 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
3251 Patch by Simon Arlott
3253 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
3254 MBX locking. Notification from Dan Rosenberg.
3260 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
3262 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
3264 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
3266 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
3268 NM/04 Bugzilla 915: Flag broken perl installation during build.
3274 TK/01 Added patch by Johannes Berg that expands the main option
3275 "spamd_address" if it starts with a dollar sign.
3277 TK/02 Write list of recipients to X-Envelope-Sender header when building
3278 the mbox-format spool file for content scanning (suggested by Jakob
3281 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
3282 (http://www.dcc-servers.net/) support via dccifd. Activated by
3283 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
3285 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
3286 by Mark Daniel Reidel <mr@df.eu>.
3288 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
3289 When building exim an external PCRE library is now needed -
3290 PCRE is a system library on the majority of modern systems.
3291 See entry on PCRE_LIBS in EDITME file.
3293 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
3294 conversation. Added nologin parameter to request.
3295 Patch contributed by Kirill Miazine.
3297 TF/01 Do not log submission mode rewrites if they do not change the address.
3299 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
3301 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
3302 log files in place. Contributed by Roberto Lima.
3304 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
3306 TF/03 Bugzilla 615: When checking the local_parts router precondition
3307 after a local_part_suffix or local_part_prefix option, Exim now
3308 does not use the address's named list lookup cache, since this
3309 contains cached lookups for the whole local part.
3311 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
3312 Robert Millan. Documentation is in experimental-spec.txt.
3314 TF/04 Bugzilla 668: Fix parallel build (make -j).
3316 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
3318 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
3319 Patch provided by Jan Srzednicki.
3321 TF/05 Leading white space used to be stripped from $spam_report which
3322 wrecked the formatting. Now it is preserved.
3324 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
3325 that they are available at delivery time.
3327 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
3329 TF/08 TLS error reporting now respects the incoming_interface and
3330 incoming_port log selectors.
3332 TF/09 Produce a more useful error message if an SMTP transport's hosts
3333 setting expands to an empty string.
3335 NM/06 Bugzilla 744: EXPN did not work under TLS.
3336 Patch provided by Phil Pennock.
3338 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
3339 Patch provided by Richard Godbee.
3341 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
3342 acl_smtp_notquit, added index entry.
3344 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
3345 Patch provided by Eugene Bujak.
3347 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
3348 accept(). Patch provided by Maxim Dounin.
3350 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
3351 Patch provided by Phil Pennock.
3353 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
3355 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
3356 Patch provided by Brad "anomie" Jorsch.
3358 NM/14 Bugzilla 622: Added timeout setting to transport filter.
3359 Patch provided by Dean Brooks.
3361 TK/05 Add native DKIM support (does not depend on external libraries).
3363 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
3364 Patch provided by Graeme Fowler.
3366 NM/16 Bugzilla 851: Documentation example syntax fix.
3368 NM/17 Changed NOTICE file to remove references to embedded PCRE.
3370 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
3373 NM/19 Bugzilla 745: TLS version reporting.
3374 Patch provided by Phil Pennock.
3376 NM/20 Bugzilla 167: bool: condition support.
3377 Patch provided by Phil Pennock.
3379 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
3380 clients. Patch provided by Phil Pennock.
3382 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
3383 Patch provided by Brad "anomie" Jorsch.
3385 NM/23 Bugzilla 687: Fix misparses in eximstats.
3386 Patch provided by Heiko Schlittermann.
3388 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
3389 Patch provided by Heiko Schlittermann.
3391 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
3392 plus update to original patch.
3394 NM/26 Bugzilla 799: Documentation correction for ratelimit.
3396 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
3397 Patch provided by David Brownlee.
3399 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
3401 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
3403 NM/30 Bugzilla 888: TLS documentation bugfixes.
3405 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
3407 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
3408 Unlike the original bugzilla I have changed all shell scripts in src tree.
3410 NM/33 Bugzilla 898: Transport filter timeout fix.
3411 Patch by Todd Rinaldo.
3413 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
3414 Patch by Serge Demonchaux.
3416 NM/35 Bugzilla 39: Base64 decode bug fixes.
3417 Patch by Jakob Hirsch.
3419 NM/36 Bugzilla 909: Correct connect() call in dcc code.
3421 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
3423 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
3425 NM/39 Bugzilla 911: Fixed MakeLinks build script.
3431 TK/01 Add preliminary DKIM support. Currently requires a forked version of
3432 ALT-N's libdkim that I have put here:
3433 http://duncanthrax.net/exim-experimental/
3435 Note to Michael Haardt: I had to rename some vars in sieve.c. They
3436 were called 'true' and it seems that C99 defines that as a reserved
3437 keyword to be used with 'bool' variable types. That means you could
3438 not include C99-style headers which use bools without triggering
3439 build errors in sieve.c.
3441 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
3442 as mailq or other aliases. Changed the --help handling significantly
3443 to do whats expected. exim_usage() emits usage/help information.
3445 SC/01 Added the -bylocaldomain option to eximstats.
3447 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
3449 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
3451 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
3457 PH/01 Another patch from the Sieve maintainer.
3459 PH/02 When an IPv6 address is converted to a string for single-key lookup
3460 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
3461 dots are used instead of colons so that keys in lsearch files need not
3462 contain colons. This was done some time before quoting was made available
3463 in lsearch files. However, iplsearch files do require colons in IPv6 keys
3464 (notated using the quote facility) so as to distinguish them from IPv4
3465 keys. This meant that lookups for IP addresses in host lists did not work
3466 for iplsearch lookups.
3468 This has been fixed by arranging for IPv6 addresses to be expressed with
3469 colons if the lookup type is iplsearch. This is not incompatible, because
3470 previously such lookups could never work.
3472 The situation is now rather anomalous, since one *can* have colons in
3473 ordinary lsearch keys. However, making the change in all cases is
3474 incompatible and would probably break a number of configurations.
3476 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
3479 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
3480 conversion specification without a maximum field width, thereby enabling
3481 a rogue spamd server to cause a buffer overflow. While nobody in their
3482 right mind would setup Exim to query an untrusted spamd server, an
3483 attacker that gains access to a server running spamd could potentially
3484 exploit this vulnerability to run arbitrary code as the Exim user.
3486 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
3487 $primary_hostname instead of what libspf2 thinks the hosts name is.
3489 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
3490 a directory entry by the name of the lookup key. Previously, if a
3491 symlink pointed to a non-existing file or a file in a directory that
3492 Exim lacked permissions to read, a lookup for a key matching that
3493 symlink would fail. Now it is enough that a matching directory entry
3494 exists, symlink or not. (Bugzilla 503.)
3496 PH/03 The body_linecount and body_zerocount variables are now exported in the
3499 PH/04 Added the $dnslist_matched variable.
3501 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
3502 This means they are set thereafter only if the connection becomes
3505 PH/06 Added the client_condition to authenticators so that some can be skipped
3506 by clients under certain conditions.
3508 PH/07 The error message for a badly-placed control=no_multiline_responses left
3509 "_responses" off the end of the name.
3511 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
3513 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
3514 (without spaces) instead of just copying the configuration text.
3516 PH/10 Added the /noupdate option to the ratelimit ACL condition.
3518 PH/11 Added $max_received_linelength.
3520 PH/12 Added +ignore_defer and +include_defer to host lists.
3522 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
3523 way in which PCRE > 7.0 is built.
3525 PH/14 Implemented queue_only_load_latch.
3527 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
3528 MAIL command. The effect was to mangle the value on 64-bit systems.
3530 PH/16 Another patch from the Sieve maintainer.
3532 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
3534 PH/18 If a system quota error occurred while trying to create the file for
3535 a maildir delivery, the message "Mailbox is full" was not appended to the
3536 bounce if the delivery eventually timed out. Change 4.67/27 below applied
3537 only to a quota excession during the actual writing of the file.
3539 PH/19 It seems that peer DN values may contain newlines (and other non-printing
3540 characters?) which causes problems in log lines. The DN values are now
3541 passed through string_printing() before being added to log lines.
3543 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
3544 and InterBase are left for another time.)
3546 PH/21 Added message_body_newlines option.
3548 PH/22 Guard against possible overflow in moan_check_errorcopy().
3550 PH/23 POSIX allows open() to be a macro; guard against that.
3552 PH/24 If the recipient of an error message contained an @ in the local part
3553 (suitably quoted, of course), incorrect values were put in $domain and
3554 $local_part during the evaluation of errors_copy.
3560 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
3561 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
3564 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
3565 issue a MAIL command.
3567 PH/02 In an ACL statement such as
3569 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
3571 if a client was not listed at all, or was listed with a value other than
3572 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
3573 the condition was not true (as it should be), so access was not denied.
3574 The bug was that the ! inversion was incorrectly passed on to the second
3575 item. This has been fixed.
3577 PH/03 Added additional dnslists conditions == and =& which are different from
3578 = and & when the dns lookup returns more than one IP address.
3580 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
3581 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
3583 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
3584 FSYNC, which compiles an option called disable_fsync that allows for
3585 bypassing fsync(). The documentation is heavily laced with warnings.
3587 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
3589 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
3590 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
3591 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
3592 including adding "make clean"; (3) Added -fPIC when compiling the test
3593 dynamically loaded module, to get rid of a warning.
3595 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
3596 message fails, move_frozen_messages = true and ignore_bounce_errors_after
3597 = 0s. The bug is otherwise harmless.
3599 PH/07 There was a bug in the dovecot authenticator such that the value of
3600 $auth1 could be overwritten, and so not correctly preserved, after a
3601 successful authentication. This usually meant that the value preserved by
3602 the server_setid option was incorrect.
3604 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
3606 PH/09 Installed PCRE release 7.0.
3608 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
3609 run for batched SMTP input. It is now run at the start of every message
3610 in the batch. While fixing this I discovered that the process information
3611 (output by running exiwhat) was not always getting set for -bs and -bS
3612 input. This is fixed, and it now also says "batched" for BSMTP.
3614 PH/11 Added control=no_pipelining.
3616 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
3617 patch, slightly modified), and move the expansion of helo_data till after
3618 the connection is made in the smtp transport (so it can use these
3621 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
3623 PH/14 Added log_selector = +pid.
3625 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
3627 PH/16 Add ${if forany and ${if forall.
3629 PH/17 Added dsn_from option to vary the From: line in DSNs.
3631 PH/18 Flush SMTP output before performing a callout, unless control =
3632 no_callout_flush is set.
3634 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
3635 was true (the default) a successful delivery failed to delete the retry
3636 item, thus causing premature timeout of the address. The bug is now
3639 PH/20 Added hosts_avoid_pipelining to the smtp transport.
3641 PH/21 Long custom messages for fakedefer and fakereject are now split up
3642 into multiline responses in the same way that messages for "deny" and
3643 other ACL rejections are.
3645 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
3646 with slight modification.
3648 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
3649 draft, changing the syntax and factoring some duplicate code".
3651 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
3652 for deliveries of the second and subsequent messages over the same SMTP
3655 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
3656 ${reduce, with only minor "tidies".
3658 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
3660 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
3661 expansion side effects.
3663 PH/27 When a message times out after an over-quota error from an Exim-imposed
3664 quota, the bounce message says "mailbox is full". This message was not
3665 being given when it was a system quota that was exceeded. It now should
3668 MH/03 Made $recipients available in local_scan(). local_scan() already has
3669 better access to the recipient list through recipients_list[], but
3670 $recipients can be useful in postmaster-provided expansion strings.
3672 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
3673 in the case of a MAIL command with additional options following the
3674 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
3675 were accidentally chopped off.
3677 PH/29 SMTP synchronization checks are implemented when a command is read -
3678 there is a check that no more input is waiting when there shouldn't be
3679 any. However, for some commands, a delay in an ACL can mean that it is
3680 some time before the response is written. In this time, more input might
3681 arrive, invalidly. So now there are extra checks after an ACL has run for
3682 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
3683 pipelining has not been advertised.
3685 PH/30 MH's patch to allow iscntrl() characters to be list separators.
3687 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
3688 returned in the SMTP response when smtp_return_error_details was false.
3689 This has been fixed.
3691 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
3692 instead of the C I/O that was originally supplied, because problems were
3693 reported on Solaris.
3695 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
3696 Exim which did not show up earlier: it was assuming that a call to
3697 SSL_CTX_set_info_callback() might give an error value. In fact, there is
3698 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
3699 was a macro that became an assignment, so it seemed to work. This has
3700 changed to a proper function call with a void return, hence the compile
3701 error. Exim's code has been fixed.
3703 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
3706 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
3708 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
3710 PH/37 If a message is not accepted after it has had an id assigned (e.g.
3711 because it turns out to be too big or there is a timeout) there is no
3712 "Completed" line in the log. When some messages of this type were
3713 selected by exigrep, they were listed as "not completed". Others were
3714 picked up by some special patterns. I have improved the selection
3715 criteria to be more general.
3717 PH/38 The host_find_failed option in the manualroute router can now be set
3718 to "ignore", to completely ignore a host whose IP address cannot be
3719 found. If all hosts are ignored, the behaviour is controlled by the new
3720 host_all_ignored option.
3722 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
3723 homing or because of multiple MX records with /mx) generated more than
3724 one IP address, and the following item turned out to be the local host,
3725 all the secondary addresses of the first item were incorrectly removed
3726 from the list, along with the local host and any following hosts (which
3727 is what is supposed to happen).
3729 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
3730 whoever called Exim into the -H file. In the case of the daemon it was
3731 behaving confusingly. When first started, it used values for whoever
3732 started the daemon, but after a SIGHUP it used the Exim user (because it
3733 calls itself on a restart). I have changed the code so that it now always
3736 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
3737 message are rejected with the same error (e.g. no authentication or bad
3738 sender address), and a DATA command is nevertheless sent (as can happen
3739 with PIPELINING or a stupid MUA), the error message that was given to the
3740 RCPT commands is included in the rejection of the DATA command. This is
3741 intended to be helpful for MUAs that show only the final error to their
3744 PH/42 Another patch from the Sieve maintainer.
3746 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
3747 Eximstats - Fixed some broken HTML links and added missing column headers
3749 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
3750 columns for Rejects, Temp Rejects, Ham, and Spam rows.
3752 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
3754 PH/43 Yet another patch from the Sieve maintainer.
3756 PH/44 I found a way to check for a TCP/IP connection going away before sending
3757 the response to the final '.' that terminates a message, but only in the
3758 case where the client has not sent further data following the '.'
3759 (unfortunately, this is allowed). However, in many cases there won't be
3760 any further data because there won't be any more messages to send. A call
3761 to select() can be used: if it shows that the input is "ready", there is
3762 either input waiting, or the socket has been closed. An attempt to read
3763 the next input character can distinguish the two cases. Previously, Exim
3764 would have sent an OK response which the client would never have see.
3765 This could lead to message repetition. This fix should cure that, at
3766 least in a lot of common cases.
3768 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
3769 advertised in response to EHLO.
3775 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
3776 fixed by 4.65/MH/01 (is this a record?) are fixed:
3778 (i) An empty string was always treated as zero by the numeric comparison
3779 operators. This behaviour has been restored.
3781 (ii) It is documented that the numeric comparison operators always treat
3782 their arguments as decimal numbers. This was broken in that numbers
3783 starting with 0 were being interpreted as octal.
3785 While fixing these problems I realized that there was another issue that
3786 hadn't been noticed. Values of message_size_limit (both the global option
3787 and the transport option) were treated as octal if they started with 0.
3788 The documentation was vague. These values are now always treated as
3789 decimal, and I will make that clear in the documentation.
3795 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
3796 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
3799 MH/01 Don't check that the operands of numeric comparison operators are
3800 integers when their expansion is in "skipping" mode (fixes bug
3801 introduced by 4.64-PH/07).
3803 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
3804 child addresses, Exim now panics and dies. Previously, because the count
3805 is held in a short int, deliveries were likely to be lost. As such a
3806 large number of recipients for a single message is ridiculous
3807 (performance will be very, very poor), I have chosen to impose a limit
3808 rather than extend the field.
3814 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
3815 leftover -K file (the existence of which was triggered by #402).
3816 While we were at it, introduced process PID as part of the -K
3817 filename. This should rule out race conditions when creating
3820 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
3821 processing considerably. Previous code took too long for large mails,
3822 triggering a timeout which in turn triggers #401.
3824 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
3825 in the DK code in transports.c. sendfile() is not really portable,
3826 hence the _LINUX specificness.
3828 TF/01 In the add_headers option to the mail command in an Exim filter,
3829 there was a bug that Exim would claim a syntax error in any
3830 header after the first one which had an odd number of characters
3833 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
3834 callout verification, Exim cached a "reject" for the entire domain. This
3835 is correct for most verifications, but it is not correct for a recipient
3836 verification with use_sender or use_postmaster set, because in that case
3837 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
3838 case of MAIL FROM:<> rejection from other early rejections (e.g.
3839 rejection of HELO). When verifying a recipient using a non-null MAIL
3840 address, the cache is ignored if it shows MAIL FROM:<> rejection.
3841 Whatever the result of the callout, the value of the domain cache is
3842 left unchanged (for any other kind of callout, getting as far as trying
3843 RCPT means that the domain itself is ok).
3845 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
3848 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
3849 manifest itself as EPIPE rather than ECONNECT. When tidying away a
3850 session, the daemon ignores ECONNECT errors and logs others; it now
3851 ignores EPIPE as well.
3853 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
3854 (quoted-printable decoding).
3856 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
3857 later the small subsequent patch to fix an introduced bug.
3859 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
3861 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
3863 PH/08 An error is now given if message_size_limit is specified negative.
3865 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
3866 to be given (somewhat) arbitrary names.
3868 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
3871 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
3872 miscellaneous code fixes
3874 PH/10 Added the log_reject_target ACL modifier to specify where to log
3877 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
3878 hostname. This is wrong, because it relates to the incoming message (and
3879 probably the interface on which it is arriving) and not to the outgoing
3880 callout (which could be using a different interface). This has been
3881 changed to use the value of the helo_data option from the smtp transport
3882 instead - this is what is used when a message is actually being sent. If
3883 there is no remote transport (possible with a router that sets up host
3884 addresses), $smtp_active_hostname is used.
3886 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
3887 tweaks were necessary in order to get it to work (see also 21 below):
3888 (a) The code assumed that strncpy() returns a negative number on buffer
3889 overflow, which isn't the case. Replaced with Exim's string_format()
3891 (b) There were several signed/unsigned issues. I just did the minimum
3892 hacking in of casts. There is scope for a larger refactoring.
3893 (c) The code used strcasecmp() which is not a standard C function.
3894 Replaced with Exim's strcmpic() function.
3895 (d) The code set only $1; it now sets $auth1 as well.
3896 (e) A simple test gave the error "authentication client didn't specify
3897 service in request". It would seem that Dovecot has changed its
3898 interface. Fortunately there's a specification; I followed it and
3899 changed what the client sends and it appears to be working now.
3901 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
3904 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
3905 address A is aliased to B and C, where B exists and C does not. Without
3906 -v the output is "A verified" because verification stops after a
3907 successful redirection if more than one address is generated. However,
3908 with -v the child addresses are also verified. Exim was outputting "A
3909 failed to verify" and then showing the successful verification for C,
3910 with its parentage. It now outputs "B failed to verify", showing B's
3911 parentage before showing the successful verification of C.
3913 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
3914 look up a TXT record in a specific list after matching in a combined
3917 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
3918 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
3919 they consult the DNS. I had assumed they would set it the way they
3920 wanted; and indeed my experiments on Linux seem to show that in some
3921 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
3922 To be on the safe side, however, I have now made the interface to
3923 host_find_byname() similar to host_find_bydns(), with an argument
3924 containing the DNS resolver options. The host_find_byname() function now
3925 sets these options at its start, just as host_find_bydns() does. The smtp
3926 transport options dns_qualify_single and dns_search_parents are passed to
3927 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
3928 of host_find_byname() use the default settings of RES_DEFNAMES
3929 (qualify_single) but not RES_DNSRCH (search_parents).
3931 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
3932 spool_read_header() do less string testing, by means of a preliminary
3933 switch on the second character of optional "-foo" lines. (This is
3934 overdue, caused by the large number of possibilities that now exist.
3935 Originally there were few.) While I was there, I also converted the
3936 str(n)cmp tests so they don't re-test the leading "-" and the first
3937 character, in the hope this might squeeze out yet more improvement.
3939 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
3940 flag allowing group syntax was set by the header_syntax check but not
3941 turned off, possible causing trouble later; (2) The flag was not being
3942 set at all for the header_verify test, causing "group"-style headers to
3943 be rejected. I have now set it in this case, and also caused header_
3944 verify to ignore an empty address taken from a group. While doing this, I
3945 came across some other cases where the code for allowing group syntax
3946 while scanning a header line wasn't quite right (mostly, not resetting
3947 the flag correctly in the right place). These bugs could have caused
3948 trouble for malformed header lines. I hope it is now all correct.
3950 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
3951 with the "reply" argument non-NULL. The code, however (which originally
3952 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
3953 but it didn't always do it. This confused somebody who was copying the
3954 code for some other use. I have removed all the tests.
3956 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
3957 feature that was used to support insecure browsers during the U.S. crypto
3958 embargo. It requires special client support, and Exim is probably the
3959 only MTA that supported it -- and would never use it because real RSA is
3960 always available. This code has been removed, because it had the bad
3961 effect of slowing Exim down by computing (never used) parameters for the
3962 RSA_EXPORT functionality.
3964 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
3965 authenticator to fail if there's a tab character in the incoming data
3966 (there should never be unless someone is messing about, as it's supposed
3967 to be base64-encoded). Also added, on Timo's advice, the "secured" option
3968 if the connection is using TLS or if the remote IP is the same as the
3969 local IP, and the "valid-client-cert option" if a client certificate has
3972 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
3973 authenticators. This can be used for authorization after authentication
3974 succeeds. (In the case of plaintext, it servers for both authentication
3977 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
3978 if any retry times were supplied.
3980 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
3981 connection, where there is no client IP address to check. In this
3982 situation, the verify now always succeeds.
3984 PH/25 Applied John Jetmore's -Mset patch.
3986 PH/26 Added -bem to be like -Mset, but loading a message from a file.
3988 PH/27 In a string expansion for a processed (not raw) header when multiple
3989 headers of the same name were present, leading whitespace was being
3990 removed from all of them, but trailing whitespace was being removed only
3991 from the last one. Now trailing whitespace is removed from each header
3992 before concatenation. Completely empty headers in a concatenation (as
3993 before) are ignored.
3995 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
3996 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
3998 PH/29 [Removed. This was a change that I later backed out, and forgot to
3999 correct the ChangeLog entry (that I had efficiently created) before
4000 committing the later change.]
4002 PH/30 Exim was sometimes attempting to deliver messages that had suffered
4003 address errors (4xx response to RCPT) over the same connection as other
4004 messages routed to the same hosts. Such deliveries are always "forced",
4005 so retry times are not inspected. This resulted in far too many retries
4006 for the affected addresses. The effect occurred only when there were more
4007 hosts than the hosts_max_try setting in the smtp transport when it had
4008 the 4xx errors. Those hosts that it had tried were not added to the list
4009 of hosts for which the message was waiting, so if all were tried, there
4010 was no problem. Two fixes have been applied:
4012 (i) If there are any address or message errors in an SMTP delivery, none
4013 of the hosts (tried or untried) are now added to the list of hosts
4014 for which the message is waiting, so the message should not be a
4015 candidate for sending over the same connection that was used for a
4016 successful delivery of some other message. This seems entirely
4017 reasonable: after all the message is NOT "waiting for some host".
4018 This is so "obvious" that I'm not sure why it wasn't done
4019 previously. Hope I haven't missed anything, but it can't do any
4020 harm, as the worst effect is to miss an optimization.
4022 (ii) If, despite (i), such a delivery is accidentally attempted, the
4023 routing retry time is respected, so at least it doesn't keep
4024 hammering the server.
4026 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
4027 in ${readsocket because some servers need this prod.
4029 PH/32 Added some extra debug output when updating a wait-xxx database.
4031 PH/33 The hint "could be header name not terminated by colon", which has been
4032 given for certain expansion errors for a long time, was not being given
4033 for the ${if def:h_colon_omitted{... case.
4035 PH/34 The spec says: "With one important exception, whenever a domain list is
4036 being scanned, $domain contains the subject domain." There was at least
4037 one case where this was not true.
4039 PH/35 The error "getsockname() failed: connection reset by peer" was being
4040 written to the panic log as well as the main log, but it isn't really
4041 panic-worthy as it just means the connection died rather early on. I have
4042 removed the panic log writing for the ECONNRESET error when getsockname()
4045 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
4046 runs only) independently of the message's sender address. This meant
4047 that, if the 4xx error was in fact related to the sender, a different
4048 message to the same recipient with a different sender could confuse
4049 things. In particular, this can happen when sending to a greylisting
4050 server, but other circumstances could also provoke similar problems.
4051 I have changed the default so that the retry time for these errors is now
4052 based a combination of the sender and recipient addresses. This change
4053 can be overridden by setting address_retry_include_sender=false in the
4056 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
4057 remote server are returned as part of bounce messages. This was not
4058 happening for LMTP over a pipe (the lmtp transport), but now it is the
4059 same for both kinds of LMTP.
4061 PH/38 Despite being documented as not happening, Exim was rewriting addresses
4062 in header lines that were in fact CNAMEs. This is no longer the case.
4064 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
4065 and queue runs started by the daemon processed all messages. This has
4066 been fixed so that -R and -S can now usefully be given with -q<time>.
4068 PH/40 Import PCRE release 6.7 (fixes some bugs).
4070 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
4072 PH/42 Give an error if -q is specified more than once.
4074 PH/43 Renamed the variables $interface_address and $interface_port as
4075 $received_ip_address and $received_port, to make it clear that these
4076 values apply to message reception, and not to the outgoing interface when
4077 a message is delivered. (The old names remain recognized, of course.)
4079 PH/44 There was no timeout on the connect() call when using a Unix domain
4080 socket in the ${readsocket expansion. There now is.
4082 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
4083 be meaningful with "accept".
4085 SC/01 Eximstats V1.43
4086 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
4088 SC/02 Eximstats V1.44
4089 Use a glob alias rather than an array ref in the generated
4090 parser. This improves both readability and performance.
4092 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
4093 Collect SpamAssassin and rejection statistics.
4094 Don't display local sender or destination tables unless
4095 there is data to show.
4096 Added average volumes into the top table text output.
4098 SC/04 Eximstats V1.46
4099 Collect data on the number of addresses (recipients)
4100 as well as the number of messages.
4102 SC/05 Eximstats V1.47
4103 Added 'Message too big' to the list of mail rejection
4104 reasons (thanks to Marco Gaiarin).
4106 SC/06 Eximstats V1.48
4107 Mainlog lines which have GMT offsets and are too short to
4108 have a flag are now skipped.
4110 SC/07 Eximstats V1.49 (Alain Williams)
4111 Added the -emptyok flag.
4113 SC/08 Eximstats V1.50
4114 Fixes for obtaining the IP address from reject messages.
4116 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
4117 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
4118 whitespace changes from 4.64-PH/27
4120 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
4123 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
4124 are found, allow negative numbers in numeric criteria)
4126 JJ/06 exipick.20061117.2, added new $message_body_missing variable
4128 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
4129 to match changes made in 4.64-PH/43
4131 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
4133 PH/47 Put in an explicit test for a DNS lookup of an address record where the
4134 "domain" is actually an IP address, and force a failure. This locks out
4135 those revolvers/nameservers that support "A-for-A" lookups, in
4136 contravention of the specifications.
4138 PH/48 When a host name was looked up from an IP address, and the subsequent
4139 forward lookup of the name timed out, the host name was left in
4140 $sender_host_name, contrary to the specification.
4142 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
4143 restricted to single-key lookups, Exim was not diagnosing an error if
4144 * or *@ was used with a query-style lookup.
4146 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
4148 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
4149 long ago, but noone interested enough thought of it. Let's just say that
4150 the "1.1" means that there are some new functions that weren't there at
4151 some point in the past.
4153 PH/51 Error processing for expansion failure of helo_data from an smtp
4154 transport during callout processing was broken.
4156 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
4157 tested/used via the -bh/-bhc/-bs options.
4159 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
4160 bug, fixed in subsequent PCRE releases).
4162 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
4163 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
4165 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
4171 SC/01 Use a glob alias rather than an array ref in eximstats generated
4172 parser. This improves both readability and performance.
4174 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
4175 Don't display local sender or destination tables in eximstats unless
4176 there is data to show.
4177 Added average volumes into the eximstats top table text output.
4179 SC/03 Collect data on the number of addresses (recipients) as well
4180 as the number of messages in eximstats.
4182 TF/01 Correct an error in the documentation for the redirect router. Exim
4183 does (usually) call initgroups() when daemonizing.
4185 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
4186 with consistent privilege compared to when running as a daemon.
4188 TF/03 Note in the spec that $authenticated_id is not set for local
4189 submissions from trusted users.
4191 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
4192 Thanks to Dean Brooks <dean@iglou.com> for the patch.
4194 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
4195 by adding some example configuration directives to the default
4196 configuration file. A little bit of work is required to uncomment the
4197 directives and define how usernames and passwords are checked, but
4198 there is now a framework to start from.
4200 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
4201 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
4202 without this. I don't know how relevant this is to other LDAP libraries.
4204 PH/02 Add the verb name to the "unknown ACL verb" error.
4206 PH/03 Magnus Holmgren's patch for filter_prepend_home.
4208 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
4210 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
4211 directory not expanded when it should be if an expanded home directory
4212 was set for the address (which is overridden by the transport).
4214 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
4217 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
4218 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
4219 because it is too late at that time, and has no effect.
4221 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
4222 security issue with \' (bugzilla #107). I could not use the
4223 PQescapeStringConn() function, because it needs a PGconn value as one of
4226 PH/08 When testing addresses using -bt, indicate those final addresses that
4227 are duplicates that would not cause an additional delivery. At least one
4228 person was confused, thinking that -bt output corresponded to deliveries.
4229 (Suppressing duplicates isn't a good idea as you lose the information
4230 about possibly different redirections that led to the duplicates.)
4232 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
4233 systems where poll() doesn't work, in particular OS X.
4235 PH/10 Added more information to debugging output for retry time not reached.
4237 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
4238 operations in malware.c.
4240 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
4243 PH/13 If write_rejectlog was set false when logging was sent to syslog with
4244 syslog_duplication set false, log lines that would normally be written
4245 both the the main log and to the reject log were not written to syslog at
4248 PH/14 In the default configuration, change the use of "message" in ACL warn
4249 statements to "add_header".
4251 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
4252 not followed by a command (e.g. "seen endif").
4254 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
4255 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
4258 PH/17 Added extra conditions to the default value of delay_warning_condition
4262 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
4263 { match{$h_precedence:}{(?i)bulk|list|junk} } \
4264 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
4267 The Auto-Submitted: and various List- headers are standardised, whereas I
4268 don't think Precedence: ever was.
4270 PH/18 Refactored debugging code in route_finduser() to show more information,
4271 in particular, the error code if getpwnam() issues one.
4273 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
4274 This is apparently needed in addition to the PH/07 change above to avoid
4275 any possible encoding problems.
4277 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
4278 but not after initializing Perl.
4280 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
4281 output them only if debugging. By default they are written stderr,
4282 apparently, which is not desirable.
4284 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
4287 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
4290 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
4292 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
4293 authenticated or an ident call has been made. Suppress the default
4294 values for $authenticated_id and $authenticated_sender (but permit -oMai
4295 and -oMas) when testing with -bh.
4297 PH/24 Re-jigged the order of the tests in the default configuration so that the
4298 tests for valid domains and recipients precede the DNS black list and CSA
4299 tests, on the grounds that those ones are more expensive.
4301 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
4302 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
4303 This bug exists in every version of Exim that I still have, right back to
4306 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
4307 However, an attempt to turn on case-sensitivity in a regex key by
4308 including (?-i) didn't work because the subject string was already
4309 lowercased, and the effects were non-intuitive. It turns out that a
4310 one-line patch can be used to allow (?-i) to work as expected.
4316 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
4317 other effects) broke the use of negated acl sub-conditions.
4319 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
4322 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
4323 "Deny" causes Exim to reject the incoming connection with a 554 error.
4324 Unfortunately, if there is a major crisis, such as a disk failure,
4325 tcp-wrappers gives "deny", whereas what one would like would be some
4326 kind of temporary error. A kludge has been added to help with this.
4327 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
4328 554 error is used if errno is still zero or contains ENOENT (which occurs
4329 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
4332 PH/03 Add -lutil to the default FreeBSD LIBS setting.
4334 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
4335 errors. Otherwise a message that provokes a temporary error (when other
4336 messages do not) can cause a whole host to time out.
4338 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
4339 addresses were routed directly to files or pipes from a redirect router.
4340 File deliveries just didn't batch; pipe deliveries might have suffered
4343 PH/06 A failure to get a lock for a hints database would erroneously always say
4344 "Failed to get write lock", even when it was really a read lock.
4346 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
4347 of 0600. This has been changed to use the value of the lockfile_mode
4348 option (which defaults to 0600).
4350 PH/08 Applied small patch from the Sieve maintainer.
4352 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
4353 folder from quota calculations, a direct delivery into this folder messed
4354 up the contents of the maildirsize file. This was because the regex was
4355 used only to exclude .Trash (or whatever) when the size of the mailbox
4356 was calculated. There was no check that a delivery was happening into an
4357 excluded directory. This bug has been fixed by ignoring all quota
4358 processing for deliveries into excluded directories.
4360 PH/10 Added the maildirfolder_create_regex option to appendfile.
4366 PH/01 The code for finding all the local interface addresses on a FreeBSD
4367 system running IPv6 was broken. This may well have applied to all BSD
4368 systems, as well as to others that have similar system calls. The broken
4369 code found IPv4 interfaces correctly, but gave incorrect values for the
4370 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
4371 that it would not match correctly against @[] and not recognize the IPv6
4374 PH/02 The ipliteral router was not recognizing addresses of the form user@
4375 [ipv6:....] because it didn't know about the "ipv6:" prefix.
4377 PH/03 Added disable_ipv6.
4379 PH/04 Changed $reply_address to use the raw form of the headers instead of the
4380 decoded form, because it is most often used to construct To: headers
4381 lines in autoreplies, and the decoded form may well be syntactically
4382 invalid. However, $reply_address has leading white space removed, and all
4383 newlines turned into spaces so that the autoreply transport does not
4386 PH/05 If group was specified without a user on a router, and no group or user
4387 was specified on a transport, the group from the router was ignored.
4389 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
4390 for visible compile-time settings that can be used to change these
4391 numbers, for those that want even more. Backwards compatibility with old
4392 spool files has been maintained. However, going back to a previous Exim
4393 release will lost any variables that are in spool files.
4395 PH/07 Two small changes when running in the test harness: increase delay when
4396 passing a TCP/IP connection to a new process, in case the original
4397 process has to generate a bounce, and remove special handling of
4398 127.0.0.2 (sic), which is no longer necessary.
4400 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
4401 be the same on different OS.
4403 PH/09 Moved a debug statement in filter processing to avoid a race problem when
4406 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
4407 whether --show-vars was specified or not
4409 JJ/02 exipick: Added support for new ACL variable spool format introduced
4412 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
4413 syntactically invalid From: or Reply-to: line, and a filter used this to
4414 generate an autoreply, and therefore failed to obtain an address for the
4415 autoreply, Exim could try to deliver to a non-existent relative file
4416 name, causing unrelated and misleading errors. What now happens is that
4417 it logs this as a hard delivery error, but does not attempt to create a
4420 PH/11 The exinext utility has a -C option for testing purposes, but although
4421 the given file was scanned by exinext itself; it wasn't being passed on
4422 when Exim was called.
4424 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
4425 an end-of-file indication when reading a command response.
4427 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
4428 compiled. In many other places in Exim, IPv6 addresses are always
4429 recognized, so I have changed this. It also means that IPv4 domain
4430 literals of the form [IPV4:n.n.n.n] are now always recognized.
4432 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
4433 used if the router is not running as root, for example, when verifying at
4434 ACL time, or when using -bh. The debugging output from this situation was
4435 non-existent - all you got was a failure to exec. I have made two
4438 (a) Failures to set uid/gid, the current directory, or a process leader
4439 in a subprocess such as that created by queryprogram now generate
4440 suitable debugging output when -d is set.
4442 (b) The queryprogram router detects when it is not running as root,
4443 outputs suitable debugging information if -d is set, and then runs
4444 the subprocess without attempting to change uid/gid.
4446 PH/15 Minor change to Makefile for building test_host (undocumented testing
4449 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
4450 additional section of a DNS packet that returns MX or SRV records.
4451 Instead, it always explicitly searches for A/AAAA records. This avoids
4452 major problems that occur when a DNS server includes only records of one
4453 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
4454 fixed another bug: if SRV records were looked up and the corresponding
4455 address records were *not* found in the additional section, the port
4456 values from the SRV records were lost.
4458 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
4459 using the correct key (the original address) when searching the retry
4460 rules in order to find which one to use for generating the retry hint.
4462 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
4463 adding the default one. Similarly, if it contains a Reply-To: header, the
4464 errors_reply_to option, if set, is not used.
4466 PH/19 When calculating a retry time, Exim used to measure the "time since
4467 failure" by looking at the "first failed" field in the retry record. Now
4468 it does not use this if it is later than than the arrival time of the
4469 message. Instead it uses the arrival time. This makes for better
4470 behaviour in cases where some deliveries succeed, thus re-setting the
4471 "first failed" field. An example is a quota failure for a huge message
4472 when small messages continue to be delivered. Without this change, the
4473 "time since failure" will always be short, possible causing more frequent
4474 delivery attempts for the huge message than are intended.
4475 [Note: This change was subsequently modified - see PH/04 for 4.62.]
4477 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
4478 $1, $2, $3) because the numerical variables can be reset during some
4479 expansion items (e.g. "match"), thereby losing the authentication data.
4481 PH/21 Make -bV show the size of off_t variables so that the test suite can
4482 decide whether to run tests for quotas > 2G.
4484 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
4485 mailbox_size, and mailbox_filecount in the appendfile transport. If a
4486 filecount value is greater than 2G or if a quota value is greater than 2G
4487 on a system where the size of off_t is not greater than 4, a panic error
4490 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
4491 never match. The debug and -bh output now contains an explicit error
4492 message indicating a malformed IPv4 address or mask.
4494 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
4495 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
4496 PH/23 above applies.
4498 PH/25 Do not write to syslog when running in the test harness. The only
4499 occasion when this arises is a failure to open the main or panic logs
4500 (for which there is an explicit test).
4502 PH/26 Added the /no_tell option to "control=freeze".
4504 PH/27 If a host name lookup failed very early in a connection, for example, if
4505 the IP address matched host_lookup and the reverse lookup yielded a name
4506 that did not have a forward lookup, an error message of the form "no IP
4507 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
4508 could be logged. Now it outputs the IP address instead of "NULL".
4510 PH/28 An enabling patch from MH: add new function child_open_exim2() which
4511 allows the sender and the authenticated sender to be set when
4512 submitting a message from within Exim. Since child_open_exim() is
4513 documented for local_scan(), the new function should be too.
4515 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
4516 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
4517 results in an empty string is now treated as unset.
4519 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
4521 PH/31 Added sender_verify_fail logging option.
4523 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
4524 needed by RFC 822 but not by RFC 2822 was commented out. I have now
4525 tidied the source and removed it altogether.
4527 PH/33 When a queue run was abandoned because the load average was too high, a
4528 log line was always written; now it is written only if the queue_run log
4529 selector is set. In addition, the log line for abandonment now contains
4530 information about the queue run such as the pid. This is always present
4531 in "start" and "stop" lines but was omitted from the "abandon" line.
4533 PH/34 Omit spaces between a header name and the colon in the error message that
4534 is given when verify = headers_syntax fails (if there are lots of them,
4535 the message gets confusing).
4537 PH/35 Change the default for dns_check_names_pattern to allow slashes within
4538 names, as there are now some PTR records that contain slashes. This check
4539 is only to protect against broken name servers that fall over on strange
4540 characters, so the fact that it applies to all lookups doesn't matter.
4542 PH/36 Now that the new test suite is complete, we can remove some of the
4543 special code in Exim that was needed for the old test suite. For example,
4544 sorting DNS records because real resolvers return them in an arbitrary
4545 order. The new test suite's fake resolver always returns records in the
4548 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
4549 bounces) except when queue_only is set, to avoid logging races between
4550 the different processes.
4552 PH/38 Panic-die if .include specifies a non-absolute path.
4554 PH/39 A tweak to the "H" retry rule from its user.
4556 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
4557 a label. They prevented compilation on older perls.
4559 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
4560 a warning to be raised on newish perls.
4562 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
4563 on queue. Changes to match documented behaviour of showing count of
4564 messages matching specified criteria.
4566 PH/40 Changed the default ident timeout from 30s to 5s.
4568 PH/41 Added support for the use of login_cap features, on those BSD systems
4569 that have them, for controlling the resources used by pipe deliveries.
4571 PH/42 The content-scanning code uses fopen() to create files in which to put
4572 message data. Previously it was not paying any attention to the mode of
4573 the files. Exim runs with umask(0) because the rest of the code creates
4574 files with open(), and sets the required mode explicitly. Thus, these
4575 files were ending up world-writeable. This was not a big issue, because,
4576 being within the spool directory, they were not world-accessible. I have
4577 created a function called modefopen, which takes an additional mode
4578 argument. It sets umask(777), creates the file, chmods it to the required
4579 mode, then resets the umask. All the relevant calls to fopen() in the
4580 content scanning code have been changed to use this function.
4582 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
4583 to 24 hours. This avoids potential overflow problems when processing G
4584 and H retry rules. I suspect nobody ever tinkers with this value.
4586 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
4588 PH/45 When the plaintext authenticator is running as a client, the server's
4589 challenges are checked to ensure they are valid base64 strings. By
4590 default, the authentication attempt is cancelled if an invalid string is
4591 received. Setting client_ignore_invalid_base64 true ignores these errors.
4592 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
4593 they are received. Thus, the responses can be made to depend on the
4594 challenges. If an invalid string is ignored, an empty string is placed in
4597 PH/46 Messages that are created by the autoreply transport now contains a
4598 References: header, in accordance with RFCs 2822 and 3834.
4600 PH/47 Added authenticated_sender_force to the smtp transport.
4602 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
4604 PH/49 Installed latest patch from the Sieve maintainer.
4606 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
4607 was also set, the appendfile transport was unnecessarily scanning a
4608 directory of message files (e.g. for maildir delivery) to find the count
4609 of files (along with the size), even though it did not need this
4610 information. It now does the scan only if it needs to find either the
4611 size of the count of files.
4613 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
4615 PH/52 Two bugs concerned with error handling when the smtp transport is
4618 (i) Exim was not creating retry information for temporary errors given
4619 for individual recipients after the DATA command when the smtp transport
4620 was used in LMTP mode. This meant that they could be retried too
4621 frequently, and not timed out correctly.
4623 (ii) Exim was setting the flag that allows error details to be returned
4624 for LMTP errors on RCPT commands, but not for LMTP errors for individual
4625 recipients that were returned after the DATA command.
4627 PH/53 This is related to PH/52, but is more general: for any failing address,
4628 when detailed error information was permitted to be returned to the
4629 sender, but the error was temporary, then after the final timeout, only
4630 "retry timeout exceeded" was returned. Now it returns the full error as
4631 well as "retry timeout exceeded".
4633 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
4634 do this, and (what is worse) MTAs that accept it.
4636 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
4637 will now be deprecated.
4639 PH/56 New os.c-cygwin from the Cygwin maintainer.
4641 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
4642 formats (previously only available in exim formats via -bpr, -bpru,
4643 and -bpra. Now also available in native and exiqgrep formats)
4645 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
4646 with very large, slow to parse queues
4648 JJ/08 exipick: added ! as generic prefix to negate any criteria format
4650 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
4652 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
4653 responses to authentication challenges, though it was showing the
4654 challenges; (ii) I've removed the CR characters from the debug output for
4657 PH/58 Allow for the insertion of a newline as well as a space when a string
4658 is turned into more than one encoded-word during RFC 2047 encoding. The
4659 Sieve code now uses this.
4661 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
4662 data_4xx, lost_connection, tls_required.
4664 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
4665 message was being sent as an SMTP response.
4667 PH/61 Add -l and -k options to exicyclog.
4669 PH/62 When verifying, if an address was redirected to one new address, so that
4670 verification continued, and the new address failed or deferred after
4671 having set something in $address_data, the value of $address_data was not
4672 passed back to the ACL. This was different to the case when no
4673 redirection occurred. The value is now passed back in both cases.
4675 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
4676 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
4677 use login_cap.h, so on its own it isn't the distinguishing feature. The
4678 new name refers directly to the setclassresources() function.
4680 PH/65 Added configuration files for NetBSD3.
4682 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
4684 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
4685 is preferred over IPv4.
4687 PH/68 The bounce_return_message and bounce_return_body options were not being
4688 honoured for bounces generated during the reception of non-SMTP messages.
4689 In particular, this applied to messages rejected by the ACL. This bug has
4690 been fixed. However, if bounce_return_message is true and bounce_return_
4691 body is false, the headers that are returned for a non-SMTP message
4692 include only those that have been read before the error was detected.
4693 (In the case of an ACL rejection, they have all been read.)
4695 PH/69 The HTML version of the specification is now built in a directory called
4696 spec_html instead of spec.html, because the latter looks like a path with
4697 a MIME-type, and this confuses some software.
4699 PH/70 Catch two compiler warnings in sieve.c.
4701 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
4702 function verify_get_ident() calls ip_connect() to connect a socket, but
4703 if the "connect()" function timed out, ip_connect() used to close the
4704 socket. However, verify_get_ident() also closes the socket later, and in
4705 between Exim writes to the log, which may get opened at this point. When
4706 the socket was closed in ip_connect(), the log could get the same file
4707 descriptor number as the socket. This naturally causes chaos. The fix is
4708 not to close the socket in ip_connect(); the socket should be closed by
4709 the function that creates it. There was only one place in the code where
4710 this was missing, in the iplookup router, which I don't think anybody now
4711 uses, but I've fixed it anyway.
4713 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
4714 well as to direct DNS lookups. Otherwise the handling of names in host
4715 lists is inconsistent and therefore confusing.
4721 PH/01 Two changes to the default runtime configuration:
4723 (1) Move the checks for relay_from_hosts and authenticated clients from
4724 after to before the (commented out) DNS black list checks.
4726 (2) Add control=submission to the relay_from_hosts and authenticated
4727 clients checks, on the grounds that messages accepted by these
4728 statements are most likely to be submissions.
4730 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
4732 (1) Generate an error if the third argument for the ${prvs expansion is
4735 (2) Treat a missing third argument of ${prvscheck as if it were an empty
4738 (3) Reset the variables that are obtained from the first argument of
4739 ${prvscheck and used in the second argument before leaving the code,
4740 because their memory is reclaimed, so using them afterwards may do
4743 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
4744 one (it's much easier than Tom thought :-).
4746 (5) Because of (4), we can now allow for the use of $prvscheck_result
4747 inside the third argument.
4749 PH/03 For some reason, the default setting of PATH when running a command from
4750 a pipe transport was just "/usr/bin". I have changed it to
4753 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
4754 anything to be listed in the output from -bV.
4756 PH/05 When a filter generated an autoreply, the entire To: header line was
4757 quoted in the delivery log line, like this:
4759 => >A.N.Other <ano@some.domain> <original@ddress> ...
4761 This has been changed so that it extracts the operative address. There
4762 may be more than one such address. If so, they are comma-separated, like
4765 => >ano@some.domain,ona@other.domain <original@ddress> ...
4767 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
4768 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
4769 not being looked up in the rDNS to get a host name, Exim was showing the
4770 IP address twice in Received: lines, even though the IP addresses were
4771 identical. For example:
4773 Received: from [1.2.3.4] (helo=[1.2.3.4])
4775 However, if the real host name was known, it was omitting the HELO data
4776 if it matched the actual IP address. This has been tidied up so that it
4777 doesn't show the same IP address twice.
4779 PH/07 When both +timestamp and +memory debugging was on, the value given by
4780 $tod_xxx expansions could be wrong, because the tod_stamp() function was
4781 called by the debug printing, thereby overwriting the timestamp buffer.
4782 Debugging no longer uses the tod_stamp() function when +timestamp is set.
4784 PH/08 When the original message was included in an autoreply transport, it
4785 always said "this is a copy of the message, including all the headers",
4786 even if body_only or headers_only was set. It now gives an appropriate
4789 PH/09 Applied a patch from the Sieve maintainer which:
4791 o fixes some comments
4792 o adds the (disabled) notify extension core
4793 o adds some debug output for the result of if/elsif tests
4794 o points to the current vacation draft in the documentation
4795 and documents the missing references header update
4799 o fixes a bug in processing the envelope test (when testing
4800 multiple envelope elements, the last element determined the
4803 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
4804 Electronic Mail") by including:
4806 Auto-submitted: auto-generated
4808 in the messages that it generates (bounce messages and others, such as
4809 warnings). In the case of bounce messages for non-SMTP messages, there was
4810 also a typo: it was using "Auto_submitted" (underscore instead of
4811 hyphen). Since every message generated by Exim is necessarily in response
4812 to another message, thes have all been changed to:
4814 Auto-Submitted: auto-replied
4816 in accordance with these statements in the RFC:
4818 The auto-replied keyword:
4820 - SHOULD be used on messages sent in direct response to another
4821 message by an automatic process,
4823 - MUST NOT be used on manually-generated messages,
4825 - MAY be used on Delivery Status Notifications (DSNs) and Message
4826 Disposition Notifications (MDNs),
4828 - MUST NOT be used on messages generated by automatic or periodic
4829 processes, except for messages which are automatic responses to
4832 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
4833 to the default Received: header definition.
4835 PH/12 Added log selector acl_warn_skipped (default on).
4837 PH/13 After a successful wildlsearch lookup, discard the values of numeric
4838 variables because (a) they are in the wrong storage pool and (b) even if
4839 they were copied, it wouldn't work properly because of the caching.
4841 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
4842 checking when decoding. Apparently there are clients that generate
4843 overlong encoded strings. Why am I not surprised?
4845 PH/15 If the first argument of "${if match_address" was not empty, but did not
4846 contain an "@" character, Exim crashed. Now it writes a panic log message
4847 and treats the condition as false.
4849 PH/16 In autoreply, treat an empty string for "once" the same as unset.
4851 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
4852 extension "envelope-auth". The code is finished and in agreement with
4853 other implementations, but there is no documentation so far and in fact,
4854 nobody wrote the draft yet. This extension is currently #undef'ed, thus
4855 not changing the active code.
4857 Print executed "if" and "elsif" statements when debugging is used. This
4858 helps a great deal to understand what a filter does.
4860 Document more things not specified clearly in RFC3028. I had all this
4861 sorted out, when out of a sudden new issues came to my mind. Oops."
4863 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
4866 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
4867 canonical form (as documented). However, after a host name lookup from
4868 the IP address, check_host() was doing a simple string comparison with
4869 addresses acquired from the DNS when checking that the found name did
4870 have the original IP as one of its addresses. Since any found IPv6
4871 addresses are likely to be in abbreviated form, the comparison could
4872 fail. Luckily, there already exists a function for doing the comparison
4873 by converting both addresses to binary, so now that is used instead of
4874 the text comparison.
4876 PH/20 There was another similar case to PH/19, when a complete host name was
4877 given in a host list; looking up its IP address could give an abbreviated
4878 form, whereas the current host's name might or might not be abbreviated.
4879 The same fix has been applied.
4885 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
4886 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
4889 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
4890 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
4892 PH/03 Typo: missing ".o" in src/pcre/Makefile.
4894 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
4895 header line, restrict the check to what is listed in RFCs 2369 and 2929.
4896 Also, for "Auto-Submitted", treat anything other than "no" as
4897 non-personal, in accordance with RFC 3834. (Previously it treated
4898 anything starting "auto-" as non-personal.)
4900 TF/01 The control=submission/name=... option had a problem with syntax
4901 errors if the name included a slash character. The /name= option
4902 now slurps the rest of the string, so it can include any characters
4903 but it must come last in the list of options (after /sender_retain
4906 PH/05 Some modifications to the interface to the fake nameserver for the new
4914 TK/01 Added the "success_on_redirect" address verification option. See
4915 NewStuff for rationale and an example.
4917 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
4919 PH/02 Patch to exigrep to allow it to work on syslog lines.
4921 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
4922 fread() to skip over the body file's header line, because in Cygwin the
4923 header line is locked and is inaccessible.
4925 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
4926 co-exist for some time) to make it clear that it is the Exim ID that is
4927 referenced, not the Message-ID: header line.
4929 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
4930 string_format() function, because snprintf() does not exist on all
4933 PH/06 The use of forbid_filter_existstest now also locks out the use of the
4934 ${stat: expansion item.
4936 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
4937 protocol synchronization error", to keep the pedants happy.
4939 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
4940 well as for IRIX systems, when gcc is being used. See the host.c source
4943 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
4945 PH/10 Named domain lists were not working if used in a queue_smtp_domains
4948 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
4949 transport and to the smtp transport in LMTP mode.
4951 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
4953 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
4954 run a filter in a subprocess. This could lead to confusion in subsequent
4955 lookups in the parent process. There should also be a search_tidyup() at
4956 the end of the subprocess.
4958 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
4959 only if the host matched helo_try_verify_hosts, which caused the
4960 verification to occur when the EHLO/HELO command was issued. The ACL just
4961 tested the remembered result. Now, if a previous verification attempt has
4962 not happened, "verify = helo" does it there and then.
4964 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
4966 TK/03 Fix log output including CR from clamd.
4968 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
4969 exist provoked a memory error which could cause a segfault.
4971 PH/15 Installed PCRE 6.2
4973 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
4975 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
4976 of the problem. Specifically, suggested +O2 rather than +O1 for the
4979 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
4981 PH/20 If a delivery was routed to a non-standard port by means of an SRV
4982 record, the port was not correctly logged when the outgoing_port log
4983 selector was set (it logged the transort's default port).
4985 PH/21 Added support for host-specific ports to manualroute, queryprogram,
4986 fallback_hosts, and "hosts" in the smtp transport.
4988 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
4989 host errors such as "Connection refused".
4991 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
4992 authentication with radiusclient 0.4.9:
4994 - Error returned from rc_read_config was caught wrongly
4995 - Username/password not passed on to radius server due to wrong length.
4997 The presumption is that some radiusclient API changes for 4.51/PH/17
4998 were not taken care of correctly. The code is still untested by me (my
4999 Linux distribution still has 0.3.2 of radiusclient), but it was
5000 contributed by a Radius user.
5002 PH/24 When doing a callout, the value of $domain wasn't set correctly when
5003 expanding the "port" option of the smtp transport.
5005 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
5006 while reading a MIME header. Thanks to Tom Hughes for a patch.
5008 PH/24 Include config.h inside local_scan.h so that configuration settings are
5011 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
5012 that in an ACL for RCPT (for example), you can examine exactly what was
5015 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
5016 commands, but it was not correctly comparing the address with the actual
5017 client host address. Thus, it would show the EHLO address in Received:
5018 header lines when this was not necessary.
5020 PH/27 Added the % operator to ${eval:}.
5022 PH/28 Exim tries to create and chdir to its spool directory when it starts;
5023 it should be ignoring failures (because with -C, for example, it has lost
5024 privilege). It wasn't ignoring creation failures other than "already
5027 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
5030 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
5031 because an input line was too long, either on its own, or by virtue of
5032 too many continuations, the temporary file was not being removed, and the
5033 return code was incorrect.
5035 PH/31 Missing "BOOL" in function definition in filtertest.c.
5037 PH/32 Applied Sieve patches from the maintainer.
5039 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
5041 PH/33 Added "verify = not_blind".
5043 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
5044 Local/Makefile (with some defaults set). These are used in built scripts
5045 such as exicyclog, but they have never been used in the exim_install
5046 script (though there are many overriding facilities there). I have
5047 arranged that the exim_install script now takes note of these two
5050 PH/35 Installed configuration files for Dragonfly.
5052 PH/36 When a locally submitted message by a trusted user did not contain a
5053 From: header, and the sender address was obtained from -f or from an SMTP
5054 MAIL command, and the trusted user did not use -F to supply a sender
5055 name, $originator_name was incorrectly used when constructing a From:
5056 header. Furthermore, $originator_name was used for submission mode
5057 messages from external hosts without From: headers in a similar way,
5058 which is clearly wrong.
5060 PH/37 Added control=suppress_local_fixups.
5062 PH/38 When log_selector = +received_sender was set, and the addition of the
5063 sender made the log line's construction buffer exactly full, or one byte
5064 less than full, an overflow happened when the terminating "\n" was
5067 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
5068 when the result of a list match is failure because a DNS lookup failed.
5070 PH/40 RM_COMMAND is now used in the building process.
5072 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
5073 the "build-* directories that it finds.
5075 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
5076 address in a domain literal was a prefix of an interface address.
5078 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
5079 when verifying a sender address, unless rewrite_headers is false.
5081 PH/44 Wrote a long comment about why errors_to addresses are verified as
5082 recipients, not senders.
5084 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
5085 the ratelimit ACL was added.
5087 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
5089 PH/46 Added extra information about PostgreSQL errors to the error string.
5091 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
5092 avoiding the need to install special zones in a real server. This is
5093 backwards compatible; if it can't find the fake resolver, it drops back.
5094 Thus, both old and new test suites can be run.
5096 TF/02 Added util/ratelimit.pl
5098 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
5099 clock is set back in time.
5101 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
5102 Candler <B.Candler@pobox.com>.
5104 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
5105 OK for addresses that are the result of redirections.
5107 PH/48 A number of further additions for the benefit of the new test suite,
5108 including a fake gethostbyname() that interfaces to the fake DNS resolver
5111 TF/06 The fix for widen_domains has also been applied to qualify_single and
5112 search_parents which are the other dnslookup options that can cause
5115 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
5118 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
5120 TF/07 Exim produced the error message "an SRV record indicated no SMTP
5121 service" if it encountered an MX record with an empty target hostname.
5122 The message is now "an MX or SRV record indicated no SMTP service".
5124 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
5125 if the DNS of the sending site is misconfigured. This is quite a
5126 common situation. This change restores the behaviour of treating a
5127 helo verification defer as a failure.
5129 PH/51 If self=fail was set on a router, the bounce message did not include the
5130 actual error message.
5136 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
5138 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
5139 command itself ended in error, the underlying message about the transport
5140 filter timeout was being overwritten with the pipe command error. Now the
5141 underlying error message should be appended to the second error message.
5143 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
5145 PH/02 Reduce the amount of output that "make" produces by default. Full output
5146 can still be requested.
5148 PH/03 The warning log line about a condition test deferring for a "warn" verb
5149 was being output only once per connection, rather than after each
5150 occurrence (because it was using the same function as for successful
5151 "warn" verbs). This seems wrong, so I have changed it.
5153 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
5154 it should not have, which might have caused a crash in the right
5155 circumstances, but probably never did.
5157 PH/04 Installed a modified version of Tony Finch's patch to make submission
5158 mode fix the return path as well as the Sender: header line, and to
5159 add a /name= option so that you can make the user's friendly name appear
5162 TF/03 Added the control = fakedefer ACL modifier.
5164 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
5165 Mark Lowes for thorough testing.
5167 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
5169 TK/03 Merged latest SRS patch from Miles Wilton.
5171 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
5172 with the definition in sysexits.h (which is #included earlier).
5173 Fortunately, Exim does not actually use EX_OK. The code used to try to
5174 preserve the sysexits.h value, by assuming that macro definitions were
5175 scanned for macro replacements. I have been disabused of this notion,
5176 so now the code just undefines EX_OK before #including unistd.h.
5178 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
5179 in the smtp transport. When a block could not be written in a single
5180 write() function, the timeout was being re-applied to each part-write.
5181 This seems wrong - if the receiver was accepting one byte at a time it
5182 would take for ever. The timeout is now adjusted when this happens. It
5183 doesn't have to be particularly precise.
5185 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
5186 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
5188 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
5189 without a domain if the check to <postmaster@domain> fails.
5191 SC/01 Eximstats: added -xls and the ability to specify output files
5192 (patch written by Frank Heydlauf).
5194 SC/02 Eximstats: use FileHandles for outputting results.
5196 SC/03 Eximstats: allow any combination of xls, txt, and html output.
5198 SC/04 Eximstats: fixed display of large numbers with -nvr option
5200 SC/05 Eximstats: fixed merging of reports with empty tables.
5202 SC/06 Eximstats: added the -include_original_destination flag
5204 SC/07 Eximstats: removed tabs and trailing whitespace.
5206 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
5208 TK/06 MBOX spool code: Add real "From " MBOX separator line
5209 so the .eml file is really in mbox format (even though
5210 most programs do not really care). Patch from Alex Miller.
5212 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
5213 The latter is generated from $received_to and is only set if the
5214 message has one envelope recipient. SA can use these headers,
5215 obviously out-of-the-box. Patch from Alex Miller.
5217 PH/08 The ${def test on a variable was returning false if the variable's
5218 value was "0", contrary to what the specification has always said!
5219 The result should be true unless the variable is empty.
5221 PH/09 The syntax error of a character other than { following "${if
5222 def:variable_name" (after optional whitespace) was not being diagnosed.
5223 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
5224 accidental colon was present, for example, could give incorrect results.
5226 PH/10 Tidied the code in a number of places where the st_size field of a stat()
5227 result is used (not including appendfile, where other changes are about
5230 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
5231 This involved changing a lot of size variables from int to off_t. It
5232 should work with maildirs and everything.
5234 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
5235 spamd dying while we are connected to it.
5237 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
5240 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
5241 function for BDB 4.3.
5243 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
5245 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
5246 the sources for PCRE have been re-arranged and more files are now
5249 PH/15 The code I had for printing potentially long long variables in PH/11
5250 above was not the best (it lost precision). The length of off_t variables
5251 is now inspected at build time, and an appropriate printing format (%ld
5252 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
5253 to be "long long int" or "long int". This is needed for the internal
5254 formatting function string_vformat().
5256 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
5257 the configuration file to be ":syslog", then the script "guesses" where
5258 the logs files are, rather than using the compiled in default. In our
5259 case the guess is not the same as the compiled default, so the script
5260 suddenly stopped working when I started to use syslog. The patch checks
5261 to see if log_file_path is "". If so, it attempts to read it from exim
5262 with no configuration file to get the compiled in version, before it
5263 falls back to the previous guessing code."
5265 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
5266 implementing BATV in an Exim configuration. See NewStuff for the gory
5269 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
5270 Makefile that are specific to HP-UX.
5272 PH/18 If the "use_postmaster" option was set for a recipient callout together
5273 with the "random" option, the postmaster address was used as the MAIL
5274 FROM address for the random test, but not for the subsequent recipient
5275 test. It is now used for both.
5277 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
5278 patch removes a few documentation additions to RFC 3028, because the
5279 latest draft now contains them. It adds the new en;ascii-case comparator
5280 and a new error check for 8bit text in MIME parts. Comparator and
5281 require names are now matched exactly. I enabled the subaddress
5282 extension, but it is not well tested yet (read: it works for me)."
5284 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
5285 rework some of the code of TK/09 above to avoid the hardwired use of
5286 "%lld" and "long long". Replaced the call to snprintf() with a call to
5289 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
5290 records point to non-existent hosts", "retry timeout exceeded", and
5291 "retry time not reached for any host after a long failure period".
5293 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
5294 experimental DomainKeys support:
5296 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
5297 (2) On an error such as an illegally used "control", the wrong name for
5298 the control was given.
5300 These problems did NOT occur unless DomainKeys support was compiled.
5302 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
5304 PH/24 Added ${if match_ip condition.
5306 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
5307 databases so that it will be absolutely obvious if a crash occurs in the
5308 DB library. This is a regular occurrence (often caused by mis-matched
5311 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
5312 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
5313 picked up on a user's system that detects such things. There doesn't seem
5314 to be a gcc warning option for this - only an attribute that has to be
5315 put on the function's prototype. It seems that in Fedora Core 4 they have
5316 set this on a number of new functions. No doubt there will be more in due
5319 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
5320 specify a transport. However, if an address that was verified by such a
5321 router was the subject of a callout, Exim crashed because it tried to
5322 read the rcpt_include_affixes from the non-existent transport. Now it
5323 just assumes that the setting of that option is false. This bug was
5324 introduced by 4.51/PH/31.
5326 PH/28 Changed -d+all to exclude +memory, because that information is very
5327 rarely of interest, but it makes the output a lot bigger. People tend to
5328 do -d+all out of habit.
5330 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
5331 code in os-type was giving problems when libc.so lives in lib64, like on
5334 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
5335 aren't the modern standard, and it seems that some systems' include files
5336 don't always have them. Exim was already checking for some of the newer
5337 ones like T_AAAA, and defining it itself. I've added checks for all the
5338 record types that Exim uses.
5340 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
5341 not automatically generating a new one, as it is supposed to. This
5342 prevented TLS from working. If the file did exist, but contained invalid
5343 data, a new version was generated, as expected. It was only the case of a
5344 non-existent file that was broken.
5346 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
5347 with a change in libdomainkeys > 0.64.
5349 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
5350 from DNS. If the selector record carries the flag, it now has
5351 precedence over the domain-wide flag.
5353 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
5355 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
5356 the use of a port name that isn't defined in /etc/services) occurred, the
5357 message was deferred as in a normal delivery, and thus remained on the
5358 spool, instead of being failed because of the mua_wrapper setting. This
5359 is now fixed, and I tidied up some of the mua_wrapper messages at the
5362 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
5363 the messages in a hash of arrays rather than using individual hashes.
5364 This is a bit cleaner and results in dramatic memory savings, albeit
5365 at a slight CPU cost.
5367 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
5368 as requested by Marc Sherman.
5370 SC/10 Eximstats: added histograms for user specified patterns as requested
5373 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
5375 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
5376 fopen() in the content-scanning modules that did not already have it.
5382 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
5383 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
5385 TK/02 Fix ACL "control" statement not being available in MIME ACL.
5387 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
5389 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
5390 to test Sieve filters that use "vacation".
5392 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
5393 that changes the way the GnuTLS parameters are stored in the cache file.
5394 The new format can be generated externally. For backward compatibility,
5395 if the data in the cache doesn't make sense, Exim assumes it has read an
5396 old-format file, and it generates new data and writes a new file. This
5397 means that you can't go back to an older release without removing the
5400 PH/03 A redirect router that has both "unseen" and "one_time" set does not
5401 work if there are any delivery delays because "one_time" forces the
5402 parent to be marked "delivered", so its unseen clone is never tried
5403 again. For this reason, Exim now forbids the simultaneous setting of
5406 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
5407 redirected to themselves ("homonym" addresses). Read the long ChangeLog
5408 entry if you want to know the details. The fix, however, neglected to
5409 consider the case when local delivery batching is involved. The test for
5410 "previously delivered" was not happening when checking to see if an
5411 address could be batched with a previous (undelivered) one; under
5412 certain circumstances this could lead to multiple deliveries to the same
5415 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
5416 in its include files, and this causes problems building Exim.
5418 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
5419 header_syntax) but Exim was just ignoring anything given after a slash.
5420 In particular, this caused confusion with an attempt to use "verify =
5421 reverse_host_lookup/defer_ok". An error is now given when options are
5422 supplied for verify items that do not have them. (Maybe reverse_host_
5423 lookup should have a defer_ok option, but that's a different point.)
5425 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
5426 defined by RFC 821) to 2048, because there were problems with some AUTH
5427 commands, and RFC 1869 says the size should be increased for extended
5428 SMTP commands that take arguments.
5430 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
5433 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
5434 "unknown" error; now it says that the functionality isn't in the binary.
5436 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
5437 an address' error message when a string expansion fails (syntax or
5438 whatever). Otherwise the password may appear in the log. Following change
5439 PH/42 below, there is no longer a chance of it appearing in a bounce
5442 PH/11 Installed exipick version 20050225.0 from John Jetmore.
5444 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
5445 of its addresses was ever tried. (Bugzilla bug #2.)
5447 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
5448 the result incorrectly in the debug output. (It correctly added a newline
5449 to what was transported.)
5451 TF/01 Added $received_time.
5453 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
5454 commented out examples of how to interface to a virus scanner and to
5455 SpamAssassin. Also added commented examples of av_scanner and
5456 spamd_address settings.
5458 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
5459 and controls are allowed in which ACLs. There were a couple of minor
5460 errors. Some of the entries in the conditions table (which is a table of
5461 where they are NOT allowed) were getting very unwieldy; rewrote them as a
5462 negation of where the condition IS allowed.
5464 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
5466 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
5467 header file does not have a version number, so I've had to invent a new
5468 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
5469 API. The code is untested by me (my Linux distribution still has 0.3.2 of
5470 radiusclient), but it was contributed by a Radius user.
5472 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
5473 files or directories, for OpenSSL.
5475 PH/19 When an Exim process that is running as root has to create an Exim log
5476 file, it does so in a subprocess that runs as exim:exim so as to get the
5477 ownership right at creation (otherwise, other Exim processes might see
5478 the file with the wrong ownership). There was no test for failure of this
5479 fork() call, which would lead to the process getting stuck as it waited
5480 for a non-existent subprocess. Forks do occasionally fail when resources
5481 run out. I reviewed all the other calls to fork(); they all seem to check
5484 PH/20 When checking for unexpected SMTP input at connect time (before writing
5485 the banner), Exim was not dealing correctly with a non-positive return
5486 from the read() function. If the client had disconnected by this time,
5487 the result was a log entry for a synchronization error with an empty
5488 string after "input=" when read() returned zero. If read() returned -1
5489 (an event I could not check), uninitialized data bytes were printed.
5490 There were reports of junk text (parts of files, etc) appearing after
5493 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
5495 PH/22 Added support for macro redefinition, and (re)definition in between
5496 driver and ACL definitions.
5498 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
5499 forgetting to use the resulting value; it was using the unexpanded value.
5501 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
5502 hadn't been configured. The fix is from Juergen Kreileder, who
5503 understands it better than I do:
5505 "Here's what I see happening with three configured cyrus_sasl
5506 authenticators configured (plain, login, cram-md5):
5508 On startup auth_cyrus_sasl_init() gets called for each of these.
5509 This means three calls to sasl_listmech() without a specified mech_list.
5510 => SASL tests which mechs of all available mechs actually work
5511 => three warnings about OTP not working
5512 => the returned list contains: plain, login, cram-md5, digest-md5, ...
5514 With the patch, sasl_listmech() also gets called three times. But now
5515 SASL's mech_list option is set to the server_mech specified in the the
5516 authenticator. Or in other words, the answer from sasl_listmech()
5517 gets limited to just the mech you're testing for (which is different
5519 => the return list contains just 'plain' or 'login', 'cram-md5' or
5520 nothing depending on the value of ob->server_mech.
5522 I've just tested the patch: Authentication still works fine,
5523 unavailable mechs specified in the exim configuration are still
5524 caught, and the auth.log warnings about OTP are gone."
5526 PH/25 When debugging is enabled, the contents of the command line are added
5527 to the debugging output, even when log_selector=+arguments is not
5530 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
5531 answer is "GNU", and only if the return is "GNU/something" is the answer
5534 PH/27 $acl_verify_message is now set immediately after the failure of a
5535 verification in an ACL, and so is available in subsequent modifiers. In
5536 particular, the message can be preserved by coding like this:
5538 warn !verify = sender
5539 set acl_m0 = $acl_verify_message
5541 Previously, $acl_verify_message was set only while expanding "message"
5542 and "log_message" when a very denied access.
5544 PH/28 Modified OS/os.c-Linux with
5546 -#ifndef OS_LOAD_AVERAGE
5547 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
5549 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
5550 nomenclature these days.)
5552 PH/29 Installed patch from the Sieve maintainer that adds the options
5553 sieve_useraddress and sieve_subaddress to the redirect router.
5555 PH/30 In these circumstances:
5556 . Two addresses routed to the same list of hosts;
5557 . First host does not offer TLS;
5558 . First host accepts first address;
5559 . First host gives temporary error to second address;
5560 . Second host offers TLS and a TLS session is established;
5561 . Second host accepts second address.
5562 Exim incorrectly logged both deliveries with the TLS parameters (cipher
5563 and peerdn, if requested) that were in fact used only for the second
5566 PH/31 When doing a callout as part of verifying an address, Exim was not paying
5567 attention to any local part prefix or suffix that was matched by the
5568 router that accepted the address. It now behaves in the same way as it
5569 does for delivery: the affixes are removed from the local part unless
5570 rcpt_include_affixes is set on the transport.
5572 PH/32 Add the sender address, as F=<...>, to the log line when logging a
5573 timeout during the DATA phase of an incoming message.
5575 PH/33 Sieve envelope tests were broken for match types other than :is. I have
5576 applied a patch sanctioned by the Sieve maintainer.
5578 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
5579 the uid or gid is negative. A case of a negative gid caused this to be
5580 noticed. The fix allows for either to be negative.
5582 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
5583 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
5586 PH/36 The change PH/12 above was broken. Fixed it.
5588 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
5589 the grounds that routing the same address twice would always produce the
5590 same answer. This might have been true once, but it is certainly no
5591 longer true now. Routing a child address may depend on the previous
5592 routing that produced that child. Some complicated redirection strategies
5593 went wrong when messages had multiple recipients, and made Exim's
5594 behaviour dependent on the order in which the addresses were given.
5596 I have moved the duplicate checking until after the routing is complete.
5597 Exim scans the addresses that are assigned to local and remote
5598 transports, and removes any duplicates. This means that more work will be
5599 done, as duplicates will always all be routed, but duplicates are
5600 presumably rare, so I don't expect this is of any significance.
5602 For deliveries to pipes, files, and autoreplies, the duplicate checking
5603 still happens during the routing process, since they are not going to be
5606 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
5607 It corrects a timeout issue with spamd. This is Ian's comment: "The
5608 background is that sometimes spamd either never reads data from a
5609 connection it has accepted, or it never writes response data. The exiscan
5610 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
5611 blindly assumes that writes won't block so it may never time out."
5613 PH/39 Allow G after quota size as well as K and M.
5615 PH/40 The value set for $authenticated_id in an authenticator may not contain
5616 binary zeroes or newlines because the value is written to log lines and
5617 to spool files. There was no check on this. Now the value is run through
5618 the string_printing() function so that such characters are converted to
5619 printable escape sequences.
5621 PH/41 $message_linecount is a new variable that contains the total number of
5622 lines in the message. Compare $body_linecount, which is the count for the
5625 PH/42 Exim no longer gives details of delivery errors for specific addresses in
5626 bounce and delay warning messages, except in certain special cases, which
5629 (a) An SMTP error message from a remote host;
5630 (b) A message specified in a :fail: redirection;
5631 (c) A message specified in a "fail" command in a system filter;
5632 (d) A message specified in a FAIL return from the queryprogram router;
5633 (e) A message specified by the cannot_route_message router option.
5635 In these cases only, Exim does include the error details in bounce and
5636 warning messages. There are also a few cases where bland messages such
5637 as "unrouteable address" or "local delivery error" are given.
5639 PH/43 $value is now also set for the "else" part of a ${run expansion.
5641 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
5642 being worked on, but at least Exim now implements the latest version to
5645 PH/45 In a pipe transport, although a timeout while waiting for the pipe
5646 process to complete was treated as a delivery failure, a timeout while
5647 writing the message to the pipe was logged, but erroneously treated as a
5648 successful delivery. Such timeouts include transport filter timeouts. For
5649 consistency with the overall process timeout, these timeouts are now
5650 treated as errors, giving rise to delivery failures by default. However,
5651 there is now a new Boolean option for the pipe transport called
5652 timeout_defer, which, if set TRUE, converts the failures into defers for
5653 both kinds of timeout. A transport filter timeout is now identified in
5656 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
5657 systems where "make" and "gmake" are different, calling "gmake" at top
5658 level broke things. I've arranged for the value of $(MAKE) to be passed
5659 from the Makefile to this script so that it can call the same version of
5663 A note about Exim versions 4.44 and 4.50
5664 ----------------------------------------
5666 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
5667 changes of various kinds. As a consequence, a big documentation update was
5668 needed. This delayed the release for rather longer than seemed good, especially
5669 in the light of a couple of (minor) security issues. Therefore, the changes
5670 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
5671 release. So 4.44 and 4.50 are in effect two different branches that both start
5674 I have left the 4.50 change log unchanged; it contains all the changes since
5675 4.43. The change log for 4.44 is below; many of its items are identical to
5676 those for 4.50. This seems to be the most sensible way to preserve the
5677 historical information.
5683 1. Minor wording change to the doc/README.SIEVE file.
5685 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
5686 computation of the current number of files was incorrect.
5688 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
5689 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
5692 4. Give more explanation in the error message when the command for a transport
5693 filter fails to execute.
5695 5. There are several places where Exim runs a non-Exim command in a
5696 subprocess. The SIGUSR1 signal should be disabled for these processes. This
5697 was being done only for the command run by the queryprogram router. It is
5698 now done for all such subprocesses. The other cases are: ${run, transport
5699 filters, and the commands run by the lmtp and pipe transports.
5701 6. Added CONFIGURE_GROUP build-time option.
5703 7. Some older OS have a limit of 256 on the maximum number of file
5704 descriptors. Exim was using setrlimit() to set 1000 as a large value
5705 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
5706 systems. I've change it so that if it can't get 1000, it tries for 256.
5708 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
5709 was an oversight, and furthermore, ever since the addition of extra
5710 controls (e.g. 4.43/32), the checks on when to allow different forms of
5711 "control" were broken. There should now be diagnostics for all cases when a
5712 control that does not make sense is encountered.
5714 9. Added the /retain_sender option to "control=submission".
5716 10. $recipients is now available in the predata ACL (oversight).
5718 11. Tidy the search cache before the fork to do a delivery from a message
5719 received from the command line. Otherwise the child will trigger a lookup
5720 failure and thereby defer the delivery if it tries to use (for example) a
5721 cached ldap connection that the parent has called unbind on.
5723 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
5724 of $address_data from the recipient verification was clobbered by the
5725 sender verification.
5727 13. The value of address_data from a sender verification is now available in
5728 $sender_address_data in subsequent conditions in the ACL statement.
5730 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
5732 15. Added a new option "connect=<time>" to callout options, to set a different
5735 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
5736 was its contents. (It was OK if the option was not defined at all.)
5738 17. A "Completed" log line is now written for messages that are removed from
5739 the spool by the -Mrm option.
5741 18. New variables $sender_verify_failure and $recipient_verify_failure contain
5742 information about exactly what failed.
5744 19. Added -dd to debug only the daemon process.
5746 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
5747 handles timeouts, both on the server side and network timeouts. Renamed the
5748 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
5750 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
5751 It is now set to "smtps".
5753 22. $host_address is now set to the target address during the checking of
5754 ignore_target_hosts.
5756 23. When checking ignore_target_hosts for an ipliteral router, no host name was
5757 being passed; this would have caused $sender_host_name to have been used if
5758 matching the list had actually called for a host name (not very likely,
5759 since this list is usually IP addresses). A host name is now passed as
5762 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
5763 code that specifies a non-restarting handler (typically sigaction() in
5764 modern systems) in an attempt to fix a rare and obscure crash bug.
5766 25. Narrowed the window for a race in the daemon that could cause it to ignore
5767 SIGCHLD signals. This is not a major problem, because they are used only to
5768 wake it up if nothing else does.
5770 26. A malformed maildirsize file could cause Exim to calculate negative values
5771 for the mailbox size or file count. Odd effects could occur as a result.
5772 The maildirsize information is now recalculated if the size or filecount
5775 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
5776 support for a long time. Removed HAVE_SYS_VFS_H.
5778 28. Installed the latest version of exipick from John Jetmore.
5780 29. In an address list, if the pattern was not a regular expression, an empty
5781 subject address (from a bounce message) matched only if the pattern was an
5782 empty string. Non-empty patterns were not even tested. This was the wrong
5783 because it is perfectly reasonable to use an empty address as part of a
5784 database query. An empty address is now tested by patterns that are
5785 lookups. However, all the other forms of pattern expect the subject to
5786 contain a local part and a domain, and therefore, for them, an empty
5787 address still always fails if the pattern is not itself empty.
5789 30. Exim went into a mad DNS loop when attempting to do a callout where the
5790 host was specified on an smtp transport, and looking it up yielded more
5791 than one IP address.
5793 31. Re-factored the code for checking spool and log partition space into a
5794 function that finds that data and another that does the check. The former
5795 is then used to implement four new variables: $spool_space, $log_space,
5796 $spool_inodes, and $log_inodes.
5798 32. The RFC2047 encoding function was originally intended for short strings
5799 such as real names; it was not keeping to the 75-character limit for
5800 encoded words that the RFC imposes. It now respects the limit, and
5801 generates multiple encoded words if necessary. To be on the safe side, I
5802 have increased the buffer size for the ${rfc2047: expansion operator from
5805 33. It is now permitted to omit both strings after an "if" condition; if the
5806 condition is true, the result is "true". As before, when the second string
5807 is omitted, a false condition yields an empty string. This makes it less
5808 cumbersome to write custom ACL and router conditions.
5810 34. Failure to deliver a bounce message always caused it to be frozen, even if
5811 there was an errors_to setting on the router. The errors_to setting is now
5814 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
5815 canonical form (fully expanded) before being placed in
5816 $sender_host_address.
5818 36. The table in the code that translates DNS record types into text (T_A to
5819 "A" for instance) was missing entries for NS and CNAME. It is just possible
5820 that this could have caused confusion if both these types were looked up
5821 for the same domain, because the text type is used as part of Exim's
5822 per-process caching. But the chance of anyone hitting this buglet seems
5825 37. The dnsdb lookup has been extended in a number of ways.
5827 (1) There is a new type, "zns", which walks up the domain tree until it
5828 finds some nameserver records. It should be used with care.
5830 (2) There is a new type, "mxh", which is like "mx" except that it returns
5831 just the host names, not the priorities.
5833 (3) It is now possible to give a list of domains (or IP addresses) to be
5834 looked up. The behaviour when one of the lookups defers can be
5835 controlled by a keyword.
5837 (4) It is now possible to specify the separator character for use when
5838 multiple records are returned.
5840 38. The dnslists ACL condition has been extended: it is now possible to supply
5841 a list of IP addresses and/or domains to be looked up in a particular DNS
5844 39. Added log_selector=+queue_time_overall.
5846 40. When running the queue in the test harness, wait just a tad after forking a
5847 delivery process, to get repeatability of debugging output.
5849 41. Include certificate and key file names in error message when GnuTLS fails
5850 to set them up, because the GnuTLS error message doesn't include the name
5851 of the failing file when there is a problem reading it.
5853 42. Allow both -bf and -bF in the same test run.
5855 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
5857 44. The "Exiscan patch" is now merged into the mainline Exim source.
5859 45. Sometimes the final signoff response after QUIT could fail to get
5860 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
5861 < 0 before doing a fflush(). This bug looks as though it goes back to the
5862 introduction of TLS in release 3.20, but "sometimes" must have been rare
5863 because the tests only now provoked it.
5865 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
5866 (this can affect the format of dates).
5868 47. exim_tidydb, when checking for the continued existence of a message for
5869 which it has found a message-specific retry record, was not finding
5870 messages that were in split spool directories. Consequently, it was
5871 deleting retry records that should have stayed in existence.
5873 48. Steve fixed some bugs in eximstats.
5875 49. The SPA authentication driver was not abandoning authentication and moving
5876 on to the next authenticator when an expansion was forced to fail,
5877 contradicting the general specification for all authenticators. Instead it
5878 was generating a temporary error. It now behaves as specified.
5880 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
5881 (the order specifies the preference for clients). The order is now AES256,
5882 AES128, 3DES, ARCFOUR128.
5884 51. Small patch to Sieve code - explicitly set From: when generating an
5887 52. Exim crashed if a remote delivery caused a very long error message to be
5888 recorded - for instance if somebody sent an entire SpamAssassin report back
5889 as a large number of 550 error lines. This bug was coincidentally fixed by
5890 increasing the size of one of Exim's internal buffers (big_buffer) that
5891 happened as part of the Exiscan merge. However, to be on the safe side, I
5892 have made the code more robust (and fixed the comments that describe what
5895 53. Now that there can be additional text after "Completed" in log lines (if
5896 the queue_time_overall log selector is set), a one-byte patch to exigrep
5897 was needed to allow it to recognize "Completed" as not the last thing in
5900 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
5901 patch that reportedly fixes this has been added. I am not expert enough to
5902 create a test for it. This is what the patch creator wrote:
5904 "I found a little strange behaviour of ldap code when working with
5905 Windows 2003 AD Domain, where users was placed in more than one
5906 Organization Units. When I tried to give exim partial DN, the exit code
5907 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
5908 But simultaneously result of request was absolutely normal ldap result,
5909 so I produce this patch..."
5911 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
5912 REFERENCE, so I have modified the code to exclude the patch when that macro
5915 55. Some experimental protocols are using DNS PTR records for new purposes. The
5916 keys for these records are domain names, not reversed IP addresses. The
5917 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
5918 leaves it alone. Component reversal etc. now happens only for IP addresses.
5921 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
5923 57. Double the size of the debug message buffer (to 2048) so that more of very
5924 long debug lines gets shown.
5926 58. The exicyclog utility now does better if the number of log files to keep
5927 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
5929 59. Two changes related to the smtp_active_hostname option:
5931 (1) $smtp_active_hostname is now available as a variable.
5932 (2) The default for smtp_banner uses $smtp_active_hostname instead
5933 of $primary_hostname.
5935 60. The host_aton() function is supposed to be passed a string that is known
5936 to be a valid IP address. However, in the case of IPv6 addresses, it was
5937 not checking this. This is a hostage to fortune. Exim now panics and dies
5938 if the condition is not met. A case was found where this could be provoked
5939 from a dnsdb PTR lookup with an IPv6 address that had more than 8
5940 components; fortuitously, this particular loophole had already been fixed
5941 by change 4.50/55 above.
5943 If there are any other similar loopholes, the new check in host_aton()
5944 itself should stop them being exploited. The report I received stated that
5945 data on the command line could provoke the exploit when Exim was running as
5946 exim, but did not say which command line option was involved. All I could
5947 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
5948 running as the user.
5951 61. There was a buffer overflow vulnerability in the SPA authentication code
5952 (which came originally from the Samba project). I have added a test to the
5953 spa_base64_to_bits() function which I hope fixes it.
5956 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
5957 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
5959 63. The daemon start-up calls getloadavg() while still root for those OS that
5960 need the first call to be done as root, but it missed one case: when
5961 deliver_queue_load_max is set with deliver_drop_privilege. This is
5962 necessary for the benefit of the queue runner, because there is no re-exec
5963 when deliver_drop_privilege is set.
5965 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
5966 This has been fixed.
5968 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
5969 was in use, was not putting the data itself into the right store pool;
5970 consequently, it could be overwritten for a subsequent message in the same
5971 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
5974 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
5976 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
5977 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
5978 the function were treating the return as a boolean value, which happened to
5979 work because 0=false and not-0=true, but is not correct code.
5981 68. The host_aton() function was not handling scoped IPv6 addresses (those
5982 with, for example, "%eth0" on the end) correctly.
5984 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
5985 negated items (that is, ~something) in unsigned ints. Some compilers
5986 apparently mutter when there is no cast.
5988 70. If an address verification called from an ACL failed, and did not produce a
5989 user-specific message (i.e. there was only a "system" message), nothing was
5990 put in $acl_verify_message. In this situation, it now puts the system
5993 71. Change 4.23/11 added synchronization checking at the start of an SMTP
5994 session; change 4.31/43 added the unwanted input to the log line - except
5995 that it did not do this in the start of session case. It now does.
5997 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
5998 This is wrong and can cause the other end to generate a synchronization
5999 error if it is another Exim or anything else that does the synchronization
6000 check. A QUIT command is no longer sent after a timeout.
6002 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
6003 during host lookups.
6005 74. The defer_ok option of callout verification was not working if it was used
6006 when verifying addresses in header lines, that is, for this case:
6008 verify = header_sender/callout=defer_ok
6010 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
6011 those file descriptors could be used for SMTP connections. If anything
6012 wrote to stderr (the example that came up was "warn" in embedded Perl), it
6013 could be sent to the SMTP client, causing chaos. The daemon now opens
6014 stdin, stdout, and stderr to /dev/null when it puts itself into the
6017 76. Arrange for output from Perl's "warn" command to be written to Exim's main
6018 log by default. The user can override this with suitable Perl magic.
6020 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
6021 the log message when discard triggers, was not working for the DATA ACL or
6022 for the non-SMTP ACL.
6024 78. Error message wording change in sieve.c.
6026 79. If smtp_accept_max_per_host was set, the number of connections could be
6027 restricted to fewer than expected, because the daemon was trying to set up
6028 a new connection before checking whether the processes handling previous
6029 connections had finished. The check for completed processes is now done
6030 earlier. On busy systems, this bug wouldn't be noticed because something
6031 else would have woken the daemon, and it would have reaped the completed
6034 80. If a message was submitted locally by a user whose login name contained one
6035 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
6036 It caused a spool format error. I have fixed the spool reading code. A
6037 related problem was that the "from" clause in the Received: line became
6038 illegal because of the space(s). It is now covered by ${quote_local_part.
6040 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
6043 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
6045 83. Patch from Sieve maintainer to fix unterminated string problem in
6046 "vacation" handling.
6048 84. Some minor changes to the Linux configuration files to help with other
6049 OS variants using glibc.
6051 85. One more patch for Sieve to update vacation handling to latest spec.
6054 ----------------------------------------------------
6055 See the note above about the 4.44 and 4.50 releases.
6056 ----------------------------------------------------
6062 1. Change 4.43/35 introduced a bug that caused file counts to be
6063 incorrectly computed when quota_filecount was set in an appendfile
6066 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
6067 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
6070 3. Give more explanation in the error message when the command for a transport
6071 filter fails to execute.
6073 4. There are several places where Exim runs a non-Exim command in a
6074 subprocess. The SIGUSR1 signal should be disabled for these processes. This
6075 was being done only for the command run by the queryprogram router. It is
6076 now done for all such subprocesses. The other cases are: ${run, transport
6077 filters, and the commands run by the lmtp and pipe transports.
6079 5. Some older OS have a limit of 256 on the maximum number of file
6080 descriptors. Exim was using setrlimit() to set 1000 as a large value
6081 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
6082 systems. I've change it so that if it can't get 1000, it tries for 256.
6084 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
6085 was an oversight, and furthermore, ever since the addition of extra
6086 controls (e.g. 4.43/32), the checks on when to allow different forms of
6087 "control" were broken. There should now be diagnostics for all cases when a
6088 control that does not make sense is encountered.
6090 7. $recipients is now available in the predata ACL (oversight).
6092 8. Tidy the search cache before the fork to do a delivery from a message
6093 received from the command line. Otherwise the child will trigger a lookup
6094 failure and thereby defer the delivery if it tries to use (for example) a
6095 cached ldap connection that the parent has called unbind on.
6097 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
6098 of $address_data from the recipient verification was clobbered by the
6099 sender verification.
6101 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
6102 was its contents. (It was OK if the option was not defined at all.)
6104 11. A "Completed" log line is now written for messages that are removed from
6105 the spool by the -Mrm option.
6107 12. $host_address is now set to the target address during the checking of
6108 ignore_target_hosts.
6110 13. When checking ignore_target_hosts for an ipliteral router, no host name was
6111 being passed; this would have caused $sender_host_name to have been used if
6112 matching the list had actually called for a host name (not very likely,
6113 since this list is usually IP addresses). A host name is now passed as
6116 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
6117 code that specifies a non-restarting handler (typically sigaction() in
6118 modern systems) in an attempt to fix a rare and obscure crash bug.
6120 15. Narrowed the window for a race in the daemon that could cause it to ignore
6121 SIGCHLD signals. This is not a major problem, because they are used only to
6122 wake it up if nothing else does.
6124 16. A malformed maildirsize file could cause Exim to calculate negative values
6125 for the mailbox size or file count. Odd effects could occur as a result.
6126 The maildirsize information is now recalculated if the size or filecount
6129 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
6130 support for a long time. Removed HAVE_SYS_VFS_H.
6132 18. Updated exipick to current release from John Jetmore.
6134 19. Allow an empty sender to be matched against a lookup in an address list.
6135 Previously the only cases considered were a regular expression, or an
6138 20. Exim went into a mad DNS lookup loop when doing a callout where the
6139 host was specified on the transport, if the DNS lookup yielded more than
6142 21. The RFC2047 encoding function was originally intended for short strings
6143 such as real names; it was not keeping to the 75-character limit for
6144 encoded words that the RFC imposes. It now respects the limit, and
6145 generates multiple encoded words if necessary. To be on the safe side, I
6146 have increased the buffer size for the ${rfc2047: expansion operator from
6149 22. Failure to deliver a bounce message always caused it to be frozen, even if
6150 there was an errors_to setting on the router. The errors_to setting is now
6153 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
6154 canonical form (fully expanded) before being placed in
6155 $sender_host_address.
6157 24. Updated eximstats to version 1.33
6159 25. Include certificate and key file names in error message when GnuTLS fails
6160 to set them up, because the GnuTLS error message doesn't include the name
6161 of the failing file when there is a problem reading it.
6163 26. Expand error message when OpenSSL has problems setting up cert/key files.
6166 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
6167 (this can affect the format of dates).
6169 28. exim_tidydb, when checking for the continued existence of a message for
6170 which it has found a message-specific retry record, was not finding
6171 messages that were in split spool directories. Consequently, it was
6172 deleting retry records that should have stayed in existence.
6174 29. eximstats updated to version 1.35
6175 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
6176 1.35 - bugfix such that pie charts by volume are generated correctly
6178 30. The SPA authentication driver was not abandoning authentication and moving
6179 on to the next authenticator when an expansion was forced to fail,
6180 contradicting the general specification for all authenticators. Instead it
6181 was generating a temporary error. It now behaves as specified.
6183 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
6184 (the order specifies the preference for clients). The order is now AES256,
6185 AES128, 3DES, ARCFOUR128.
6187 31. Small patch to Sieve code - explicitly set From: when generating an
6190 32. Exim crashed if a remote delivery caused a very long error message to be
6191 recorded - for instance if somebody sent an entire SpamAssassin report back
6192 as a large number of 550 error lines. This bug was coincidentally fixed by
6193 increasing the size of one of Exim's internal buffers (big_buffer) that
6194 happened as part of the Exiscan merge. However, to be on the safe side, I
6195 have made the code more robust (and fixed the comments that describe what
6198 33. Some experimental protocols are using DNS PTR records for new purposes. The
6199 keys for these records are domain names, not reversed IP addresses. The
6200 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
6201 leaves it alone. Component reversal etc. now happens only for IP addresses.
6204 34. The host_aton() function is supposed to be passed a string that is known
6205 to be a valid IP address. However, in the case of IPv6 addresses, it was
6206 not checking this. This is a hostage to fortune. Exim now panics and dies
6207 if the condition is not met. A case was found where this could be provoked
6208 from a dnsdb PTR lookup with an IPv6 address that had more than 8
6209 components; fortuitously, this particular loophole had already been fixed
6210 by change 4.50/55 or 4.44/33 above.
6212 If there are any other similar loopholes, the new check in host_aton()
6213 itself should stop them being exploited. The report I received stated that
6214 data on the command line could provoke the exploit when Exim was running as
6215 exim, but did not say which command line option was involved. All I could
6216 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
6217 running as the user.
6220 35. There was a buffer overflow vulnerability in the SPA authentication code
6221 (which came originally from the Samba project). I have added a test to the
6222 spa_base64_to_bits() function which I hope fixes it.
6225 36. The daemon start-up calls getloadavg() while still root for those OS that
6226 need the first call to be done as root, but it missed one case: when
6227 deliver_queue_load_max is set with deliver_drop_privilege. This is
6228 necessary for the benefit of the queue runner, because there is no re-exec
6229 when deliver_drop_privilege is set.
6231 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
6232 was in use, was not putting the data itself into the right store pool;
6233 consequently, it could be overwritten for a subsequent message in the same
6234 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
6237 38. Sometimes the final signoff response after QUIT could fail to get
6238 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
6239 < 0 before doing a fflush(). This bug looks as though it goes back to the
6240 introduction of TLS in release 3.20, but "sometimes" must have been rare
6241 because the tests only now provoked it.
6247 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
6248 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
6249 Now it returns an integer. A number of calls were still expecting a T/F
6250 return. Fortuitously, in all cases, the tests worked in OK situations,
6251 which is the norm. However, things would have gone wrong on any write
6252 failures on the smtp file descriptor. This function is used when sending
6253 messages over SMTP and also when doing verify callouts.
6255 2. When Exim is called to do synchronous delivery of a locally submitted
6256 message (the -odf or -odi options), it no longer closes stderr before doing
6259 3. Implemented the mua_wrapper option.
6261 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
6263 5. Implemented the functions header_remove(), header_testname(),
6264 header_add_at_position(), and receive_remove_recipient(), and exported them
6267 6. If an ACL "warn" statement specified the addition of headers, Exim already
6268 inserted X-ACL-Warn: at the start if there was no header name. However, it
6269 was not making this test for the second and subsequent header lines if
6270 there were newlines in the string. This meant that an invalid header could
6271 be inserted if Exim was badly configured.
6273 7. Allow an ACL "warn" statement to add header lines at the start or after all
6274 the Received: headers, as well as at the end.
6276 8. Added the rcpt_4xx retry error code.
6278 9. Added postmaster_mailfrom=xxx to callout verification option.
6280 10. Added mailfrom=xxxx to the callout verification option, for verify=
6283 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
6284 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
6286 12. Inserted some casts to stop certain compilers warning when using pointer
6287 differences as field lengths or precisions in printf-type calls (mostly
6288 affecting debugging statements).
6290 13. Added optional readline() support for -be (dynamically loaded).
6292 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
6293 same clock tick as a message's arrival, so that its received time was the
6294 same as the "first fail" time on the retry record, and that message
6295 remained on the queue past the ultimate address timeout, every queue runner
6296 would try a delivery (because it was past the ultimate address timeout) but
6297 after another failure, the ultimate address timeout, which should have then
6298 bounced the address, did not kick in. This was a "< instead of <=" error;
6299 in most cases the first failure would have been in the next clock tick
6300 after the received time, and all would be well.
6302 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
6303 being recognized when the domain list was tested by the match_domain
6304 condition in an expansion string.
6306 16. Added the ${str2b64: operator.
6308 17. Exim was always calling setrlimit() to set a large limit for the number of
6309 processes, without checking whether the existing limit was already
6310 adequate. (It did check for the limit on file descriptors.) Furthermore,
6311 errors from getrlimit() and setrlimit() were being ignored. Now they are
6312 logged to the main and panic logs, but Exim does carry on, to try to do its
6313 job under whatever limits there are.
6315 18. Imported PCRE 5.0.
6317 19. Trivial typo in log message " temporarily refused connection" (the leading
6320 20. If the log selector return_path_on_delivery was set and an address was
6321 redirected to /dev/null, the delivery process crashed because it assumed
6322 that a return path would always be set for a "successful" delivery. In this
6323 case, the whole delivery is bypassed as an optimization, and therefore no
6326 21. Internal re-arrangement: the function for sending a challenge and reading
6327 a response while authentication was assuming a zero-terminated challenge
6328 string. It's now changed to take a pointer and a length, to allow for
6329 binary data in such strings.
6331 22. Added the cyrus_sasl authenticator (code supplied by MBM).
6333 23. Exim was not respecting finduser_retries when seeking the login of the
6334 uid under which it was called; it was always trying 10 times. (The default
6335 setting of finduser_retries is zero.) Also, it was sleeping after the final
6336 failure, which is pointless.
6338 24. Implemented tls_on_connect_ports.
6340 25. Implemented acl_smtp_predata.
6342 26. If the domain in control=submission is set empty, Exim assumes that the
6343 authenticated id is a complete email address when it generates From: or
6344 Sender: header lines.
6346 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
6347 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
6348 chown and chgrp in /bin and hostname in /usr/bin.
6350 28. Exim was keeping the "process log" file open after each use, just as it
6351 does for the main log. This opens the possibility of it remaining open for
6352 long periods when the USR1 signal hits a daemon. Occasional processlog
6353 errors were reported, that could have been caused by this. Anyway, it seems
6354 much more sensible not to leave this file open at all, so that is what now
6357 29. The long-running daemon process does not normally write to the log once it
6358 has entered its main loop, and it closes the log before doing so. This is
6359 so that log files can straightforwardly be renamed and moved. However,
6360 there are a couple of unusual error situations where the daemon does write
6361 log entries, and I had neglected to close the log afterwards.
6363 30. The text of an SMTP error response that was received during a remote
6364 delivery was being truncated at 512 bytes. This is too short for some of
6365 the long messages that one sometimes sees. I've increased the limit to
6368 31. It is now possible to make retry rules that apply only when a message has a
6369 specific sender, in particular, an empty sender.
6371 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
6372 it possible to be selective about when SMTP synchronization is enforced.
6374 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
6376 32. Implemented hosts_connection_nolog.
6378 33. Added an ACL for QUIT.
6380 34. Setting "delay_warning=" to disable warnings was not working; it gave a
6383 35. Added mailbox_size and mailbox_filecount to appendfile.
6385 36. Added control = no_multiline_responses to ACLs.
6387 37. There was a bug in the logic of the code that waits for the clock to tick
6388 in the case where the clock went backwards by a substantial amount such
6389 that the microsecond fraction of "now" was more than the microsecond
6390 fraction of "then" (but the whole seconds number was less).
6392 38. Added support for the libradius Radius client library this is found on
6393 FreeBSD (previously only the radiusclient library was supported).
6399 1. When certain lookups returned multiple values in the form name=value, the
6400 quoting of the values was not always being done properly. Specifically:
6401 (a) If the value started with a double quote, but contained no whitespace,
6403 (b) If the value contained whitespace other than a space character (i.e.
6404 tabs or newlines or carriage returns) it was not quoted.
6405 This fix has been applied to the mysql and pgsql lookups by writing a
6406 separate quoting function and calling it from the lookup code. The fix
6407 should probably also be applied to nisplus, ibase and oracle lookups, but
6408 since I cannot test any of those, I have not disturbed their existing code.
6410 2. A hit in the callout cache for a specific address caused a log line with no
6411 reason for rejecting RCPT. Now it says "Previous (cached) callout
6412 verification failure".
6414 3. There was an off-by-one bug in the queryprogram router. An over-long
6415 return line was truncated at 256 instead of 255 characters, thereby
6416 overflowing its buffer with the terminating zero. As well as fixing this, I
6417 have increased the buffer size to 1024 (and made a note to document this).
6419 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
6420 when Exim is waiting for an SMTP response from a remote server, Exim
6421 restarts its select() call on the socket, thereby resetting its timeout.
6422 This is not a problem when such interrupts are rare. Somebody set up a cron
6423 job to run exiwhat every 2 minutes, which is less than the normal select()
6424 timeout (5 or 10 minutes). This meant that the select() timeout never
6425 kicked in because it was always reset. I have fixed this by comparing the
6426 time when an interrupt arrives with the time at the start of the first call
6427 to select(). If more time than the timeout has elapsed, the interrupt is
6428 treated as a timeout.
6430 5. Some internal re-factoring in preparation for the addition of Sieve
6431 extensions (by MH). In particular, the "personal" test is moved to a
6432 separate function, and given an option for scanning Cc: and Bcc: (which is
6433 not set for Exim filters).
6435 6. When Exim created an email address using the login of the caller as the
6436 local part (e.g. when creating a From: or Sender: header line), it was not
6437 quoting the local part when it contained special characters such as @.
6439 7. Installed new OpenBSD configuration files.
6441 8. Reworded some messages for syntax errors in "and" and "or" conditions to
6442 try to make them clearer.
6444 9. Callout options, other than the timeout value, were being ignored when
6445 verifying sender addresses in header lines. For example, when using
6447 verify = header_sender/callout=no_cache
6449 the cache was (incorrectly) being used.
6451 10. Added a missing instance of ${EXE} to the exim_install script; this affects
6452 only the Cygwin environment.
6454 11. When return_path_on_delivery was set as a log selector, if different remote
6455 addresses in the same message used different return paths and parallel
6456 remote delivery occurred, the wrong values would sometimes be logged.
6457 (Whenever a remote delivery process finished, the return path value from
6458 the most recently started remote delivery process was logged.)
6460 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
6461 lines when AUTH and/or TLS are in use. This is the "received protocol"
6462 field. Exim used to use "asmtp" for authenticated SMTP, without any
6463 indication (in the protocol name) for TLS use. Now it follows the RFC and
6464 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
6465 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
6466 names appear in log lines as well as in Received: header lines.
6468 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
6469 extensions, and comparison tests, and to fix some bugs.
6471 14. Changes to the "personal" filter test:
6473 (1) The test was buggy in that it was just doing the equivalent of
6474 "contains" tests on header lines. For example, if a user's address was
6475 anne@some.where, the "personal" test would incorrectly be true for
6477 To: susanne@some.where
6479 This test is now done by extracting each address from the header in turn,
6480 and checking the entire address. Other tests that are part of "personal"
6481 are now done using regular expressions (for example, to check local parts
6482 of addresses in From: header lines).
6484 (2) The list of non-personal local parts in From: addresses has been
6485 extended to include "listserv", "majordomo", "*-request", and "owner-*",
6486 taken from the Sieve specification recommendations.
6488 (3) If the message contains any header line starting with "List-" it is
6489 treated as non-personal.
6491 (4) The test for "circular" in the Subject: header line has been removed
6492 because it now seems ill-conceived.
6494 15. Minor typos in src/EDITME comments corrected.
6496 16. Installed latest exipick from John Jetmore.
6498 17. If headers_add on a router specified a text string that was too long for
6499 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
6500 of string_sprintf() is now avoided.
6502 18. $message_body_size was not set (it was always zero) when running the DATA
6503 ACL and the local_scan() function.
6505 19. For the "mail" command in an Exim filter, no default was being set for
6506 the once_repeat time, causing a random time value to be used if "once" was
6507 specified. (If the value happened to be <= 0, no repeat happened.) The
6508 default is now 0s, meaning "never repeat". The "vacation" command was OK
6509 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
6510 (I found it when inspecting the code).
6512 20. There is now an overall timeout for performing a callout verification. It
6513 defaults to 4 times the callout timeout, which applies to individual SMTP
6514 commands during the callout. The overall timeout applies when there is more
6515 than one host that can be tried. The timeout is checked before trying the
6516 next host. This prevents very long delays if there are a large number of
6517 hosts and all are timing out (e.g. when the network connections are timing
6518 out). The value of the overall timeout can be changed by specifying an
6519 additional sub-option for "callout", called "maxwait". For example:
6521 verify = sender/callout=5s,maxwait=20s
6523 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
6524 to the end before writing, but this should make it even safer).
6526 22. Exim was forgetting that it had advertised PIPELINING for the second and
6527 subsequent messages on an SMTP connection. It was also not resetting its
6528 memory on STARTTLS and an internal HELO.
6530 23. When Exim logs an SMTP synchronization error within a session, it now
6531 records whether PIPELINING has been advertised or not.
6533 24. Added 3 instances of "(long int)" casts to time_t variables that were being
6534 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
6535 rather than long int.
6537 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
6539 26. Added the never_mail option to autoreply.
6545 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
6546 crash if the getsockname() call failed; this can happen if a connection is
6547 closed very soon after it is established. The problem was simply in the
6548 order in which certain operations were done, causing Exim to try to write
6549 to the SMTP stream before it had set up the file descriptor. The bug has
6550 been fixed by making things happen in the correct order.
6556 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
6557 before the connection was closed, thus losing the rejection response.
6559 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
6560 some early Solaris releases, but causes trouble in current releases where
6561 socklen_t is defined.
6563 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
6566 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
6569 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
6570 that releasing the top of it at the end releases what was used for sub-
6571 expansions (unless the block got too big). However, discard this block if
6572 the first thing is a variable or header, so that we can use its block when
6573 it is dynamic (useful for very large $message_headers, for example).
6575 6. Lookups now cache *every* query, not just the most recent. A new, separate
6576 store pool is used for this. It can be recovered when all lookup caches are
6577 flushed. Lookups now release memory at the end of their result strings.
6578 This has involved some general refactoring of the lookup sources.
6580 7. Some code has been added to the store_xxx() functions to reduce the amount
6581 of flapping under certain conditions.
6583 8. log_incoming_interface used to affect only the <= reception log lines. Now
6584 it causes the local interface and port to be added to several more SMTP log
6585 lines, for example "SMTP connection from", and rejection lines.
6587 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
6589 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
6591 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
6592 could be overwritten at the end of the current message (or the start of a
6593 new message if it was set in a HELO ACL). The value is now preserved for
6594 the duration of the SMTP connection.
6596 12. If a transport had a headers_rewrite setting, and a matching header line
6597 contained an unqualified address, that address was qualified, even if it
6598 did not match any rewriting rules. The underlying bug was that the values
6599 of the flags that permit the existence of unqualified sender and recipient
6600 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
6601 non-local messages, and by -bnq for local messages) were not being
6602 preserved with the message after it was received.
6604 13. When Exim was logging an SMTP synchronization error, it could sometimes log
6605 "next input=" as part of the text comprising the host identity instead of
6606 the correct text. The code was using the same buffer for two different
6607 strings. However, depending on which order the printing function evaluated
6608 its arguments, the bug did not always show up. Under Linux, for example, my
6609 test suite worked just fine.
6611 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
6612 doesn't work with some older versions of Perl. It has been changed to "my",
6613 which in any case is probably the better facility to use.
6615 15. A really picky compiler found some instances of statements for creating
6616 error messages that either had too many or two few arguments for the format
6619 16. The size of the buffer for calls to the DNS resolver has been increased
6620 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
6621 for addresses that have a lot of PTR records. This alleviates a problem; it
6622 does not fully solve it.
6624 17. A dnsdb lookup for PTR records that receives more data than will fit in the
6625 buffer now truncates the list and logs the incident, which is the same
6626 action as happens when Exim is looking up a host name and its aliases.
6627 Previously in this situation something unpredictable would happen;
6628 sometimes it was "internal error: store_reset failed".
6630 18. If a server dropped the connection unexpectedly when an Exim client was
6631 using GnuTLS and trying to read a response, the client delivery process
6632 crashed while trying to generate an error log message.
6634 19. If a "warn" verb in an ACL added multiple headers to a message in a single
6635 string, for example:
6637 warn message = H1: something\nH2: something
6639 the text was added as a single header line from Exim's point of view
6640 though it ended up OK in the delivered message. However, searching for the
6641 second and subsequent header lines using $h_h2: did not work. This has been
6642 fixed. Similarly, if a system filter added multiple headers in this way,
6643 the routers could not see them.
6645 20. Expanded the error message when iplsearch is called with an invalid key to
6646 suggest using net-iplsearch in a host list.
6648 21. When running tests using -bh, any delays imposed by "delay" modifiers in
6649 ACLs are no longer actually imposed (and a message to that effect is
6652 22. If a "gecos" field in a passwd entry contained escaped characters, in
6653 particular, if it contained a \" sequence, Exim got it wrong when building
6654 a From: or a Sender: header from that name. A second bug also caused
6655 incorrect handling when an unquoted " was present following a character
6656 that needed quoting.
6658 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
6659 was not being matched caselessly.
6661 24. Arranged for all hyphens in the exim.8 source to be escaped with
6664 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
6665 properly. Recipient callout cache records were still being keyed to include
6666 the sender, even when use_sender was set false. This led to far more
6667 callouts that were necessary. The sender is no longer included in the key
6668 when use_sender is false.
6670 26. Added "control = submission" modifier to ACLs.
6672 27. Added the ${base62d: operator to decode base 62 numbers.
6674 28. dnsdb lookups can now access SRV records.
6676 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
6677 the configuration file.
6679 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
6680 (-v) mode. This makes the output for a verbose queue run more intelligible.
6682 31. Added a use_postmaster feature to recipient callouts.
6684 32. Added the $body_zerocount variable, containing the number of binary zero
6685 bytes in the message body.
6687 33. The time of last modification of the "new" subdirectory is now used as the
6688 "mailbox time last read" when there is a quota error for a maildir
6691 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
6693 35. Added +ignore_unknown as a special item in host lists.
6695 36. Code for decoding IPv6 addresses in host lists is now included, even if
6696 IPv6 support is not being compiled. This fixes a bug in which an IPv6
6697 address was recognized as an IP address, but was then not correctly decoded
6698 into binary, causing unexpected and incorrect effects when compared with
6705 1. Very minor rewording of debugging text in manualroute to say "list of
6706 hosts" instead of "hostlist".
6708 2. If verify=header_syntax was set, and a header line with an unqualified
6709 address (no domain) and a large number of spaces between the end of the
6710 name and the colon was received, the reception process suffered a buffer
6711 overflow, and (when I tested it) crashed. This was caused by some obsolete
6712 code that should have been removed. The fix is to remove it!
6714 3. When running in the test harness, delay a bit after writing a bounce
6715 message to get a bit more predictability in the log output.
6717 4. Added a call to search_tidyup() just before forking a reception process. In
6718 theory, someone could use a lookup in the expansion of smtp_accept_max_
6719 per_host which, without the tidyup, could leave open a database connection.
6721 5. Added the variables $recipient_data and $sender_data which get set from a
6722 lookup success in an ACL "recipients" or "senders" condition, or a router
6723 "senders" option, similar to $domain_data and $local_part_data.
6725 6. Moved the writing of debug_print from before to after the "senders" test
6728 7. Change 4.31/66 (moving the time when the Received: is generated) caused
6729 problems for message scanning, either using a data ACL, or using
6730 local_scan() because the Received: header was not generated till after they
6731 were called (in order to set the time as the time of reception completion).
6732 I have revised the way this works. The header is now generated after the
6733 body is received, but before the ACL or local_scan() are called. After they
6734 are run, the timestamp in the header is updated.
6740 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
6741 before starting a queue runner without re-exec. This happened only when
6742 deliver_drop_privilege was set or when the Exim user was set to root. The
6743 effect of the bug was that timeouts during subsequent deliveries caused
6744 crashes instead of being properly handled. The handler is now left at its
6745 default (and expected) setting.
6747 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
6748 message, again when deliver_drop_privilege is set or Exim is run as root.
6749 The bug described in (1) was not present in this case, but the tidying up
6750 of the other signals was missing. I have made the two cases consistent.
6752 3. The ignore_target_hosts setting on a manualroute router was being ignored
6753 for hosts that were looked up using the /MX notation.
6755 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
6758 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
6759 operated on the sender address. After changing the $sender_address to <>
6760 for the sender address verify, Exim was re-instated it as the original
6761 (before rewriting) address, but remembering that it had rewritten it, so it
6762 wasn't rewriting it again. This bug also had the effect of breaking the
6763 sender address verification caching when the sender address was rewritten.
6765 6. The ignore_target_hosts option was being ignored by the ipliteral router.
6766 This has been changed so that if the ip literal address matches
6767 ignore_target_hosts, the router declines.
6769 7. Added expansion conditions match_domain, match_address, and match_local_
6770 part (NOT match_host).
6772 8. The placeholder for the Received: header didn't have a length field set.
6774 9. Added code to Exim itself and to exim_lock to test for a specific race
6775 condition that could lead to file corruption when using MBX delivery. The
6776 issue is with the lockfile that is created in /tmp. If this file is removed
6777 after a process has opened it but before that process has acquired a lock,
6778 there is the potential for a second process to recreate the file and also
6779 acquire a lock. This could lead to two Exim processes writing to the file
6780 at the same time. The added code performs the same test as UW imapd; it
6781 checks after acquiring the lock that its file descriptor still refers to
6782 the same named file.
6784 10. The buffer for building added header lines was of fixed size, 8192 bytes.
6785 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
6788 11. Added the smtp_active_hostname option. If used, this will typically be made
6789 to depend on the incoming interface address. Because $interface_address is
6790 not set up until the daemon has forked a reception process, error responses
6791 that can happen earlier (such as "too many connections") no longer contain
6794 12. If an expansion in a condition on a "warn" statement fails because a lookup
6795 defers, the "warn" statement is abandoned, and the next ACL statement is
6796 processed. Previously this caused the whole ACL to be aborted.
6798 13. Added the iplsearch lookup type.
6800 14. Added ident_timeout as a log selector.
6802 15. Added tls_certificate_verified as a log selector.
6804 16. Added a global option tls_require_ciphers (compare the smtp transport
6805 option of the same name). This controls incoming TLS connections.
6807 17. I finally figured out how to make tls_require_ciphers do a similar thing
6808 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
6809 before starting the TLS session.
6811 18. Tabs are now shown as \t in -bP output.
6813 19. If the log selector return_path_on_delivery was set, Exim crashed when
6814 bouncing a message because it had too many Received: header lines.
6816 20. If two routers both had headers_remove settings, and the first one included
6817 a superfluous trailing colon, the final name in the first list and the
6818 first name in the second list were incorrectly joined into one item (with a
6819 colon in the middle).
6825 1. Added -C and -D options to the exinext utility, mainly to make it easier
6826 to include in the automated testing, but these could be helpful when
6827 multiple configurations are in use.
6829 2. The exinext utility was not formatting the output nicely when there was
6830 an alternate port involved in the retry record key, nor when there was a
6831 message id as well (for retries that were specific to a specific message
6832 and a specific host). It was also confused by IPv6 addresses, because of
6833 the additional colons they contain. I have fixed the IPv4 problem, and
6834 patched it up to do a reasonable job for IPv6.
6836 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
6837 delivery, the log line now contains "pipelined" if PIPELINING was used.
6839 4. An SMTP transport process used to panic and die if the bind() call to set
6840 an explicit outgoing interface failed. This has been changed; it is now
6841 treated in the same way as a connect() failure.
6843 5. A reference to $sender_host_name in the part of a conditional expansion
6844 that was being skipped was still causing a DNS lookup. This no longer
6847 6. The def: expansion condition was not recognizing references to header lines
6848 that used bh_ and bheader_.
6850 7. Added the _cache feature to named lists.
6852 8. The code for checking quota_filecount in the appendfile transport was
6853 allowing one more file than it should have been.
6855 9. For compatibility with Sendmail, the command line option
6863 and sets the incoming protocol and host name (for trusted callers). The
6864 host name and its colon can be omitted when only the protocol is to be set.
6865 Note the Exim already has two private options, -pd and -ps, that refer to
6866 embedded Perl. It is therefore impossible to set a protocol value of "d" or
6867 "s", but I don't think that's a major issue.
6869 10. A number of refactoring changes to the code, none of which should affect
6872 (a) The number of logging options was getting close to filling up the
6873 32-bit word that was used as a bit map. I have split them into two classes:
6874 those that are passed in the argument to log_write(), and those that are
6875 only ever tested independently outside of that function. These are now in
6876 separate 32-bit words, so there is plenty of room for expansion again.
6877 There is no change in the user interface or the logging behaviour.
6879 (b) When building, for example, log lines, the code previously used a
6880 macro that called string_cat() twice, in order to add two strings. This is
6881 not really sufficiently general. Furthermore, there was one instance where
6882 it was actually wrong because one of the argument was used twice, and in
6883 one call a function was used. (As it happened, calling the function twice
6884 did not affect the overall behaviour.) The macro has been replaced by a
6885 function that can join an arbitrary number of extra strings onto a growing
6888 (c) The code for expansion conditions now uses a table and a binary chop
6889 instead of a serial search (which was left over from when there were very
6890 few conditions). Also, it now recognizes conditions like "pam" even when
6891 the relevant support is not compiled in: a suitably worded error message is
6892 given if an attempt is made to use such a condition.
6894 11. Added ${time_interval:xxxxx}.
6896 12. A bug was causing one of the ddress fields not to be passed back correctly
6897 from remote delivery subprocesses. The field in question was not being
6898 subsequently used, so this caused to problems in practice.
6900 13. Added new log selectors queue_time and deliver_time.
6902 14. Might have fixed a bug in maildirsizefile handling that threw up
6903 "unexpected character" debug warnings, and recalculated the data
6904 unnecessarily. In any case, I expanded the warning message to give more
6907 15. Added the message "Restricted characters in address" to the statements in
6908 the default ACL that block characters like @ and % in local parts.
6910 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
6911 Three changes have been made:
6913 (a) There was a serious bug; a negative response to MAIL caused the whole
6914 recipient domain to be cached as invalid, thereby blocking all messages
6915 to all local parts at the same domain, from all senders. This bug has
6916 been fixed. The domain is no longer cached after a negative response to
6917 MAIL if the sender used is not empty.
6919 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
6922 (c) A new callout option, "use_sender" has been added for people who want
6923 the modified behaviour.
6929 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
6932 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
6933 indeed breaks things for older releases.
6935 3. Added additional logging to the case where there is a problem reading data
6936 from a filter that is running in a subprocess using a pipe, in order to
6937 try to track down a specific problem.
6939 4. Testing facility fudge: when running in the test harness and attempting
6940 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
6941 getting "No route to host". Convert this to a timeout.
6943 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
6946 6. Some OS don't have socklen_t but use size_t instead. This affects the
6947 fifth argument of getsockopt() amongst other things. This is now
6948 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
6949 can be set for individual OS. I have set it for SunOS5, OSF1, and
6950 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
6951 some earlier ones do not.
6953 7. Change 4.30/15 was not doing the test caselessly.
6955 8. The standard form for an IPv6 address literal was being rejected by address
6956 parsing in, for example, MAIL and RCPT commands. An example of this kind of
6957 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
6958 this, as well as the form without the "IPv6" on the front (but only when
6959 address literals are enabled, of course).
6961 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
6963 10. Exim crashed if a message with an empty sender address specified by -f
6964 encountered a router with an errors_to setting. This could be provoked only
6965 by a command such as
6969 where an empty string was supplied; "<>" did not hit this bug.
6971 11. Installed PCRE release 4.5.
6973 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
6974 remained set. It is now erased.
6976 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
6977 times from message ids (which are base 36 rather than the normal 62).
6979 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
6980 were being counted as actual protocol errors, and logged if the log
6981 selector +smtp_protocol_error was set. One cannot be perfect in this test,
6982 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
6983 and DATA following a set of rejected RCPTs do not count as protocol errors.
6984 In other words, Exim assumes they were pipelined, though this may not
6985 actually be the case. Of course, in all cases the client gets an
6986 appropriate error code.
6988 15. If a lookup fails in an ACL condition, a message about the failure may
6989 be available; it is used if testing the ACL cannot continue, because most
6990 such messages specify what the cause of the deferral is. However, some
6991 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
6992 that caused an old message to be retained and used if a later statement
6993 caused a defer, replacing the real cause of the deferral.
6995 16. If an IP address had so many PTR records that the DNS lookup buffer
6996 was not large enough to hold them, Exim could crash while trying to process
6997 the truncated data. It now detects and logs this case.
6999 17. Further to 4.21/58, another change has been made: if (and only if) the
7000 first line of a message (the first header line) ends with CRLF, a bare LF
7001 in a subsequent header line has a space inserted after it, so as not to
7002 terminate the header.
7004 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
7005 unnecessarily, used atoi() instead of strtol(), and didn't check the
7006 termination when getting file sizes from file names by regex.
7008 19. Completely re-implemented the support for maildirsize files, in the light
7009 of a number of problems with the previous contributed implementation
7010 (4.30/29). In particular:
7012 . If the quota is zero, the maildirsize file is maintained, but no quota is
7015 . If the maildir directory does not exist, it is created before any attempt
7016 to write a maildirsize file.
7018 . The quota value in the file is just a cache; if the quota is changed in
7019 the transport, the new value overrides.
7021 . A regular expression is available for excluding directories from the
7024 20. The autoreply transport checks the characters in options that define the
7025 message's headers; it allows continued headers, but it was checking with
7026 isspace() after an embedded newline instead of explicitly looking for a
7029 21. If all the "regular" hosts to which an address was routed had passed their
7030 expiry times, and had not reached their retry times, the address was
7031 bounced, even if fallback hosts were defined. Now Exim should go on to try
7034 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
7035 equivalent code in the SMTP transport. Some hosts send humungous responses
7036 to HELO/EHLO, more than 1024 it seems.
7038 23. Refactoring: code in filter.c used (void *) for "any old type" but this
7039 gives compiler warnings in some environments. I've now done it "properly",
7042 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
7043 (because of problems with the built-in one) was declared to return uschar *
7044 instead of char *, causing compiler failure.
7046 25. Fixed a file descriptor leak when processing alias/forward files.
7048 26. Fixed a minor format string issue in dbfn.c.
7050 27. Typo in exim.c: ("dmbnz" for "dbmnz").
7052 28. If a filter file refered to $h_xxx or $message_headers, and the headers
7053 contained RFC 2047 "words", Exim's memory could, under certain conditions,
7056 29. When a sender address is verified, it is cached, to save repeating the test
7057 when there is more than one recipient in a message. However, when the
7058 verification involves a callout, it is possible for different callout
7059 options to be set for different recipients. It is too complicated to keep
7060 track of this in the cache, so now Exim always runs a verification when a
7061 callout is required, relying on the callout cache for the optimization.
7062 The overhead is duplication of the address routing, but this should not be
7065 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
7066 to be verified with callout=postmaster, and the main callout worked but the
7067 postmaster check failed, the verification correctly failed. However, if a
7068 subsequent RCPT command asked for sender verification *without* the
7069 postmaster check, incorrect caching caused this verification also to fail,
7072 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
7073 it was not caching the DNS options (qualify_single, search_parents) that
7074 were used when the lookup failed. A subsequent lookup with different
7075 options therefore always gave the same answer, though there were cases
7076 where it should not have. (Example: a "domains = !$mx_any" option on a
7077 dnslookup router: the "domains" option is always processed without any
7078 widening, but the router might have qualify_single set.) Now Exim uses the
7079 cached value only when the same options are set.
7081 32. Added John Jetmore's "exipick" utility to the distribution.
7083 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
7084 than a timeout (e.g. a certificate is required, and is not provided), an
7085 Exim server now closes the connection immediately. Previously it waited for
7086 the client to close - but if the client is SSL, it seems that they each
7087 wait for each other, leading to a delay before one of them times out.
7089 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
7090 maintained 0.8.x compatibility because I don't think many are using it, and
7091 it is clearly obsolete.
7093 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
7096 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
7097 client certificate was expired. A simple patch fixes this, though I don't
7098 understand the full logic of why the verify callback is called multiple
7101 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
7102 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
7103 which causes problems with some clients (such as the Certicom SSL Plus
7104 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
7105 disables the coutermeasure allowing Eudora to connect."
7107 38. Exim was not checking that a write() to a log file succeeded. This could
7108 lead to Bad Things if a log got too big, in particular if it hit a file
7109 size limit. Exim now panics and dies if it cannot write to a log file, just
7110 as it does if it cannot open a log file.
7112 39. Modified OS/Makefile-Linux so that it now contains
7114 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
7116 The two -D definitions ensure that Exim is compiled with large file
7117 support, which makes it possible to handle log files that are bigger than
7120 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
7121 instance) a domain was checked against a named list that involved a lookup,
7122 causing $domain_data to be set, then another domain was checked against the
7123 same list, then the first domain was re-checked, the value of $domain_data
7124 after the final check could be wrong. In particular, if the second check
7125 failed, it could be set empty. This bug probably also applied to
7128 41. The strip_trailing_dot option was not being applied to the address given
7129 with the -f command-line option.
7131 42. The code for reading a message's header from the spool was incrementing
7132 $received_count, but never initializing it. This meant that the value was
7133 incorrect (doubled) while delivering a message in the same process in which
7134 it was received. In the most common configuration of Exim, this never
7135 happens - a fresh exec is done - but it can happen when
7136 deliver_drop_privilege is set.
7138 43. When Exim logs an SMTP synchronization error - client data sent too soon -
7139 it now includes up to 150 characters of the unexpected data in the log
7142 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
7143 and building data strings. The size of both of these buffers was 10 000
7144 bytes - far larger than anybody would *ever* want, thought I. Needless to
7145 say, somebody hit the limit. I have increased the maximum line length to
7146 20 000 and the maximum data length of concatenated lines to 100 000. I have
7147 also fixed two bugs, because there was no checking on these buffers. Tsk,
7148 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
7149 buffer is too small.
7151 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
7152 lsearch lookups. Now it does.
7154 46. When parsing a route_list item in a manualroute router, a fixed-length
7155 buffer was used for the list of hosts. I made this 1024 bytes long,
7156 thinking that nobody would ever have a list of hosts that long. Wrong.
7157 Somebody had a whole pile of complicated expansion conditions, and the
7158 string was silently truncated, leading to an expansion error. It turns out
7159 that it is easier to change to an unlimited length (owing to other changes
7160 that have happened since this code was originally written) than to build
7161 structure for giving a limitation error. The length of the item that
7162 expands into the list of hosts is now unlimited.
7164 47. The lsearch lookup could not handle data where the length of text line was
7165 more than 4095 characters. Such lines were truncated, leading to shortened
7166 data being returned. It should now handle lines of any length.
7168 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
7169 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
7172 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
7173 replacing strings such as BIN_DIRECTORY in a source file: the replacement
7174 no longer happens in comment lines. A list of replacements is now placed
7175 at the head of all of the source files, except those whose only change is
7176 to replace PERL_COMMAND in the very first #! line.
7178 50. Replaced the slow insertion sort in queue.c, for sorting the list of
7179 messages on the queue, with a bottom-up merge sort, using code contributed
7180 by Michael Haardt. This should make operations like -bp somewhat faster on
7181 large queues. It won't affect queue runners, except when queue_run_in_order
7184 51. Installed eximstats 1.31 in the distribution.
7186 52. Added support for SRV lookups to the dnslookup router.
7188 53. If an ACL referred to $message_body or $message_body_end, the value was not
7189 reset for any messages that followed in the same SMTP session.
7191 54. The store-handling optimization for building very long strings was not
7192 differentiating between the different store pools. I don't think this
7193 actually made any difference in practice, but I've tidied it.
7195 55. While running the routers to verify a sender address, $sender_address
7196 was still set to the sender address. This is wrong, because when routing to
7197 send a bounce to the sender, it would be empty. Therefore, I have changed
7198 it so that, while verifying a sender address, $sender_address is set to <>.
7199 (There is no change to what happens when verifying a recipient address.)
7201 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
7202 target A or AAAA records (if not already returned) without resetting the
7203 qualify_single or search_parents options of the DNS resolver. These are
7204 inappropriate in this case because the targets of MX and SRV records must
7205 be FQDNs. A broken DNS record could cause trouble if it happened to have a
7206 target that, when qualified, matched something in the local domain. These
7207 two options are now turned off when doing these lookups.
7209 57. It seems that at least some releases of Reiserfs (which does not have the
7210 concept of a fixed number of inodes) returns zero and not -1 for the
7211 number of available inodes. This interacted badly with check_spool_inodes,
7212 which assumed that -1 was the "no such thing" setting. What I have done is
7213 to check that the total number of inodes is greater than zero before doing
7214 the test of how many are available.
7216 58. When a "warn" ACL statement has a log_message modifier, the message is
7217 remembered, and not repeated. This is to avoid a lot of repetition when a
7218 message has many recipients that cause the same warning to be written.
7219 However, Exim was preserving the list of already written lines for an
7220 entire SMTP session, which doesn't seem right. The memory is now reset if a
7221 new message is started.
7223 59. The "rewrite" debugging flag was not showing the result of rewriting in the
7224 debugging output unless log_rewrite was also set.
7226 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
7227 of (int)(handle) when we know that handle contains (void *)(-1).
7229 61. The Exim daemon panic-logs an error return when it closes the incoming
7230 connection. However "connection reset by peer" seems to be common, and
7231 isn't really an error worthy of noting specially, so that particular error
7234 62. When Exim is trying to find all the local interfaces, it used to panic and
7235 die if the ioctl to get the interface flags failed. However, it seems that
7236 on at least one OS (Solaris 9) it is possible to have an interface that is
7237 included in the list of interfaces, but for which you get a failure error
7238 for this call. This happens when the interface is not "plumbed" into a
7239 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
7240 failure of the "get flags" call assumes that the interface is down.
7242 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
7243 makes life easier for people who are doing arithmetic on fields extracted
7244 from dates, where you often get leading zeros that should not be
7245 interpreted as octal.
7247 64. Added qualify_domain to the redirect router, to override the global
7250 65. If a pathologically long header line contained very many addresses (the
7251 report of this problem mentioned 10 000) and each of them was rewritten,
7252 Exim could use up a very large amount of memory. (It kept on making new
7253 copies of the header line as it rewrote, and never released the old ones.)
7254 At the expense of a bit more processing, the header rewriting function has
7255 been changed so that it no longer eats memory in this way.
7257 66. The generation of the Received: header has been moved from the time that a
7258 message starts to be received, to the time that it finishes. The timestamp
7259 in the Received: header should now be very close to that of the <= log
7260 line. There are two side-effects of this change:
7262 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
7263 logged header lines no longer include the local Received: line, because
7264 it has not yet been created. The same applies to a copy of the message
7265 that is returned to a non-SMTP sender when a message is rejected.
7267 (b) When a filter file is tested using -bf, no additional Received: header
7268 is added to the test message. After some thought, I decided that this
7271 This change does not affect the value of $received_for. It is still set
7272 after address rewriting, but before local_scan() is called.
7274 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
7276 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
7277 gave an unhelpful panic error message, and a defer error. I have managed to
7278 change this behaviour so that it now rejects any supplied certificate,
7279 which seems right, as the list of acceptable certificates is empty.
7281 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
7282 gave an unhelpful defer error. I have not managed to make this reject any
7283 supplied certificates, but the error message it gives is "no certificate
7284 supplied", which is not helpful.
7286 70. exigrep's output now also includes lines that are not associated with any
7287 message, but which match the given pattern. Implemented by a patch from
7288 Martin Sluka, which also tidied up the Perl a bit.
7290 71. Recipient callout verification, like sender verification, was using <> in
7291 the MAIL FROM command. This isn't really the right thing, since the actual
7292 sender may affect whether the remote host accepts the recipient or not. I
7293 have changed it to use the actual sender in the callout; this means that
7294 the cache record is now keyed on a recipient/sender pair, not just the
7295 recipient address. There doesn't seem to be a real danger of callout loops,
7296 since a callout by the remote host to check the sender would use <>.
7297 [SEE ABOVE: changed after hitting problems.]
7299 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
7300 temporary errors. However, in the case of such a code being given after
7301 the end of a data transmission (i.e. after ".") Exim was failing to write
7302 a retry record for the message. (Yes, there was some broken host that was
7303 actually sending 8xx at this point.)
7305 73. An unknown lookup type in a host list could cause Exim to panic-die when
7306 the list was checked. (An example that provoked this was putting <; in the
7307 middle of a list instead of at the start.) If this happened during a DATA
7308 ACL check, a -D file could be left lying around. This kind of configuration
7309 error no longer causes Exim to die; instead it causes a defer error. The
7310 incident is still logged to the main and panic logs.
7312 74. Buglet left over from Exim 3 conversion. The message "too many messages
7313 in one connection" was written to the rejectlog but not the mainlog, except
7314 when address rewriting (yes!) was being logged.
7316 75. Added write_rejectlog option.
7318 76. When a system filter was run not as root (that is, when system_filter_user
7319 was set), the values of the $n variables were not being returned to the
7320 main process; thus, they were not subsequently available in the $sn
7323 77. Added +return_path_on_delivery log selector.
7325 78. A connection timeout was being treated differently from recipients deferred
7326 when testing hosts_max_try with a message that was older than the host's
7327 retry timeout. (The host should not be counted, thus allowing all hosts to
7328 be tried at least once before bouncing.) This may have been the cause of an
7329 occasionally reported bug whereby a message would remain on the queue
7330 longer than the retry timeout, but would be bounced if a delivery was
7331 forced. I say "may" because I never totally pinned down the problem;
7332 setting up timeout/retry tests is difficult. See also the next item.
7334 79. The ultimate address timeout was not being applied to errors that involved
7335 a combination of host plus message (for example, a timeout on a MAIL
7336 command). When an address resolved to a number of possible hosts, and they
7337 were not all tried for each delivery (e.g. because of hosts_max_try), a
7338 message could remain on the queue longer than the retry timeout.
7340 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
7343 81. Fixed an obscure SMTP outgoing bug which required at least the following
7344 conditions: (a) there was another message waiting for the same server;
7345 (b) the server returned 5xx to all RCPT commands in the first message so
7346 that the message was not completed; (c) the server dropped the connection
7347 or gave a negative response to the RSET that Exim sends to abort the
7348 transaction. The observed case was a dropped connection after DATA that had
7349 been sent in pipelining mode. That is, the server had advertised PIPELINING
7350 but was not implementing it correctly. The effect of the bug was incorrect
7351 behaviour, such as trying another host, and this could lead to a crash.
7357 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
7358 and daemon.c were passed as pointers to ints; they should have been
7359 pointers to socklen_t variables (which are typically unsigned ints).
7361 2. Some signed/unsigned type warnings in the os.c file for Linux have been
7364 3. Fixed a really odd bug that affected only the testing scheme; patching a
7365 certain fixed string in the binary changed the value of another string that
7366 happened to be identical to the end of the original first string.
7368 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
7369 name", it returns that address as the IP address. On some operating
7370 systems (e.g. Solaris), it also passes back the IP address string as the
7371 "host name". However, on others (e.g. Linux), it passes back an empty
7372 string. Exim wasn't checking for this, and was changing the host name to an
7373 empty string, assuming it had been canonicalized.
7375 5. Although rare, it is permitted to have more than one PTR record for a given
7376 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
7377 all the names associated with an address, because they do in Solaris.
7378 However, it seems that they do not in Linux for data that comes from the
7379 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
7380 I found this out when I moved to a new Linux workstation and tried to run
7381 the Exim test suite.
7383 To get round this problem I have changed the code so that it now does its
7384 own call to the DNS to look up PTR records when searching for a host name.
7385 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
7386 addresses that are only in /etc/hosts are still found.
7388 This behaviour is, however, controlled by an option called host_lookup_
7389 order, which defaults to "bydns:byaddr". If people want to use the other
7390 order, or indeed, just use one or the other means of lookup, they can
7391 specify it in this variable.
7393 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
7394 some operating systems, this comes back from gethostbyaddr() as an empty
7395 string, and this is what Exim used to test for. However, it seems that in
7396 other systems, "." is yielded. Exim now tests for this case too.
7398 7. The values of check_spool_space and check_log_space are now held internally
7399 as a number of kilobytes instead of an absolute number of bytes. If a
7400 numbers is specified without 'K' or 'M', it is rounded up to the nearest
7401 kilobyte. This means that much larger values can be stored.
7403 8. Exim monitor: an attempt to get the action menu when not actually pointing
7404 at a message produces an empty menu entitled "No message selected". This
7405 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
7406 no entries in it ("Shell widget menu has zero width and/or height"). So I
7407 have added a single, blank menu entry in this case.
7409 9. Added ${quote_local_part.
7411 10. MIME decoding is now applied to the contents of Subject: header lines when
7414 11. Now that a reference to $sender_host_address automatically causes a reverse
7415 lookup to occur if necessary (4.13/18), there is no need to arrange for a
7416 host lookup before query-style lookups in lists that might use this
7417 variable. This has therefore been abolished, and the "net-" prefix is no
7418 longer necessary for query-style lookups.
7420 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
7421 have been a typo for LFLAGS, so it has been changed.
7423 13. The install script calls Exim with "-C /dev/null" in order to find the
7424 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
7425 to be output. However, since Exim outputs its version number before the
7426 error, it didn't break the script. It just looked ugly. I fixed this by
7427 always allowing "-C /dev/null" if the caller is root.
7429 14. Ignore overlarge ACL variable number when reading spool file - insurance
7430 against a later release with more variables having written the file.
7432 15. The standard form for an IPv6 address literal was being rejected by EHLO.
7433 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
7434 this, as well as the form without the "IPv6" on the front.
7436 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
7437 OS/Makefile-Darwin file.
7439 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
7440 only to LDAP libraries that do not have LDAP_OPT_DEREF.
7442 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
7443 variable. However, some OS use ints for inodes. I've added cast to long int
7444 to get rid of the compiler warning.
7446 19. I had forgotten to lock out "/../" in configuration file names when
7447 ALT_CONFIG_PREFIX was set.
7449 20. Routers used for verification do not need to specify transports. However,
7450 if such a router generated a host list, and callout was configured, Exim
7451 crashed, because it could not find a port number from the (non-existent)
7452 transport. It now assumes port 25 in this circumstance.
7454 21. Added the -t option to exigrep.
7456 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
7457 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
7458 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
7459 them is set, LOOKUP_LSEARCH is forced.
7461 23. "exim -bV" now outputs a list of lookups that are included in the binary.
7463 24. Added sender and host information to the "rejected by local_scan()" log
7464 line; previously there was no indication of these.
7466 25. Added .include_if_exists.
7468 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
7469 Linux. It turns out that not all file systems support this. Apparently some
7470 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
7471 do it.) To cope with this, the error EINVAL, which means that sync-ing is
7472 not supported on the file descriptor, is now ignored when Exim is trying to
7473 sync a directory. This applies only to Linux.
7475 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
7477 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
7478 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
7481 29. Added support for maildirsize files from supplied patch (modified a bit).
7483 30. The use of :fail: followed by an empty string could lead Exim to respond to
7484 sender verification failures with (e.g.):
7486 550 Verification failed for <xxx>
7487 550 Sender verify failed
7489 where the first response line was missing the '-' that indicates it is not
7490 the final line of the response.
7492 31. The loop for finding the name of the user that called Exim had a hardwired
7493 limit of 10; it now uses the value of finduser_retries, which is used for
7494 all other user lookups.
7496 32. Added $received_count variable, available in data and not_smtp ACLs, and at
7499 33. Exim was neglecting to zero errno before one call of strtol() when
7500 expanding a string and expecting an integer value. On some systems this
7501 resulted in spurious "integer overflow" errors. Also, it was casting the
7502 result into an int without checking.
7504 34. Testing for a connection timeout using "timeout_connect" in the retry rules
7505 did not work. The code looks as if it has *never* worked, though it appears
7506 to have been documented since at least release 1.62. I have made it work.
7508 35. The "timeout_DNS" error in retry rules, also documented since at least
7509 1.62, also never worked. As it isn't clear exactly what this means, and
7510 clearly it isn't a major issue, I have abolished the feature by treating it
7511 as "timeout", and writing a warning to the main and panic logs.
7513 36. The display of retry rules for -brt wasn't always showing the error code
7516 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
7517 timeout_connect_A, timeout_connect_MX.
7519 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
7520 to the empty sender.
7522 39. The daemon was not analysing the content of -oX till after it had closed
7523 stderr and disconnected from the controlling terminal. This meant that any
7524 syntax errors were only noted on the panic log, and the return code from
7525 the command was 0. By re-arranging the code a little, I've made the
7526 decoding happen first, so such errors now appear on stderr, and the return
7527 code is 1. However, the actual setting up of the sockets still happens in
7528 the disconnected process, so errors there are still only recorded on the
7531 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
7532 connections (as happens on some IP stacks) was logged at start up time as
7533 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
7534 it from "IPv6 and IPv4", which means that two separate sockets are being
7537 41. The debug output for gethostbyname2() or getipnodebyname() failures now
7538 says whether AF_INET or AF_INET6 was passed as an argument.
7540 42. Exiwhat output was messed up when time zones were included in log
7543 43. Exiwhat now gives more information about the daemon's listening ports,
7544 and whether -tls-on-connect was used.
7546 44. The "port" option of the smtp transport is now expanded.
7548 45. A "message" modifier in a "warn" statement in a non-message ACL was being
7549 silently ignored. Now an error message is written to the main and panic
7552 46. There's a new ACL modifier called "logwrite" which writes to a log file
7553 as soon as it is encountered.
7555 47. Added $local_user_uid and $local_user_gid at routing time.
7557 48. Exim crashed when trying to verify a sender address that was being
7560 49. Exim was recognizing only a space character after ".include". It now also
7561 recognizes a tab character.
7563 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
7564 extracting the relevant information from the specification. The man page no
7565 longer contains scrambled data for the -d option, and I've added a section
7566 at the front about calling Exim under different names.
7568 51. Added "extra_headers" argument to the "mail" command in filter files.
7570 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
7573 53. Installed eximstats 1.29.
7575 54. Added transport_filter_timeout as a generic transport option.
7577 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
7578 Cc: header lines. This was required by RFC 822, but it not required by RFC
7581 56. Exim used to add From:, Date:, and Message-Id: header lines to any
7582 incoming messages that did not have them. Now it does so only if the
7583 message originates locally, that is, if there is no associated remote host
7584 address. When Resent- header lines are present, this applies to the Resent-
7585 lines rather than the non-Resent- lines.
7587 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
7588 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
7590 58. Messages for configuration errors now include the name of the main
7591 configuration file - useful now that there may be more than one file in a
7592 list (.included file names were always shown).
7594 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
7595 for those rare installations that do not start the daemon as root or run it
7596 setuid root. I've cut out the call to initgroups() if the daemon is not
7599 60. The Exim user and group can now be bound into the binary as text strings
7600 that are looked up at the start of Exim's processing.
7602 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
7604 62. Added $mailstore_basename variable.
7606 63. Installed patch to sieve.c from Michael Haardt.
7608 64. When Exim failed to open the panic log after failing to open the main log,
7609 the original message it was trying to log was written to stderr and debug
7610 output, but if they were not available (the usual case in production), it
7611 was lost. Now it is written to syslog before the two lines that record the
7612 failures to open the logs.
7614 65. Users' Exim filters run in subprocesses under the user's uid. It is
7615 possible for a "deliver" command or an alias in a "personal" command to
7616 provoke an address rewrite. If logging of address rewriting is configured,
7617 this fails because the process is not running as root or exim. There may be
7618 a better way of dealing with this, but for the moment (because 4.30 needs
7619 to be released), I have disabled address rewrite logging when running a
7620 filter in a non-root, non-exim process.
7626 1. The buildconfig auxiliary program wasn't quoting the value set for
7627 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
7628 not defined. This bug was masked in 4.22 by the effect that was fixed in
7631 2. Some messages that were rejected after a message id was allocated were
7632 shown as "incomplete" by exigrep. It no longer does this for messages that
7633 are rejected by local_scan() or the DATA or non-SMTP ACLs.
7635 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
7636 have allow_domain_literals set, the ID did not get logged in the <= line.
7637 Domain literals are now always recognized in Message-ID: header lines.
7639 4. The first argument for a ${extract expansion item is the key name or field
7640 number. Leading and trailing spaces in this item were not being ignored,
7641 causing some misleading effects.
7643 5. When deliver_drop_privilege was set, single queue runner processes started
7644 manually (i.e. by the command "exim -q") or by the daemon (which uses the
7645 same command in the process it spins off) were not dropping privilege.
7647 6. When the daemon running as "exim" started a queue runner, it always
7648 re-executed Exim in the spun-off process. This is a waste of effort when
7649 deliver_drop_privilege is set. The new process now just calls the
7650 queue-runner function directly.
7656 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
7659 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
7660 ignored. Though the use of -oP was forcing the writing of a pid file, it
7661 was always written to the default place.
7663 3. If the message "no IP address found for host xxxx" is generated during
7664 incoming verification, it is now followed by identification of the incoming
7665 connection (so you can more easily find what provoked it).
7667 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
7669 5. Added some features to "harden" Exim a bit more against certain attacks:
7671 (a) There is now a build-time option called FIXED_NEVER_USERS that can
7672 be put in Local/Makefile. This is like the never_users runtime option,
7673 but it cannot be overridden. The default setting is "root".
7675 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
7676 prefix string with which any file named in a -C command line option
7679 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
7680 is retained for -C and -D only if the caller of Exim is root. Without
7681 it, the exim user may also use -C and -D and retain privilege.
7683 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
7684 command line option is disabled.
7686 6. Macro names set by the -D option must start with an upper case letter, just
7687 like macro names defined in the configuration file.
7689 7. Added "dereference=" facility to LDAP.
7691 8. Two instances of the typo "uknown" in the source files are fixed.
7693 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
7694 the Configure-Makefile script screwed up while processing it.
7696 10. Incorporated PCRE 4.4.
7698 11. The SMTP synchronization check was not operating right at the start of an
7699 SMTP session. For example, it could not catch a HELO sent before the client
7700 waited for the greeting. There is now a check for outstanding input at the
7701 point when the greeting is written. Because of the duplex, asynchronous
7702 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
7703 way, but not yet received, when the check is performed.
7705 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
7706 on TCP/IP sockets, because this apparently causes some broken clients to
7709 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
7710 unchanged) from the Cygwin maintainer.
7712 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
7713 was supported instead of testing for mbx. Effectively a typo.
7715 15. The spa authenticator server code was not checking that the input it
7716 received was valid base64.
7718 16. The debug output line for the "set" modifier in ACLs was not showing the
7719 name of the variable that was being set.
7721 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
7723 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
7724 lookup on its own. Something else was needed to trigger the lookup. For
7725 example, a match in host_lookup or the need for a host name in a host list.
7726 Now, if $sender_host_name is referenced and the host name has not yet been
7727 looked up, a lookup is performed. If the lookup fails, the variable remains
7728 empty, and $host_lookup_failed is set to "1".
7730 19. Added "eqi" as a case-independent comparison operator.
7732 20. The saslauthd authentication condition could segfault if neither service
7733 nor realm was specified.
7735 21. If an overflowing value such as "2048M" was set for message_size_limit, the
7736 error message that was logged was misleading, and incoming SMTP
7737 connections were dropped. The message is now more accurate, and temporary
7738 errors are given to SMTP connections.
7740 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
7741 (except RSET) with a 421 error, until QUIT is received. However, it was
7742 failing to send a response to QUIT.
7744 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
7745 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
7746 ACL is now run after the helo_try_verify_hosts code.
7748 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
7749 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
7750 case-independent, so other case variants are also recognized). Apparently
7751 some systems use these upper case variants.
7753 25. If more than two messages were waiting for the same host, and a transport
7754 filter was specified for the transport, Exim sent two messages over the
7755 same TCP/IP connection, and then failed with "socket operation on non-
7756 socket" when it tried to send the third.
7758 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
7760 27. The extern definition of crypt16() in expand.c was not being excluded when
7761 the OS had its own crypt16() function.
7763 28. Added bounce_return_body as a new option, and bounce_return_size_limit
7764 as a preferred synonym for return_size_limit, both as an option and as an
7767 29. Added LIBS=-liconv to OS/Makefile-OSF1.
7769 30. Changed the default configuration ACL to relax the local part checking rule
7770 for addresses that are not in any local domains. For these addresses,
7771 slashes and pipe symbols are allowed within local parts, but the sequence
7772 /../ is explicitly forbidden.
7774 31. SPA server authentication was not clearing the challenge buffer before
7777 32. log_message in a "warn" ACL statement was writing to the reject log as
7778 well as to the main log, which contradicts the documentation and doesn't
7779 seem right (because no rejection is happening). So I have stopped it.
7781 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
7782 However, I am unable to do any testing of this.
7784 34. Fixed an infelicity in the appendfile transport. When checking directories
7785 for a mailbox, to see if any needed to be created, it was accidentally
7786 using path names with one or more superfluous leading slashes; tracing
7787 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
7789 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
7790 contents are added to the log line that is written for every discarded
7791 recipient. (Previously a log_message setting was ignored.)
7793 36. The ${quote: operator now quotes the string if it is empty.
7795 37. The install script runs exim in order to find its version number. If for
7796 some reason other than non-existence or emptiness, which it checks, it
7797 could not run './exim', it was installing it with an empty version number,
7798 i.e. as "exim-". This error state is now caught, and the installation is
7801 38. An argument was missing from the function that creates an error message
7802 when Exim fails to connect to the socket for saslauthd authentication.
7803 This could cause Exim to crash, or give a corrupted message.
7805 39. Added isip, isip4, and isip6 to ${if conditions.
7807 40. The ACL variables $acl_xx are now saved with the message, and can be
7808 accessed later in routers, transports, and filters.
7810 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
7811 strings in the file are not string-expanded.
7813 42. If a MAIL command specified a SIZE value that was too large to fit into an
7814 int variable, the check against message_size_limit failed. Such values are
7815 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
7816 day this will have to be increased, but I don't think I want to be around
7817 when emails are that large.
7824 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
7825 iconv() is not standard in FreeBSD.
7827 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
7828 IPv6 enabled. The observed symptom was a segmentation fault on return from
7829 the function os_common_find_running_interfaces() in src/os.c.
7831 3. In the check_special_case() function in daemon.c I had used "errno" as an
7832 argument name, which causes warnings on some systems. This was basically a
7833 typo, since it was named "eno" in the comments!
7835 4. The code that waits for the clock to tick (at a resolution of some fraction
7836 of a second) so as to ensure message-id uniqueness was always waiting for
7837 at least one whole tick, when it could have waited for less. [This is
7838 almost certainly not relevant at current processor speeds, where it is
7839 unlikely to ever wait at all. But we try to future-proof.]
7841 5. The function that sleeps for a time interval that includes fractions of a
7842 second contained a race. It did not block SIGALRM between setting the
7843 timer, and suspending (a couple of lines later). If the interval was short
7844 and the sigsuspend() was delayed until after it had expired, the suspension
7845 never ended. On busy systems this could lead to processes getting stuck for
7848 6. Some uncommon configurations may cause a lookup to happen in a queue runner
7849 process, before it forks any delivery processes. The open lookup caching
7850 mechanism meant that the open file or database connection was passed into
7851 the delivery process. The problem was that delivery processes always tidy
7852 up cached lookup data. This could cause a problem for the next delivery
7853 process started by the queue runner, because the external queue runner
7854 process does not know about the closure. So the next delivery process
7855 still has data in the lookup cache. In the case of a file lookup, there was
7856 no problem because closing a file descriptor in a subprocess doesn't affect
7857 the parent. However, if the lookup was caching a connection to a database,
7858 the connection was closed, and the second delivery process was likely to
7859 see errors such as "PGSQL: query failed: server closed the connection
7860 unexpectedly". The problem has been fixed by closing all cached lookups
7861 in a queue runner before running a delivery process.
7863 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
7864 seem to have the "const" qualifier which it has on other OS. I've
7867 8. Change 4.21/2 was too strict. It is only if there are two authenticators
7868 *of the same type* (client or server) with the same public name that an
7869 error should be diagnosed.
7871 9. When Exim looked up a host name for an IP address, but failed to find the
7872 original IP address when looking up the host name (a safety check), it
7873 output the message "<ip address> does not match any IP for NULL", which was
7874 confusing, to say the least. The bug was that the host name should have
7875 appeared instead of "NULL".
7877 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
7878 user that is built into the binary, and the -C or -D options is used, root
7879 privilege is dropped before the configuration file is read. In addition,
7880 logging is switched to stderr instead of the normal log files. If the
7881 configuration then re-defines the Exim user, the unprivileged environment
7882 is probably not what is expected, so Exim logs a panic warning message (but
7885 However, if deliver_drop_privilege is set, the unprivileged state may well
7886 be exactly what is intended, so the warning has been cut out in that case,
7887 and Exim is allowed to try to write to its normal log files.
7893 1. smtp_return_error_details was not giving details for temporary sender
7894 or receiver verification errors.
7896 2. Diagnose a configuration error if two authenticators have the same public
7899 3. Exim used not to create the message log file for a message until the first
7900 delivery attempt. This could be confusing when incoming messages were held
7901 for policy or load reasons. The message log file is now created at the time
7902 the message is received, and an initial "Received" line is written to it.
7904 4. The automatically generated man page for command line options had a minor
7905 bug that caused no ill effects; however, a more serious problem was that
7906 the procedure for building the man page automatically didn't always
7907 operate. Consequently, release 4.20 contains an out-of-date version. This
7908 shouldn't happen again.
7910 5. When building Exim with embedded Perl support, the script that builds the
7911 Makefile was calling 'perl' to find its compile-time parameters, ignoring
7912 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
7914 6. The freeze_tell option was not being used for messages that were frozen on
7915 arrival, either by an ACL or by local_scan().
7917 7. Added the smtp_incomplete_transaction log selector.
7919 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
7920 was accepting AUTH without a new EHLO.
7922 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
7923 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
7926 10. Logging of TCP/IP connections (when configured) now happens in the main
7927 daemon process instead of the child process, so that the TCP/IP connection
7928 count is more accurate (but it can never be perfect).
7930 11. The use of "drop" in a nested ACL was not being handled correctly in the
7931 outer ACL. Now, if condition failure induced by the nested "drop" causes
7932 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
7933 or "require"), the connection is dropped.
7935 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
7936 that yield "discard" can now be used with an "accept" or a "discard" verb,
7937 but an error is generated for any others (because I can't see a useful way
7938 to define what should happen).
7940 13. When an ACL is read dynamically from a file (or anywhere else), the lines
7941 are now processed in the same way as lines in the Exim configuration file.
7942 In particular, continuation lines are supported.
7944 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
7946 15. Added -ti meaning -t -i.
7948 16. Check for letters, digits, hyphens, and dots in the names of dnslist
7949 domains, and warn by logging if others are found.
7951 17. At least on BSD, alignment is not guaranteed for the array of ifreq's
7952 returned from GIFCONF when Exim is trying to find the list of interfaces on
7953 a host. The code in os.c has been modified to copy each ifreq to an aligned
7954 structure in all cases.
7956 Also, in some cases, the returned ifreq's were being copied to a 'struct
7957 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
7958 means the last couple of bytes of an IPv6 address could be chopped if the
7959 ifreq contained only a normal sockaddr (14 bytes storage).
7961 18. Named domain lists were not supported in the hosts_treat_as_local option.
7962 An entry such as +xxxx was not recognized, and was treated as a literal
7965 19. Ensure that header lines added by a DATA ACL are included in the reject log
7966 if the ACL subsequently rejects the message.
7968 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
7969 MD5 (which is deprecated).
7971 21. When testing a filter file using -bf, Exim was writing a message when it
7972 took the sender from a "From " line in the message, but it was not doing so
7973 when it took $return_path from a Return-Path: header line. It now does.
7975 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
7976 with a valid header line field name (a series of printing characters
7977 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
7979 23. Changed "disc" in the source to "disk" to conform to the documentation and
7980 the book and for uniformity.
7982 24. Ignore Sendmail's -Ooption=value command line item.
7984 25. When execve() failed while trying to run a command in a pipe transport,
7985 Exim was returning EX_UNAVAILABLE (69) from the subprocess. However, this
7986 could be confused with a return value of 69 from the command itself. This
7987 has been changed to 127, the value the shell returns if it is asked to run
7988 a non-existent command. The wording for the related log line suggests a
7989 non-existent command as the problem.
7991 26. If received_header_text expands to an empty string, do not add a Received:
7992 header line to the message. (Well, it adds a token one on the spool, but
7993 marks it "old" so that it doesn't get used or transmitted.)
7995 27. Installed eximstats 1.28 (addition of -nt option).
7997 28. There was no check for failure on the call to getsockname() in the daemon
7998 code. This can fail if there is a shortage of resources on the system, with
7999 ENOMEM, for example. A temporary error is now given on failure.
8001 29. Contrary to the C standard, it seems that in some environments, the
8002 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
8003 program. Exim now does this explicitly; it affects the formatting of
8004 timestamps using strftime().
8006 30. If exiqsumm was given junk data, it threw up some uninitialized variable
8007 complaints. I've now initialized all the variables, to avoid this.
8009 32. Header lines added by a system filter were not being "seen" during
8010 transport-time rewrites.
8012 33. The info_callback() function passed to OpenSSL is set up with type void
8013 (*)(SSL *, int, int), as described somewhere. However, when calling the
8014 function (actually a macro) that sets it up, the type void(*)() is
8015 expected. I've put in a cast to prevent warnings from picky compilers.
8017 34. If a DNS black list lookup found a CNAME record, but there were no A
8018 records associated with the domain it pointed at, Exim crashed.
8020 35. If a DNS black list lookup returned more than one A record, Exim ignored
8021 all but the first. It now scans all returned addresses if a particular IP
8022 value is being sought. In this situation, the contents of the
8023 $dnslist_value variable are a list of all the addresses, separated by a
8026 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
8027 to accept a host name and all its aliases if the forward lookup for any of
8028 them yielded the IP address of the incoming connection. Now it accepts only
8029 those names whose forward lookup yields the correct IP address. Any other
8030 names are discarded. This closes a loophole whereby a rogue DNS
8031 administrator could create reverse DNS records to break through a
8032 wildcarded host restriction in an ACL.
8034 37. If a user filter or a system filter that ran in a subprocess used any of
8035 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
8036 the wrong values were passed to the pipe command ($thisaddress had the
8037 value of $0, $0 had the value of $1, etc). This bug was introduced by
8038 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
8040 38. Improved the line breaking for long SMTP error messages from ACLs.
8041 Previously, if there was no break point between 40 and 75 characters, Exim
8042 left the rest of the message alone. Two changes have been made: (a) I've
8043 reduced the minimum length to 35 characters; (b) if it can't find a break
8044 point between 35 and 75 characters, it looks ahead and uses the first one
8045 that it finds. This may give the occasional overlong line, but at least the
8046 remaining text gets split now.
8048 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
8049 file descriptors might be low, and that setting 1000 would always raise it.
8050 It turns out that in some environments, the limit is already over 1000 and
8051 that lowering it causes trouble. So now Exim takes care not to decrease it.
8053 40. When delivering a message, the value of $return_path is set to $sender_
8054 address at the start of routing (routers may change the value). By an
8055 oversight, this default was not being set up when an address was tested by
8056 -bt or -bv, which affected the outcome if any router or filter referred to
8059 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
8060 log only when "message" or "log_message" are set. However, if one of the
8061 conditions was an address verification, or a call to a nested ACL, the
8062 messages generated by the underlying test were being passed through. This
8063 no longer happens. The underlying message is available in $acl_verify_
8064 message for both "message" and "log_message" expansions, so it can be
8065 passed through if needed.
8067 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
8068 new expansion $bh_ to give the encoded byte string without charset
8069 translation. Translation happens only if iconv() is available; HAVE_ICONV
8070 indicates this at build time. HEADERS_CHARSET gives the charset to
8071 translate to; headers_charset can change it in the configuration, and
8072 "headers charset" can change it in an individual filter file.
8074 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
8075 that creates RFC 2047 encoded "words" labels them as that charset instead
8076 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
8077 expansion operator; (ii) when Exim creates a From: line for a local
8078 message; (iii) when a header line is rewritten to include a "phrase" part.
8080 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
8081 buggy, causing it to skip the first lines of messages whose message ID
8082 ended in 'D'. This would not have bitten before Exim release 4.14, because
8083 message IDs were unlikely to end in 'D' before then. The effect was to have
8084 incorrect size information for certain domains.
8086 45. #include "config.h" was missing at the start of the crypt16.c module. This
8087 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
8090 46. If there was a timeout during a "random" callout check, Exim treated it as
8091 a failure of the random address, and carried on sending RSET and the real
8092 address. If the delay was just some slowness somewhere, the response to the
8093 original RCPT would be taken as a response to RSET and so on, causing
8094 mayhem of various kinds.
8096 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
8097 when I implemented it. It didn't allow for the fact that some option values
8098 may legitimately be negative (e.g. size_addition), and it didn't even do
8099 the right test for positive values.
8101 48. Domain names in DNS records are case-independent. Exim always looks them up
8102 in lower case. Some resolvers return domain names in exactly the case they
8103 appear in the zone file, that is, they may contain uppercase letters. Not
8104 all resolvers do this - some return always lower case. Exim was treating a
8105 change of case by a resolver as a change of domain, similar to a widening
8106 of a domain abbreviation. This triggered its re-routing code and so it was
8107 trying to route what was effectively the same domain again. This normally
8108 caused routing to fail (because the router wouldn't handle the domain
8109 twice). Now Exim checks for this case specially, and just changes the
8110 casing of the domain that it ultimately uses when it transmits the message
8113 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
8116 50. If a filter generated a file delivery with a non-absolute name (possible if
8117 no home directory exists for the router), the forbid_file option was not
8120 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
8121 the existing equality matching.
8123 52. Exim was using ints instead of ino_t variables in some places where it was
8124 dealing with inode numbers.
8126 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
8127 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
8128 called TMPDIR, and if it finds it is different, it changes its value.
8130 54. The smtp_printf() function is now made available to local_scan() so
8131 additional output lines can be written before returning. There is also an
8132 smtp_fflush() function to enable the detection of a dropped connection.
8133 The variables smtp_input and smtp_batched_input are exported to
8136 55. Changed the default runtime configuration: the message "Unknown user"
8137 has been removed from the ACL, and instead placed on the localuser router,
8138 using the cannot_route_message feature. This means that any verification
8139 failures that generate their own messages won't get overridden. Similarly,
8140 the "Unrouteable address" message that was in the ACL for unverifiable
8141 relay addresses has also been removed.
8143 56. Added hosts_avoid_esmtp to the smtp transport.
8145 57. The exicyclog script was not checking for the esoteric option
8146 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
8147 will work only if exicyclog is run under the appropriate euid.
8149 58. Following a discussion on the list, the rules by which Exim recognises line
8150 endings on incoming messages have been changed. The -dropcr and drop_cr
8151 options are now no-ops, retained only for backwards compatibility. The
8152 following line terminators are recognized: LF CRLF CR. However, special
8153 processing applies to CR:
8155 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
8156 nor a local message in the state where . is a terminator.
8158 (ii) If a bare CR is encountered in a header line, an extra space is added
8159 after the line terminator so as not to end the header. The reasoning
8160 behind this is that bare CRs in header lines are most likely either
8161 to be mistakes, or people trying to play silly games.
8163 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
8164 was being incorrectly given as 18 bytes larger than it should have been.
8165 This is a VOB (very old bug).
8167 60. This may never have affected anything current, but just in case it has:
8168 When the local host is found other than at the start of a list of hosts,
8169 the local host, those with the same MX, and any that follow, are discarded.
8170 When the list in question was part of a longer list of hosts, the following
8171 hosts (not currently being processed) were also being discarded. This no
8172 longer happens. I'm not sure if this situation could ever has previously
8175 61. Added the "/MX" feature to lists of hosts in the manualroute and query
8178 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
8179 header. This is something that is recommended in a new Internet Draft, and
8180 is something that is documented as being done by Sendmail. There are two
8181 possible values. For messages generated by the autoreply transport, Exim
8184 Auto-Submitted: auto-replied
8186 whereas for all other generated messages (e.g. bounces) it adds
8188 Auto-Submitted: auto-generated
8190 63. The "personal" condition in filters now includes a test for the
8191 Auto-Submitted: header. If it contains the string "auto-" the message it
8192 not considered personal.
8194 64. Added rcpt_include_affixes as a generic transport option.
8196 65. Added queue_only_override (default true).
8198 66. Added the syslog_duplication option.
8200 67. If what should have been the first header line of a message consisted of
8201 a space followed by a colon, Exim was mis-interpreting it as a header line.
8202 It isn't of course - it is syntactically invalid and should therefore be
8203 treated as the start of the message body. The misbehaviour could have
8204 caused a number of strange effects, including loss of data in subsequent
8205 header lines, and spool format errors.
8207 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
8208 client host had authenticated. This control can now be exercised by an ACL
8209 for more flexibility.
8211 69. By default, callouts do not happen when testing with -bh. There is now a
8212 variant, -bhc, which does actually run the callout code, including
8213 consulting and updating the callout cache.
8215 70. Added support for saslauthd authentication, courtesy of Alexander
8218 71. If statvfs() failed on the spool or log directories while checking their
8219 size for availability, Exim confusingly gave the error "space shortage".
8220 Furthermore, in debugging mode it crashed with a floating point exception.
8221 These checks are done if check_{spool,log}_{space,inodes} are set, and when
8222 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
8223 serious problem, Exim now writes to the main and panic logs when this
8224 happens, with details of the failure. It then refuses to accept the
8225 incoming message, giving the message "spool directory problem" or "log
8226 directory problem" with a 421 code for SMTP messages.
8228 72. When Exim is about to re-exec itself, it ensures that the file descriptors
8229 0, 1, and 2 exist, because some OS complain for execs without them (see
8230 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
8231 descriptors. However, the code omitted to check that the open succeeded,
8232 causing mysterious errors if for some reason the permissions on /dev/null
8233 got screwed. Now Exim writes a message to the main and panic logs, and
8234 bombs out if it can't open /dev/null.
8236 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
8237 interact so that it is all more flexible. It is supposed to remain
8238 backwards compatible. Also added extra_local_interfaces.
8240 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
8241 to bomb out with an assertion failure - to the client this appears as a
8242 connection drop. This problem occurs in the part of the code that was taken
8243 from the Samba project. Fortunately, the assertion is in a very simple
8244 function, so I have fixed this by reproducing the function inline in the
8245 one place where it is called, and arranging for authentication to fail
8246 instead of killing the process with assert().
8248 75. The SPA client code was not working when the server requested OEM rather
8249 than Unicode encoding.
8251 76. Added code to make require_files with a specific uid setting more usable in
8252 the case where statting the file as root fails - usually a non-root-mounted
8253 NFS file system. When this happens and the failure is EACCES, Exim now
8254 forks a subprocess and does the per-uid checking as the relevant uid.
8256 77. Added process_log_path.
8258 78. If log_file_path was not explicitly set, a setting of check_log_space or
8259 check_log_inodes was ignored.
8261 79. If a space check for the spool or log partitions fails, the incident is now
8262 logged. Of course, in the latter case the data may get lost...
8264 80. Added the %p formatting code to string_format() so that it can be used to
8265 print addresses in debug_print(). Adjusted all the address printing in the
8266 debugging in store.c to use %p rather than %d.
8268 81. There was a concern that a line of code in smtp_in.c could overflow a
8269 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
8270 initially expressed, the concern was not well-founded, because trailing
8271 spaces are removed early. However, if the trailing spaces were followed by
8272 a NULL, they did not get removed, so the overflow was possible. Two fixes
8275 (a) I re-wrote the offending code in a cleaner fashion.
8276 (b) If an incoming SMTP command contains a NULL character, it is rejected
8279 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
8280 runs initgroups(), so that if the Exim user is in any additional groups,
8281 they will be used during message reception.
8287 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / exim.git / blob