1 Change log file for Exim from version 4.21
2 ------------------------------------------
3 This document describes *changes* to previous versions, that might
4 affect Exim's operation, with an unchanged configuration file. For new
5 options, and new features, see the NewStuff file next to this ChangeLog.
11 JH/01 Remove code calling the customisable local_scan function, unless a new
12 definition "HAVE_LOCAL_SCAN=yes" is present in the Local/Makefile.
14 JH/02 Bug 1007: Avoid doing logging from signal-handlers, as that can result in
15 non-signal-safe functions being used.
17 JH/03 Bug 2269: When presented with a received message having a stupidly large
18 number of DKIM-Signature headers, disable DKIM verification to avoid
19 a resource-consumption attack. The limit is set at twenty.
21 JH/04 Add variables $arc_domains, $arc_oldest_pass for ARC verify. Fix the
22 report of oldest_pass in ${authres } in consequence, and separate out
23 some descriptions of reasons for verification fail.
25 JH/05 Bug 2273: Cutthrough delivery left a window where the received messsage
26 files in the spool were present and unlocked. A queue-runner could spot
27 them, resulting in a duplicate delivery. Fix that by doing the unlock
28 after the unlink. Investigation by Tim Stewart. Take the opportunity to
29 add more error-checking on spoolfile handling while that code is being
32 PP/01 Refuse to open a spool data file (*-D) if it's a symlink.
33 No known attacks, no CVE, this is defensive hardening.
35 JH/06 Bug 2275: The MIME ACL unlocked the received message files early, and
36 a queue-runner could start a delivery while other operations were ongoing.
37 Cutthrough delivery was a common victim, resulting in duplicate delivery.
38 Found and investigated by Tim Stewart. Fix by using the open message data
39 file handle rather than opening another, and not locally closing it (which
40 releases a lock) for that case, while creating the temporary .eml format
41 file for the MIME ACL. Also applies to "regex" and "spam" ACL conditions.
43 JH/07 Bug 177: Make a random-recipient callout success visible in ACL, by setting
44 $sender_verify_failure/$recipient_verify_failure to "random".
46 JH/08 When generating a selfsigned cert, use serial number 1 since zero is not
49 JH/09 Bug 2274: Fix logging of cmdline args when starting in an unlinked cwd.
50 Previously this would segfault.
52 JH/10 Fix ARC signing for case when DKIM signing failed. Previously this would
55 JH/11 Bug 2264: Exim now only follows CNAME chains one step by default. We'd
56 like zero, since the resolver should be doing this for us, But we need one
57 as a CNAME but no MX presence gets the CNAME returned; we need to check
58 that doesn't point to an MX to declare it "no MX returned" rather than
59 "error, loop". A new main option is added so the older capability of
60 following some limited number of chain links is maintained.
62 JH/12 Add client-ip info to non-pass iprev ${authres } lines.
64 JH/13 For receent Openssl versions (1.1 onward) use modern generic protocol
65 methods. These should support TLS 1.3; they arrived with TLS 1.3 and the
66 now-deprecated earlier definitions used only specified the range up to TLS
67 1.2 (in the older-version library docs).
69 JH/14 Bug 2284: Fix DKIM signing for body lines starting with a pair of dots.
71 JH/15 Rework TLS client-side context management. Stop using a global, and
72 explicitly pass a context around. This enables future use of TLS for
73 connections to service-daemons (eg. malware scanning) while a client smtp
74 connection is using TLS; with cutthrough connections this is quite likely.
76 JH/16 Fix ARC verification to do AS checks in reverse order.
78 JH/17 Support a "tls" option on the ${readsocket } expansion item.
80 JH/18 Bug 2287: Fix the protocol name (eg utf8esmtp) for multiple messages
81 using the SMTPUTF8 option on their MAIL FROM commands, in one connection.
82 Previously the "utf8" would be re-prepended for every additional message.
84 JH/19 Reject MAIL FROM commands with SMTPUTF8 when the facility was not advertised.
85 Previously thery were accepted, resulting in issues when attempting to
86 forward messages to a non-supporting MTA.
88 PP/02 Let -n work with printing macros too, not just options.
90 JH/20 Bug 2296: Fix cutthrough for >1 address redirection. Previously only
91 one parent address was copied, and bogus data was used at delivery-logging
92 time. Either a crash (after delivery) or bogus log data could result.
93 Discovery and analysis by Tim Stewart.
95 PP/03 Make ${utf8clean:} expansion operator detect incomplete final character.
96 Previously if the string ended mid-character, we did not insert the
97 promised '?' replacement.
99 PP/04 Documentation: current string operators work on bytes, not codepoints.
101 JH/21 Change as many as possible of the global flags into one-bit bitfields; these
102 should pack well giving a smaller memory footprint so better caching and
103 therefore performance. Group the declarations where this can't be done so
104 that the byte-sized flag variables are not interspersed among pointer
105 variables, giving a better chance of good packing by the compiler.
107 JH/22 Bug 1896: Fix the envelope from for DMARC forensic reports to be possibly
108 non-null, to avoid issues with sites running BATV. Previously reports were
109 sent with an empty envelope sender so looked like bounces.
115 GF/01 DEFER rather than ERROR on redis cluster MOVED response.
116 When redis_servers is set to a list of > 1 element, and the Redis servers
117 in that list are in cluster configuration, convert the REDIS_REPLY_ERROR
118 case of MOVED into a DEFER case instead, thus moving the query onto the
119 next server in the list. For a cluster of N elements, all N servers must
120 be defined in redis_servers.
122 GF/02 Catch and remove uninitialized value warning in exiqsumm
123 Check for existence of @ARGV before looking at $ARGV[0]
125 JH/01 Replace the store_release() internal interface with store_newblock(),
126 which internalises the check required to safely use the old one, plus
127 the allocate and data copy operations duplicated in both (!) of the
128 extant use locations.
130 JH/02 Disallow '/' characters in queue names specified for the "queue=" ACL
131 modifier. This matches the restriction on the commandline.
133 JH/03 Fix pgsql lookup for multiple result-tuples with a single column.
134 Previously only the last row was returned.
136 JH/04 Bug 2217: Tighten up the parsing of DKIM signature headers. Previously
137 we assumed that tags in the header were well-formed, and parsed the
138 element content after inspecting only the first char of the tag.
139 Assumptions at that stage could crash the receive process on malformed
142 JH/05 Bug 2215: Fix crash associated with dnsdb lookup done from DKIM ACL.
143 While running the DKIM ACL we operate on the Permanent memory pool so that
144 variables created with "set" persist to the DATA ACL. Also (at any time)
145 DNS lookups that fail create cache records using the Permanent pool. But
146 expansions release any allocations made on the current pool - so a dnsdb
147 lookup expansion done in the DKIM ACL releases the memory used for the
148 DNS negative-cache, and bad things result. Solution is to switch to the
149 Main pool for expansions.
150 While we're in that code, add checks on the DNS cache during store_reset,
151 active in the testsuite.
152 Problem spotted, and debugging aided, by Wolfgang Breyha.
154 JH/06 Fix issue with continued-connections when the DNS shifts unreliably.
155 When none of the hosts presented to a transport match an already-open
156 connection, close it and proceed with the list. Previously we would
157 queue the message. Spotted by Lena with Yahoo, probably involving
160 JH/07 Bug 2214: Fix SMTP responses resulting from non-accept result of MIME ACL.
161 Previously a spurious "250 OK id=" response was appended to the proper
164 JH/08 The "support for" informational output now, which built with Content
165 Scanning support, has a line for the malware scanner interfaces compiled
166 in. Interface can be individually included or not at build time.
168 JH/09 The "aveserver", "kavdaemon" and "mksd" interfaces are now not included
169 by the template makefile "src/EDITME". The "STREAM" support for an older
170 ClamAV interface method is removed.
172 JH/10 Bug 2223: Fix mysql lookup returns for the no-data case (when the number of
173 rows affected is given instead).
175 JH/11 The runtime Berkeley DB library version is now additionally output by
176 "exim -d -bV". Previously only the compile-time version was shown.
178 JH/12 Bug 2230: Fix cutthrough routing for nonfirst messages in an initiating
179 SMTP connection. Previously, when one had more receipients than the
180 first, an abortive onward connection was made. Move to full support for
181 multiple onward connections in sequence, handling cutthrough connection
182 for all multi-message initiating connections.
184 JH/13 Bug 2229: Fix cutthrough routing for nonstandard port numbers defined by
185 routers. Previously, a multi-recipient message would fail to match the
186 onward-connection opened for the first recipient, and cause its closure.
188 JH/14 Bug 2174: A timeout on connect for a callout was also erroneously seen as
189 a timeout on read on a GnuTLS initiating connection, resulting in the
190 initiating connection being dropped. This mattered most when the callout
191 was marked defer_ok. Fix to keep the two timeout-detection methods
194 JH/15 Relax results from ACL control request to enable cutthrough, in
195 unsupported situations, from error to silently (except under debug)
196 ignoring. This covers use with PRDR, frozen messages, queue-only and
199 HS/01 Fix Buffer overflow in base64d() (CVE-2018-6789)
201 JH/16 Fix bug in DKIM verify: a buffer overflow could corrupt the malloc
202 metadata, resulting in a crash in free().
204 PP/01 Fix broken Heimdal GSSAPI authenticator integration.
205 Broken in f2ed27cf5, missing an equals sign for specified-initialisers.
206 Broken also in d185889f4, with init system revamp.
208 JH/17 Bug 2113: Fix conversation closedown with the Avast malware scanner.
209 Previously we abruptly closed the connection after reading a malware-
210 found indication; now we go on to read the "scan ok" response line,
213 JH/18 Bug 2239: Enforce non-usability of control=utf8_downconvert in the mail
214 ACL. Previously, a crash would result.
216 JH/19 Speed up macro lookups during configuration file read, by skipping non-
217 macro text after a replacement (previously it was only once per line) and
218 by skipping builtin macros when searching for an uppercase lead character.
220 JH/20 DANE support moved from Experimental to mainline. The Makefile control
221 for the build is renamed.
223 JH/21 Fix memory leak during multi-message connections using STARTTLS. A buffer
224 was allocated for every new TLS startup, meaning one per message. Fix
225 by only allocating once (OpenSSL) or freeing on TLS-close (GnuTLS).
227 JH/22 Bug 2236: When a DKIM verification result is overridden by ACL, DMARC
228 reported the original. Fix to report (as far as possible) the ACL
229 result replacing the original.
231 JH/23 Fix memory leak during multi-message connections using STARTTLS under
232 OpenSSL. Certificate information is loaded for every new TLS startup,
233 and the resources needed to be freed.
235 JH/24 Bug 2242: Fix exim_dbmbuild to permit directoryless filenames.
237 JH/25 Fix utf8_downconvert propagation through a redirect router. Previously it
240 JH/26 Bug 2253: For logging delivery lines under PRDR, append the overall
241 DATA response info to the (existing) per-recipient response info for
242 the "C=" log element. It can have useful tracking info from the
243 destination system. Patch from Simon Arlott.
245 JH/27 Bug 2251: Fix ldap lookups that return a single attribute having zero-
246 length value. Previously this would segfault.
248 HS/02 Support Avast multiline protoocol, this allows passing flags to
249 newer versions of the scanner.
251 JH/28 Ensure that variables possibly set during message acceptance are marked
252 dead before release of memory in the daemon loop. This stops complaints
253 about them when the debug_store option is enabled. Discovered specifically
254 for sender_rate_period, but applies to a whole set of variables.
255 Do the same for the queue-runner and queue-list loops, for variables set
256 from spool message files. Do the same for the SMTP per-message loop, for
257 certain variables indirectly set in ACL operations.
259 JH/29 Bug 2250: Fix a longstanding bug in heavily-pipelined SMTP input (such
260 as a multi-recipient message from a mailinglist manager). The coding had
261 an arbitrary cutoff number of characters while checking for more input;
262 enforced by writing a NUL into the buffer. This corrupted long / fast
263 input. The problem was exposed more widely when more pipelineing of SMTP
264 responses was introduced, and one Exim system was feeding another.
265 The symptom is log complaints of SMTP syntax error (NUL chars) on the
266 receiving system, and refused recipients seen by the sending system
267 (propating to people being dropped from mailing lists).
268 Discovered and pinpointed by David Carter.
270 JH/30 The (EXPERIMENTAL_DMARC) variable $dmarc_ar_header is withdrawn, being
271 replaced by the ${authresults } expansion.
273 JH/31 Bug 2257: Fix pipe transport to not use a socket-only syscall.
275 HS/03 Set a handler for SIGTERM and call exit(3) if running as PID 1. This
276 allows proper process termination in container environments.
278 JH/32 Bug 2258: Fix spool_wireformat in combination with LMTP transport.
279 Previously the "final dot" had a newline after it; ensure it is CR,LF.
281 JH/33 SPF: remove support for the "spf" ACL condition outcome values "err_temp"
282 and "err_perm", deprecated since 4.83 when the RFC-defined words
283 "temperror" and "permerror" were introduced.
285 JH/34 Re-introduce enforcement of no cutthrough delivery on transports having
286 transport-filters or DKIM-signing. The restriction was lost in the
287 consolidation of verify-callout and delivery SMTP handling.
288 Extend the restriction to also cover ARC-signing.
290 JH/35 Cutthrough: for a final-dot response timeout (and nonunderstood responses)
291 in defer=pass mode supply a 450 to the initiator. Previously the message
294 PP/02 DANE: add dane_require_tls_ciphers SMTP Transport option; if unset,
295 tls_require_ciphers is used as before.
297 HS/03 Malware Avast: Better match the Avast multiline protocol. Add
298 "pass_unscanned". Only tmpfails from the scanner are written to
299 the paniclog, as they may require admin intervention (permission
300 denied, license issues). Other scanner errors (like decompression
301 bombs) do not cause a paniclog entry.
303 JH/36 Fix reinitialisation of DKIM logging variable between messages.
304 Previously it was possible to log spurious information in receive log
307 JH/37 Bug 2255: Revert the disable of the OpenSSL session caching. This
308 triggered odd behaviour from Outlook Express clients.
310 PP/03 Add util/renew-opendmarc-tlds.sh script for safe renewal of public
313 JH/38 DKIM: accept Ed25519 pubkeys in SubjectPublicKeyInfo-wrapped form,
314 since the IETF WG has not yet settled on that versus the original
315 "bare" representation.
317 JH/39 Fix syslog logging for syslog_timestamp=no and log_selector +millisec.
318 Previously the millisecond value corrupted the output.
319 Fix also for syslog_pid=no and log_selector +pid, for which the pid
320 corrupted the output.
326 JH/01 Rework error string handling in TLS interface so that the caller in
327 more cases is responsible for logging. This permits library-sourced
328 string to be attached to addresses during delivery, and collapses
329 pairs of long lines into single ones.
331 PP/01 Allow PKG_CONFIG_PATH to be set in Local/Makefile and use it correctly
332 during configuration. Wildcards are allowed and expanded.
334 JH/02 Rework error string handling in DKIM to pass more info back to callers.
335 This permits better logging.
337 JH/03 Rework the transport continued-connection mechanism: when TLS is active,
338 do not close it down and have the child transport start it up again on
339 the passed-on TCP connection. Instead, proxy the child (and any
340 subsequent ones) for TLS via a unix-domain socket channel. Logging is
341 affected: the continued delivery log lines do not have any DNSSEC, TLS
342 Certificate or OCSP information. TLS cipher information is still logged.
344 JH/04 Shorten the log line for daemon startup by collapsing adjacent sets of
345 identical IP addresses on different listening ports. Will also affect
348 PP/02 Bug 2070: uClibc defines __GLIBC__ without providing glibc headers;
349 add noisy ifdef guards to special-case this sillyness.
350 Patch from Bernd Kuhls.
352 JH/05 Tighten up the checking in isip4 (et al): dotted-quad components larger
353 than 255 are no longer allowed.
355 JH/06 Default openssl_options to include +no_ticket, to reduce load on peers.
356 Disable the session-cache too, which might reduce our load. Since we
357 currrectly use a new context for every connection, both as server and
358 client, there is no benefit for these.
359 GnuTLS appears to not support tickets server-side by default (we don't
360 call gnutls_session_ticket_enable_server()) but client side is enabled
361 by default on recent versions (3.1.3 +) unless the PFS priority string
364 PP/03 Add $SOURCE_DATE_EPOCH support for reproducible builds, per spec at
365 <https://reproducible-builds.org/specs/source-date-epoch/>.
367 JH/07 Fix smtp transport use of limited max_rcpt under mua_wrapper. Previously
368 the check for any unsuccessful recipients did not notice the limit, and
369 erroneously found still-pending ones.
371 JH/08 Pipeline CHUNKING command and data together, on kernels that support
372 MSG_MORE. Only in-clear (not on TLS connections).
374 JH/09 Avoid using a temporary file during transport using dkim. Unless a
375 transport-filter is involved we can buffer the headers in memory for
376 creating the signature, and read the spool data file once for the
377 signature and again for transmission.
379 JH/10 Enable use of sendfile in Linux builds as default. It was disabled in
380 4.77 as the kernel support then wasn't solid, having issues in 64bit
381 mode. Now, it's been long enough. Add support for FreeBSD also.
383 JH/11 Bug 2104: Fix continued use of a transport connection with TLS. In the
384 case where the routing stage had gathered several addresses to send to
385 a host before calling the transport for the first, we previously failed
386 to close down TLS in the old transport process before passing the TCP
387 connection to the new process. The new one sent a STARTTLS command
388 which naturally failed, giving a failed delivery and bloating the retry
389 database. Investigation and fix prototype from Wolfgang Breyha.
391 JH/12 Fix check on SMTP command input synchronisation. Previously there were
392 false-negatives in the check that the sender had not preempted a response
393 or prompt from Exim (running as a server), due to that code's lack of
394 awareness of the SMTP input buffering.
396 PP/04 Add commandline_checks_require_admin option.
397 Exim drops privileges sanely, various checks such as -be aren't a
398 security problem, as long as you trust local users with access to their
399 own account. When invoked by services which pass untrusted data to
400 Exim, this might be an issue. Set this option in main configuration
401 AND make fixes to the calling application, such as using `--` to stop
404 JH/13 Do pipelining under TLS. Previously, although safe, no advantage was
405 taken. Now take care to pack both (client) MAIL,RCPT,DATA, and (server)
406 responses to those, into a single TLS record each way (this usually means
407 a single packet). As a side issue, smtp_enforce_sync now works on TLS
410 PP/05 OpenSSL/1.1: use DH_bits() for more accurate DH param sizes. This
411 affects you only if you're dancing at the edge of the param size limits.
412 If you are, and this message makes sense to you, then: raise the
413 configured limit or use OpenSSL 1.1. Nothing we can do for older
416 JH/14 For the "sock" variant of the malware scanner interface, accept an empty
417 cmdline element to get the documented default one. Previously it was
420 JH/15 Fix a crash in the smtp transport caused when two hosts in succession
421 are unsuable for non-message-specific reasons - eg. connection timeout,
422 banner-time rejection.
424 JH/16 Fix logging of delivery remote port, when specified by router, under
427 PP/06 Repair manualroute's ability to take options in any order, even if one
428 is the name of a transport.
431 HS/01 Cleanup, prevent repeated use of -p/-oMr (CVE-2017-1000369)
433 JH/17 Change the list-building routines interface to use the expanding-string
434 triplet model, for better allocation and copying behaviour.
436 JH/18 Prebuild the data-structure for "builtin" macros, for faster startup.
437 Previously it was constructed the first time a possibly-matching string
438 was met in the configuration file input during startup; now it is done
441 JH/19 Bug 2141: Use the full-complex API for Berkeley DB rather than the legacy-
442 compatible one, to avoid the (poorly documented) possibility of a config
443 file in the working directory redirecting the DB files, possibly correpting
444 some existing file. CVE-2017-10140 assigned for BDB.
446 JH/20 Bug 2147: Do not defer for a verify-with-callout-and-random which is not
447 cache-hot. Previously, although the result was properly cached, the
448 initial verify call returned a defer.
450 JH/21 Bug 2151: Avoid using SIZE on the MAIL for a callout verify, on any but
451 the main verify for receipient in uncached-mode.
453 JH/22 Retire historical build files to an "unsupported" subdir. These are
454 defined as "ones for which we have no current evidence of testing".
456 JH/23 DKIM: enforce the DNS pubkey record "h" permitted-hashes optional field,
457 if present. Previously it was ignored.
459 JH/24 Start using specified-initialisers in C structure init coding. This is
460 a C99 feature (it's 2017, so now considered safe).
462 JH/25 Use one-bit bitfields for flags in the "addr" data structure. Previously
463 if was a fixed-sized field and bitmask ops via macros; it is now more
466 PP/07 GitHub PR 56: Apply MariaDB build fix.
467 Patch provided by Jaroslav Škarvada.
469 PP/08 Bug 2161: Fix regression in sieve quoted-printable handling introduced
470 during Coverity cleanups [4.87 JH/47]
471 Diagnosis and fix provided by Michael Fischer v. Mollard.
473 JH/26 Fix DKIM bug: when the pseudoheader generated for signing was exactly
474 the right size to place the terminating semicolon on its own folded
475 line, the header hash was calculated to an incorrect value thanks to
476 the (relaxed) space the fold became.
478 HS/02 Fix Bug 2130: large writes from the transport subprocess where chunked
479 and confused the parent.
481 JH/27 Fix SOCKS bug: an unitialized pointer was deref'd by the transport process
482 which could crash as a result. This could lead to undeliverable messages.
484 JH/28 Logging: "next input sent too soon" now shows where input was truncated
487 JH/29 Fix queue_run_in_order to ignore the PID portion of the message ID. This
488 matters on fast-turnover and PID-randomising systems, which were getting
489 out-of-order delivery.
491 JH/30 Fix a logging bug on aarch64: an unsafe routine was previously used for
492 a possibly-overlapping copy. The symptom was that "Remote host closed
493 connection in response to HELO" was logged instead of the actual 4xx
496 JH/31 Fix CHUNKING code to properly flush the unwanted chunk after an error.
497 Previously only that bufferd was discarded, resulting in SYMTP command
500 JH/32 DKIM: when a message has multiple signatures matching an identity given
501 in dkim_verify_signers, run the dkim acl once for each. Previously only
502 one run was done. Bug 2189.
504 JH/33 Downgrade an unfound-list name (usually a typo in the config file) from
505 "panic the current process" to "deliberately defer". The panic log is
506 still written with the problem list name; the mail and reject logs now
507 get a temp-reject line for the message that was being handled, saying
508 something like "domains check lookup or other defer". The SMTP 451
509 message is still "Temporary local problem".
511 JH/34 Bug 2199: Fix a use-after-free while reading smtp input for header lines.
512 A crafted sequence of BDAT commands could result in in-use memory beeing
513 freed. CVE-2017-16943.
515 HS/03 Bug 2201: Fix checking for leading-dot on a line during headers reading
516 from SMTP input. Previously it was always done; now only done for DATA
517 and not BDAT commands. CVE-2017-16944.
519 JH/35 Bug 2201: Flush received data in BDAT mode after detecting an error fatal
520 to the message (such as an overlong header line). Previously this was
521 not done and we did not exit BDAT mode. Followon from the previous item
522 though a different problem.
528 JH/01 Bug 1922: Support IDNA2008. This has slightly different conversion rules
529 than -2003 did; needs libidn2 in addition to libidn.
531 JH/02 The path option on a pipe transport is now expanded before use.
533 PP/01 GitHub PR 50: Do not call ldap_start_tls_s on ldapi:// connections.
534 Patch provided by "Björn", documentation fix added too.
536 JH/03 Bug 2003: fix Proxy Protocol v2 handling: the address size field was
537 missing a wire-to-host endian conversion.
539 JH/04 Bug 2004: fix CHUNKING in non-PIPELINEING mode. Chunk data following
540 close after a BDAT command line could be taken as a following command,
541 giving a synch failure. Fix by only checking for synch immediately
542 before acknowledging the chunk.
544 PP/02 GitHub PR 52: many spelling fixes, which include fixing parsing of
545 no_require_dnssec option and creation of _HAVE_TRANSPORT_APPEND_MAILDIR
546 macro. Patches provided by Josh Soref.
548 JH/05 Have the EHLO response advertise VRFY, if there is a vrfy ACL defined.
549 Previously we did not; the RFC seems ambiguous and VRFY is not listed
550 by IANA as a service extension. However, John Klensin suggests that we
553 JH/06 Bug 2017: Fix DKIM verification in -bh test mode. The data feed into
554 the dkim code may be unix-mode line endings rather than smtp wire-format
555 CRLF, so prepend a CR to any bare LF.
557 JH/07 Rationalise the coding for callout smtp conversations and transport ones.
558 As a side-benfit, callouts can now use PIPELINING hence fewer round-trips.
560 JH/08 Bug 2016: Fix DKIM verification vs. CHUNKING. Any BDAT commands after
561 the first were themselves being wrongly included in the feed into dkim
562 processing; with most chunk sizes in use this resulted in an incorrect
563 body hash calculated value.
565 JH/09 Bug 2014: permit inclusion of a DKIM-Signature header in a received
566 DKIM signature block, for verification. Although advised against by
567 standards it is specifically not ruled illegal.
569 JH/10 Bug 2025: Fix reception of (quoted) local-parts with embedded spaces.
571 JH/11 Bug 2029: Fix crash in DKIM verification when a message signature block is
572 missing a body hash (the bh= tag).
574 JH/12 Bug 2018: Re-order Proxy Protocol startup versus TLS-on-connect startup.
575 It seems that HAProxy sends the Proxy Protocol information in clear and
576 only then does a TLS startup, so do the same.
578 JH/13 Bug 2027: Avoid attempting to use TCP Fast Open for non-transport client
579 TCP connections (such as for Spamd) unless the daemon successfully set
580 Fast Open mode on its listening sockets. This fixes breakage seen on
581 too-old kernels or those not configured for Fast Open, at the cost of
582 requiring both directions being enabled for TFO, and TFO never being used
583 by non-daemon-related Exim processes.
585 JH/14 Bug 2000: Reject messages recieved with CHUNKING but with malformed line
586 endings, at least on the first header line. Try to canonify any that get
587 past that check, despite the cost.
589 JH/15 Angle-bracket nesting (an error inserted by broken sendmails) levels are
590 now limited to an arbitrary five deep, while parsing addresses with the
591 strip_excess_angle_brackets option enabled.
593 PP/03 Bug 2018: For Proxy Protocol and TLS-on-connect, do not over-read and
594 instead leave the unprompted TLS handshake in socket buffer for the
595 TLS library to consume.
597 PP/04 Bug 2018: Also handle Proxy Protocol v2 safely.
599 PP/05 FreeBSD compat: handle that Ports no longer create /usr/bin/perl
601 JH/16 Drop variables when they go out of scope. Memory management drops a whole
602 region in one operation, for speed, and this leaves assigned pointers
603 dangling. Add checks run only under the testsuite which checks all
604 variables at a store-reset and panics on a dangling pointer; add code
605 explicitly nulling out all the variables discovered. Fixes one known
606 bug: a transport crash, where a dangling pointer for $sending_ip_address
607 originally assigned in a verify callout, is re-used.
609 PP/06 Drop '.' from @INC in various Perl scripts.
611 PP/07 Switch FreeBSD iconv to always use the base-system libc functions.
613 PP/08 Reduce a number of compilation warnings under clang; building with
614 CC=clang CFLAGS+=-Wno-dangling-else -Wno-logical-op-parentheses
615 should be warning-free.
617 JH/17 Fix inbound CHUNKING when DKIM disabled at runtime.
619 HS/01 Fix portability problems introduced by PP/08 for platforms where
620 realloc(NULL) is not equivalent to malloc() [SunOS et al].
622 HS/02 Bug 1974: Fix missing line terminator on the last received BDAT
623 chunk. This allows us to accept broken chunked messages. We need a more
624 general solution here.
626 PP/09 Wrote util/chunking_fixqueue_finalnewlines.pl to help recover
627 already-broken messages in the queue.
629 JH/18 Bug 2061: Fix ${extract } corrupting an enclosing ${reduce } $value.
631 JH/19 Fix reference counting bug in routing-generated-address tracking.
637 JH/01 Use SIZE on MAIL FROM in a cutthrough connection, if the destination
638 supports it and a size is available (ie. the sending peer gave us one).
640 JH/02 The obsolete acl condition "demime" is removed (finally, after ten
641 years of being deprecated). The replacements are the ACLs
642 acl_smtp_mime and acl_not_smtp_mime.
644 JH/03 Upgrade security requirements imposed for hosts_try_dane: previously
645 a downgraded non-dane trust-anchor for the TLS connection (CA-style)
646 or even an in-clear connection were permitted. Now, if the host lookup
647 was dnssec and dane was requested then the host is only used if the
648 TLSA lookup succeeds and is dnssec. Further hosts (eg. lower priority
649 MXs) will be tried (for hosts_try_dane though not for hosts_require_dane)
650 if one fails this test.
651 This means that a poorly-configured remote DNS will make it incommunicado;
652 but it protects against a DNS-interception attack on it.
654 JH/04 Bug 1810: make continued-use of an open smtp transport connection
655 non-noisy when a race steals the message being considered.
657 JH/05 If main configuration option tls_certificate is unset, generate a
658 self-signed certificate for inbound TLS connections.
660 JH/06 Bug 165: hide more cases of password exposure - this time in expansions
661 in rewrites and routers.
663 JH/07 Retire gnutls_require_mac et.al. These were nonfunctional since 4.80
664 and logged a warning sing 4.83; now they are a configuration file error.
666 JH/08 Bug 1836: Fix crash in VRFY handling when handed an unqualified name
667 (lacking @domain). Apply the same qualification processing as RCPT.
669 JH/09 Bug 1804: Avoid writing msglog files when in -bh or -bhc mode.
671 JH/10 Support ${sha256:} applied to a string (as well as the previous
674 JH/11 Cutthrough: avoid using the callout hints db on a verify callout when
675 a cutthrough deliver is pending, as we always want to make a connection.
676 This also avoids re-routing the message when later placing the cutthrough
677 connection after a verify cache hit.
678 Do not update it with the verify result either.
680 JH/12 Cutthrough: disable when verify option success_on_redirect is used, and
681 when routing results in more than one destination address.
683 JH/13 Cutthrough: expand transport dkim_domain option when testing for dkim
684 signing (which inhibits the cutthrough capability). Previously only
685 the presence of an option was tested; now an expansion evaluating as
686 empty is permissible (obviously it should depend only on data available
687 when the cutthrough connection is made).
689 JH/14 Fix logging of errors under PIPELINING. Previously the log line giving
690 the relevant preceding SMTP command did not note the pipelining mode.
692 JH/15 Fix counting of empty lines in $body_linecount and $message_linecount.
693 Previously they were not counted.
695 JH/16 DANE: treat a TLSA lookup response having all non-TLSA RRs, the same
696 as one having no matching records. Previously we deferred the message
697 that needed the lookup.
699 JH/17 Fakereject: previously logged as a normal message arrival "<="; now
700 distinguished as "(=".
702 JH/18 Bug 1867: make the fail_defer_domains option on a dnslookup router work
703 for missing MX records. Previously it only worked for missing A records.
705 JH/19 Bug 1850: support Radius libraries that return REJECT_RC.
707 JH/20 Bug 1872: Ensure that acl_smtp_notquit is run when the connection drops
708 after the data-go-ahead and data-ack. Patch from Jason Betts.
710 JH/21 Bug 1846: Send DMARC forensic reports for reject and quarantine results,
711 even for a "none" policy. Patch from Tony Meyer.
713 JH/22 Fix continued use of a connection for further deliveries. If a port was
714 specified by a router, it must also match for the delivery to be
717 JH/23 Bug 1874: fix continued use of a connection for further deliveries.
718 When one of the recipients of a message was unsuitable for the connection
719 (has no matching addresses), we lost track of needing to mark it
720 deferred. As a result mail would be lost.
722 JH/24 Bug 1832: Log EHLO response on getting conn-close response for HELO.
724 JH/25 Decoding ACL controls is now done using a binary search; the source code
725 takes up less space and should be simpler to maintain. Merge the ACL
726 condition decode tables also, with similar effect.
728 JH/26 Fix problem with one_time used on a redirect router which returned the
729 parent address unchanged. A retry would see the parent address marked as
730 delivered, so not attempt the (identical) child. As a result mail would
733 JH/27 Fix a possible security hole, wherein a process operating with the Exim
734 UID can gain a root shell. Credit to http://www.halfdog.net/ for
735 discovery and writeup. Ubuntu bug 1580454; no bug raised against Exim
738 JH/28 Enable {spool,log} filesystem space and inode checks as default.
739 Main config options check_{log,spool}_{inodes,space} are now
740 100 inodes, 10MB unless set otherwise in the configuration.
742 JH/29 Fix the connection_reject log selector to apply to the connect ACL.
743 Previously it only applied to the main-section connection policy
746 JH/30 Bug 1897: fix callouts connection fallback from TLS to cleartext.
748 PP/01 Changed default Diffie-Hellman parameters to be Exim-specific, created
749 by me. Added RFC7919 DH primes as an alternative.
751 PP/02 Unbreak build via pkg-config with new hash support when crypto headers
752 are not in the system include path.
754 JH/31 Fix longstanding bug with aborted TLS server connection handling. Under
755 GnuTLS, when a session startup failed (eg because the client disconnected)
756 Exim did stdio operations after fclose. This was exposed by a recent
757 change which nulled out the file handle after the fclose.
759 JH/32 Bug 1909: Fix OCSP proof verification for cases where the proof is
760 signed directly by the cert-signing cert, rather than an intermediate
761 OCSP-signing cert. This is the model used by LetsEncrypt.
763 JH/33 Bug 1914: Ensure socket is nonblocking before draining after SMTP QUIT.
765 HS/01 Fix leak in verify callout under GnuTLS, about 3MB per recipient on
766 an incoming connection.
768 HS/02 Bug 1802: Do not half-close the connection after sending a request
771 HS/03 Use "auto" as the default EC curve parameter. For OpenSSL < 1.0.2
772 fallback to "prime256v1".
774 JH/34 SECURITY: Use proper copy of DATA command in error message.
775 Could leak key material. Remotely exploitable. CVE-2016-9963.
781 JH/01 Bug 1664: Disable OCSP for GnuTLS library versions at/before 3.3.16
782 and 3.4.4 - once the server is enabled to respond to an OCSP request
783 it does even when not requested, resulting in a stapling non-aware
784 client dropping the TLS connection.
786 TF/01 Code cleanup: Overhaul the debug_selector and log_selector machinery to
787 support variable-length bit vectors. No functional change.
789 TF/02 Improve the consistency of logging incoming and outgoing interfaces.
790 The I= interface field on outgoing lines is now after the H= remote
791 host field, same as incoming lines. There is a separate
792 outgoing_interface log selector which allows you to disable the
795 JH/02 Bug 728: Close logfiles after a daemon-process "exceptional" log write.
796 If not running log_selector +smtp_connection the mainlog would be held
797 open indefinitely after a "too many connections" event, including to a
798 deleted file after a log rotate. Leave the per net connection logging
799 leaving it open for efficiency as that will be quickly detected by the
800 check on the next write.
802 HS/01 Bug 1671: Fix post transport crash.
803 Processing the wait-<transport> messages could crash the delivery
804 process if the message IDs didn't exist for some reason. When
805 using 'split_spool_directory=yes' the construction of the spool
806 file name failed already, exposing the same netto behaviour.
808 JH/03 Bug 425: Capture substrings in $regex1, $regex2 etc from regex &
809 mime_regex ACL conditions.
811 JH/04 Bug 1686: When compiled with EXPERIMENTAL_DSN_INFO: Add extra information
812 to DSN fail messages (bounces): remote IP, remote greeting, remote response
813 to HELO, local diagnostic string.
815 JH/05 Downgrade message for a TLS-certificate-based authentication fail from
816 log line to debug. Even when configured with a tls authenticator many
817 client connections are expected to not authenticate in this way, so
818 an authenticate fail is not an error.
820 HS/02 Add the Exim version string to the process info. This way exiwhat
821 gives some more detail about the running daemon.
823 JH/06 Bug 1395: time-limit caching of DNS lookups, to the TTL value. This may
824 matter for fast-change records such as DNSBLs.
826 JH/07 Bug 1678: Always record an interface option value, if set, as part of a
827 retry record, even if constant. There may be multiple transports with
828 different interface settings and the retry behaviour needs to be kept
831 JH/08 Bug 1586: exiqgrep now refuses to run if there are unexpected arguments.
833 JH/09 Bug 1700: ignore space & tab embedded in base64 during decode.
835 JH/10 Bug 840: fix log_defer_output option of pipe transport
837 JH/11 Bug 830: use same host for all RCPTS of a message, even under
838 hosts_randomize. This matters a lot when combined with mua_wrapper.
840 JH/12 Bug 1706: percent and underbar characters are no longer escaped by the
841 ${quote_pgsql:<string>} operator.
843 JH/13 Bug 1708: avoid misaligned access in cached lookup.
845 JH/14 Change header file name for freeradius-client. Relevant if compiling
846 with Radius support; from the Gentoo tree and checked under Fedora.
848 JH/15 Bug 1712: Introduce $prdr_requested flag variable
850 JH/16 Bug 1714: Permit an empty string as expansion result for transport
851 option transport_filter, meaning no filtering.
853 JH/17 Bug 1713: Fix non-PDKIM_DEBUG build. Patch from Jasen Betts.
855 JH/18 Bug 1709: When built with TLS support, the tls_advertise_hosts option now
856 defaults to "*" (all hosts). The variable is now available when not built
857 with TLS, default unset, mainly to enable keeping the testsuite sane.
858 If a server certificate is not supplied (via tls_certificate) an error is
859 logged, and clients will find TLS connections fail on startup. Presumably
860 they will retry in-clear.
861 Packagers of Exim are strongly encouraged to create a server certificate
862 at installation time.
864 HS/03 Add -bP config_file as a synonym for -bP configure_file, for consistency
865 with the $config_file variable.
867 JH/19 Two additional event types: msg:rcpt:defer and msg:rcpt:host:defer. Both
868 in transport context, after the attempt, and per-recipient. The latter type
869 is per host attempted. The event data is the error message, and the errno
870 information encodes the lookup type (A vs. MX) used for the (first) host,
871 and the trailing two digits of the smtp 4xx response.
873 GF/01 Bug 1715: Fix for race condition in exicyclog, where exim could attempt
874 to write to mainlog (or rejectlog, paniclog) in the window between file
875 creation and permissions/ownership being changed. Particularly affects
876 installations where exicyclog is run as root, rather than exim user;
877 result is that the running daemon panics and dies.
879 JH/20 Bug 1701: For MySQL lookups, support MySQL config file option group names.
881 JH/21 Bug 1720: Add support for priority groups and weighted-random proxy
882 selection for the EXPERIMENTAL_SOCKS feature, via new per-proxy options
883 "pri" and "weight". Note that the previous implicit priority given by the
884 list order is no longer honoured.
886 JH/22 Bugs 963, 1721: Fix some corner cases in message body canonicalization
889 JH/23 Move SOCKS5 support from Experimental to mainline, enabled for a build
890 by defining SUPPORT_SOCKS.
892 JH/26 Move PROXY support from Experimental to mainline, enabled for a build
893 by defining SUPPORT_PROXY. Note that the proxy_required_hosts option
894 is renamed to hosts_proxy, and the proxy_{host,target}_{address,port}.
895 variables are renamed to proxy_{local,external}_{address,port}.
897 JH/27 Move Internationalisation support from Experimental to mainline, enabled
898 for a build by defining SUPPORT_I18N
900 JH/28 Bug 1745: Fix redis lookups to handle (quoted) spaces embedded in parts
901 of the query string, and make ${quote_redis:} do that quoting.
903 JH/29 Move Events support from Experimental to mainline, enabled by default
904 and removable for a build by defining DISABLE_EVENT.
906 JH/30 Updated DANE implementation code to current from Viktor Dukhovni.
908 JH/31 Fix bug with hosts_connection_nolog and named-lists which were wrongly
909 cached by the daemon.
911 JH/32 Move Redis support from Experimental to mainline, enabled for a build
912 by defining LOOKUP_REDIS. The libhiredis library is required.
914 JH/33 Bug 1748: Permit ACL dnslists= condition in non-smtp ACLs if explicit
915 keys are given for lookup.
917 JH/34 Bug 1192: replace the embedded copy of PolarSSL RSA routines in the DKIM
918 support, by using OpenSSL or GnuTLS library ones. This means DKIM is
919 only supported when built with TLS support. The PolarSSL SHA routines
920 are still used when the TLS library is too old for convenient support.
922 JH/35 Require SINGLE_DH_USE by default in OpenSSL (main config option
923 openssl_options), for security. OpenSSL forces this from version 1.1.0
924 server-side so match that on older versions.
926 JH/36 Bug 1778: longstanding bug in memory use by the ${run } expansion: A fresh
927 allocation for $value could be released as the expansion processing
928 concluded, but leaving the global pointer active for it.
930 JH/37 Bug 1769: Permit a VRFY ACL to override the default 252 response,
931 and to use the domains and local_parts ACL conditions.
933 JH/38 Fix cutthrough bug with body lines having a single dot. The dot was
934 incorrectly not doubled on cutthrough transmission, hence seen as a
935 body-termination at the receiving system - resulting in truncated mails.
936 Commonly the sender saw a TCP-level error, and retransmitted the message
937 via the normal store-and-forward channel. This could result in duplicates
938 received - but deduplicating mailstores were liable to retain only the
939 initial truncated version.
941 JH/39 Bug 1781: Fix use of DKIM private-keys having trailing '=' in the base-64.
943 JH/40 Fix crash in queryprogram router when compiled with EXPERIMENTAL_SRS.
945 JH/41 Bug 1792: Fix selection of headers to sign for DKIM: bottom-up. While
946 we're in there, support oversigning also; bug 1309.
948 JH/42 Bug 1796: Fix error logged on a malware scanner connection failure.
950 HS/04 Add support for keep_environment and add_environment options.
952 JH/43 Tidy coding issues detected by gcc --fsanitize=undefined. Some remain;
953 either intentional arithmetic overflow during PRNG, or testing config-
956 JH/44 Bug 1800: The combination of a -bhc commandline option and cutthrough
957 delivery resulted in actual delivery. Cancel cutthrough before DATA
960 JH/45 Fix cutthrough, when connection not opened by verify and target hard-
961 rejects a recipient: pass the reject to the originator.
963 JH/46 Multiple issues raised by Coverity. Some were obvious or plausible bugs.
964 Many were false-positives and ignorable, but it's worth fixing the
967 JH/47 Fix build on HP-UX and older Solaris, which need (un)setenv now also
968 for the new environment-manipulation done at startup. Move the routines
969 from being local to tls.c to being global via the os.c file.
971 JH/48 Bug 1807: Fix ${extract } for the numeric/3-string case. While preparsing
972 an extract embedded as result-arg for a map, the first arg for extract
973 is unavailable so we cannot tell if this is a numbered or keyed
974 extraction. Accept either.
980 JH/01 Bug 1545: The smtp transport option "retry_include_ip_address" is now
983 JH/02 The smtp transport option "multi_domain" is now expanded.
985 JH/03 The smtp transport now requests PRDR by default, if the server offers
988 JH/04 Certificate name checking on server certificates, when exim is a client,
989 is now done by default. The transport option tls_verify_cert_hostnames
990 can be used to disable this per-host. The build option
991 EXPERIMENTAL_CERTNAMES is withdrawn.
993 JH/05 The value of the tls_verify_certificates smtp transport and main options
994 default to the word "system" to access the system default CA bundle.
995 For GnuTLS, only version 3.0.20 or later.
997 JH/06 Verification of the server certificate for a TLS connection is now tried
998 (but not required) by default. The verification status is now logged by
999 default, for both outbound TLS and client-certificate supplying inbound
1002 JH/07 Changed the default rfc1413 lookup settings to disable calls. Few
1005 JH/08 The EXPERIMENTAL_DSN compile option is no longer needed; all Delivery
1006 Status Notification (bounce) messages are now MIME format per RFC 3464.
1007 Support for RFC 3461 DSN options NOTIFY,ENVID,RET,ORCPT can be advertised
1008 under the control of the dsn_advertise_hosts option, and routers may
1009 have a dsn_lasthop option.
1011 JH/09 A timeout of 2 minutes is now applied to all malware scanner types by
1012 default, modifiable by a malware= option. The list separator for
1013 the options can now be changed in the usual way. Bug 68.
1015 JH/10 The smtp_receive_timeout main option is now expanded before use.
1017 JH/11 The incoming_interface log option now also enables logging of the
1018 local interface on delivery outgoing connections.
1020 JH/12 The cutthrough-routing facility now supports multi-recipient mails,
1021 if the interface and destination host and port all match.
1023 JH/13 Bug 344: The verify = reverse_host_lookup ACL condition now accepts a
1026 JH/14 Bug 1573: The spam= ACL condition now additionally supports Rspamd.
1027 Patch from Andrew Lewis.
1029 JH/15 Bug 670: The spamd_address main option (for the spam= ACL condition)
1030 now supports optional time-restrictions, weighting, and priority
1031 modifiers per server. Patch originally by <rommer@active.by>.
1033 JH/16 The spamd_address main option now supports a mixed list of local
1034 and remote servers. Remote servers can be IPv6 addresses, and
1035 specify a port-range.
1037 JH/17 Bug 68: The spamd_address main option now supports an optional
1038 timeout value per server.
1040 JH/18 Bug 1581: Router and transport options headers_add/remove can
1041 now have the list separator specified.
1043 JH/19 Bug 392: spamd_address, and clamd av_scanner, now support retry
1046 JH/20 Bug 1571: Ensure that $tls_in_peerdn is set, when verification fails
1049 JH/21 Support for the A6 type of dns record is withdrawn.
1051 JH/22 Bug 608: The result of a QUIT or not-QUIT toplevel ACL now matters
1052 rather than the verbs used.
1054 JH/23 Bug 1572: Increase limit on SMTP confirmation message copy size
1055 from 255 to 1024 chars.
1057 JH/24 Verification callouts now attempt to use TLS by default.
1059 HS/01 DNSSEC options (dnssec_require_domains, dnssec_request_domains)
1060 are generic router options now. The defaults didn't change.
1062 JH/25 Bug 466: Add RFC2322 support for MIME attachment filenames.
1063 Original patch from Alexander Shikoff, worked over by JH.
1065 HS/02 Bug 1575: exigrep falls back to autodetection of compressed
1066 files if ZCAT_COMMAND is not executable.
1068 JH/26 Bug 1539: Add timeout/retry options on dnsdb lookups.
1070 JH/27 Bug 286: Support SOA lookup in dnsdb lookups.
1072 JH/28 Bug 1588: Do not use the A lookup following an AAAA for setting the FQDN.
1073 Normally benign, it bites when the pair was led to by a CNAME;
1074 modern usage is to not canonicalize the domain to a CNAME target
1075 (and we were inconsistent anyway for A-only vs AAAA+A).
1077 JH/29 Bug 1632: Removed the word "rejected" from line logged for ACL discards.
1079 JH/30 Check the forward DNS lookup for DNSSEC, in addition to the reverse,
1080 when evaluating $sender_host_dnssec.
1082 JH/31 Check the HELO verification lookup for DNSSEC, adding new
1083 $sender_helo_dnssec variable.
1085 JH/32 Bug 1397: Enable ECDHE on OpenSSL, just the NIST P-256 curve.
1087 JH/33 Bug 1346: Note MAIL cmd seen in -bS batch, to avoid smtp_no_mail log.
1089 JH/34 Bug 1648: Fix a memory leak seen with "mailq" and large queues.
1091 JH/35 Bug 1642: Fix support of $spam_ variables at delivery time. Was
1092 documented as working, but never had. Support all but $spam_report.
1094 JH/36 Bug 1659: Guard checking of input smtp commands again pseudo-command
1095 added for tls authenticator.
1097 HS/03 Add perl_taintmode main config option
1103 TL/01 When running the test suite, the README says that variables such as
1104 no_msglog_check are global and can be placed anywhere in a specific
1105 test's script, however it was observed that placement needed to be near
1106 the beginning for it to behave that way. Changed the runtest perl
1107 script to read through the entire script once to detect and set these
1108 variables, reset to the beginning of the script, and then run through
1109 the script parsing/test process like normal.
1111 TL/02 The BSD's have an arc4random API. One of the functions to induce
1112 adding randomness was arc4random_stir(), but it has been removed in
1113 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
1114 function when detected.
1116 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
1117 cause callback expansion.
1119 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
1120 syntax errors in an expansion can be treated as a string instead of
1121 logging or causing an error, due to the internal use of bool_lax
1122 instead of bool when processing it.
1124 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
1125 server certificates when making smtp deliveries.
1127 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
1129 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
1131 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
1133 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
1134 Merged patch from Sebastian Wiedenroth.
1136 JH/05 Fix results-pipe from transport process. Several recipients, combined
1137 with certificate use, exposed issues where response data items split
1138 over buffer boundaries were not parsed properly. This eventually
1139 resulted in duplicates being sent. This issue only became common enough
1140 to notice due to the introduction of connection certificate information,
1141 the item size being so much larger. Found and fixed by Wolfgang Breyha.
1143 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
1144 size buffer was used, resulting in syntax errors when an expansion
1147 JH/07 Add support for directories of certificates when compiled with a GnuTLS
1148 version 3.3.6 or later.
1150 JH/08 Rename the TPDA experimental facility to Event Actions. The #ifdef
1151 is EXPERIMENTAL_EVENT, the main-configuration and transport options
1152 both become "event_action", the variables become $event_name, $event_data
1153 and $event_defer_errno. There is a new variable $verify_mode, usable in
1154 routers, transports and related events. The tls:cert event is now also
1155 raised for inbound connections, if the main configuration event_action
1158 TL/06 In test suite, disable OCSP for old versions of openssl which contained
1159 early OCSP support, but no stapling (appears to be less than 1.0.0).
1161 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
1162 server certificate names available under the smtp transport option
1163 "tls_verify_cert_hostname" now do not permit multi-component wildcard
1166 JH/10 Time-related extraction expansions from certificates now use the main
1167 option "timezone" setting for output formatting, and are consistent
1168 between OpenSSL and GnuTLS compilations. Bug 1541.
1170 JH/11 Fix a crash in mime ACL when meeting a zero-length, quoted or RFC2047-
1171 encoded parameter in the incoming message. Bug 1558.
1173 JH/12 Bug 1527: Autogrow buffer used in reading spool files. Since they now
1174 include certificate info, eximon was claiming there were spoolfile
1177 JH/13 Bug 1521: Fix ldap lookup for single-attr request, multiple-attr return.
1179 JH/14 Log delivery-related information more consistently, using the sequence
1180 "H=<name> [<ip>]" wherever possible.
1182 TL/07 Bug 1547: Omit RFCs from release. Draft and RFCs have licenses which
1183 are problematic for Debian distribution, omit them from the release
1186 JH/15 Updates and fixes to the EXPERIMENTAL_DSN feature.
1188 JH/16 Fix string representation of time values on 64bit time_t architectures.
1191 JH/17 Fix a null-indirection in certextract expansions when a nondefault
1192 output list separator was used.
1197 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
1198 checkers that were complaining about end of non-void function with no
1201 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
1202 This was a regression introduced in 4.83 by another bugfix.
1204 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
1206 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
1207 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
1213 TF/01 Correctly close the server side of TLS when forking for delivery.
1215 When a message was received over SMTP with TLS, Exim failed to clear up
1216 the incoming connection properly after forking off the child process to
1217 deliver the message. In some situations the subsequent outgoing
1218 delivery connection happened to have the same fd number as the incoming
1219 connection previously had. Exim would try to use TLS and fail, logging
1220 a "Bad file descriptor" error.
1222 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
1223 utilities have not been installed.
1225 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
1226 temporary space as the ACL may create new global variables.
1228 TL/01 LDAP support uses per connection or global context settings, depending
1229 upon the detected version of the libraries at build time.
1231 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
1232 to extract and use the src ip:port in logging and expansions as if it
1233 were a direct connection from the outside internet. PPv2 support was
1234 updated based on HAProxy spec change in May 2014.
1236 JH/02 Add ${listextract {number}{list}{success}{fail}}.
1238 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
1239 Properly escape header and check for NULL return.
1241 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
1244 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
1246 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
1247 characters in header names, implemented as a verify condition.
1248 Contributed by Michael Fischer v. Mollard.
1250 TL/05 Rename SPF condition results err_perm and err_temp to standardized
1251 results permerror and temperror. Previous values are deprecated but
1252 still accepted. In a future release, err_perm and err_temp will be
1253 completely removed, which will be a backward incompatibility if the
1254 ACL tests for either of these two old results. Patch contributed by
1255 user bes-internal on the mailing list.
1257 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
1259 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
1260 selectors, in both main and reject logs.
1262 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
1265 JH/07 Add malware type "sock" for talking to simple daemon.
1267 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
1269 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
1270 routers/transports under cutthrough routing.
1272 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
1273 numbers. Touch up "bool" conditional to keep the same definition.
1275 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
1277 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
1279 JH/12 Expand items in router/transport headers_add or headers_remove lists
1280 individually rather than the list as a whole. Bug 1452.
1282 Required for reasonable handling of multiple headers_ options when
1283 they may be empty; requires that headers_remove items with embedded
1284 colons must have them doubled (or the list-separator changed).
1286 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
1287 view the policy declared in the DMARC record. Currently, $dmarc_status
1288 is a combined value of both the record presence and the result of the
1291 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
1293 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
1294 dnslookup router and the smtp transport (applying to the forward
1297 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
1298 of ldap servers used for a specific lookup. Patch provided by Heiko
1301 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
1302 New variable $lookup_dnssec_authenticated for observability.
1304 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
1305 Patch submitted by Lars Timman.
1307 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
1309 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
1310 Requires trusted mode and valid format message id, aborts otherwise.
1311 Patch contributed by Heiko Schlichting.
1313 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
1314 certextract with support for various fields. Bug 1358.
1316 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
1317 is requested by default, modifiable by smtp transport option
1320 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
1321 operate on certificate variables to give certificate fingerprints
1322 Also new ${sha256:cert_variable}.
1324 JH/23 The PRDR feature is moved from being Experimental into the mainline.
1326 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
1327 Christian Aistleitner.
1329 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
1331 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
1332 file. Patch from Wolfgang Breyha.
1334 JH/25 Expand the coverage of the delivery $host and $host_address to
1335 client authenticators run in verify callout. Bug 1476.
1337 JH/26 Port service names are now accepted for tls_on_connect_ports, to
1338 align with daemon_smtp_ports. Bug 72.
1340 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
1341 support and error reporting did not work properly.
1343 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
1344 and is readable. Patch from Andrew Colin Kissa.
1346 TL/14 Enhance documentation of ${run expansion and how it parses the
1347 commandline after expansion, particularly in the case when an
1348 unquoted variable expansion results in an empty value.
1350 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
1352 PP/02 Fix internal collision of T_APL on systems which support RFC3123
1353 by renaming away from it. Addresses GH issue 15, reported by
1356 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
1358 TL/15 SECURITY: prevent double expansion in math comparison functions
1359 (can expand unsanitized data). Not remotely exploitable.
1366 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
1368 PP/02 Make -n do something, by making it not do something.
1369 When combined with -bP, the name of an option is not output.
1371 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
1374 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
1375 $sender_host_name and config options to manage this, and basic check
1378 PP/05 DSCP support for outbound connections and control modifier for inbound.
1380 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
1381 (Only plugin which currently uses this is kerberos4, which nobody should
1382 be using, but we should make it available and other future plugins might
1383 conceivably use it, even though it would break NAT; stuff *should* be
1384 using channel bindings instead).
1386 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
1387 name; added for Sendmail compatibility; requires admin caller.
1388 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
1389 just ignore it); requires trusted caller.
1390 Also parse but ignore: -Ac -Am -X<logfile>
1393 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
1395 TL/02 Add +smtp_confirmation as a default logging option.
1397 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
1398 Patch by Magnus Holmgren from 2007-02-20.
1400 TL/04 Bugzilla 1281 - Spec typo.
1401 Bugzilla 1283 - Spec typo.
1402 Bugzilla 1290 - Spec grammar fixes.
1404 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
1406 TL/06 Add Experimental DMARC support using libopendmarc libraries.
1408 TL/07 Fix an out of order global option causing a segfault. Reported to dev
1409 mailing list by by Dmitry Isaikin.
1411 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
1413 JH/02 Support "G" suffix to numbers in ${if comparisons.
1415 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
1417 NM/01 Bugzilla 1197 - Spec typo
1418 Bugzilla 1196 - Spec examples corrections
1420 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
1422 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
1423 gnutls_enable_pkcs11, but renamed to more accurately indicate its
1426 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
1427 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
1429 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
1430 "acl {{name}{arg}...}", and optional args on acl condition
1433 JH/05 Permit multiple router/transport headers_add/remove lines.
1435 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
1437 JH/07 Avoid using a waiting database for a single-message-only transport.
1438 Performance patch from Paul Fisher. Bugzilla 1262.
1440 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
1443 JH/09 Add $headers_added variable, with content from use of ACL modifier
1444 add_header (but not yet added to the message). Bugzilla 199.
1446 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
1447 Pulled from Bugzilla 817 by Wolfgang Breyha.
1449 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
1451 (nb: this is the same fix as in Exim 4.80.1)
1453 JH/11 Add A= logging on delivery lines, and a client_set_id option on
1456 JH/12 Add optional authenticated_sender logging to A= and a log_selector
1459 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
1461 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
1462 advertise SMTP AUTH mechanism to us, instead of a generic
1463 protocol violation error. Also, make Exim more robust to bad
1464 data from the Dovecot auth socket.
1466 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
1468 When a queue runner is handling a message, Exim first routes the
1469 recipient addresses, during which it prunes them based on the retry
1470 hints database. After that it attempts to deliver the message to
1471 any remaining recipients. It then updates the hints database using
1474 So if a recipient address works intermittently, it can get repeatedly
1475 deferred at routing time. The retry hints record remains fresh so the
1476 address never reaches the final cutoff time.
1478 This is a fairly common occurrence when a user is bumping up against
1479 their storage quota. Exim had some logic in its local delivery code
1480 to deal with this. However it did not apply to per-recipient defers
1481 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
1483 This change adds a proper retry rule check during routing so that the
1484 final cutoff time is checked against the message's age. We only do
1485 this check if there is an address retry record and there is not a
1486 domain retry record; this implies that previous attempts to handle
1487 the address had the retry_use_local_parts option turned on. We use
1488 this as an approximation for the destination being like a local
1489 delivery, as in LMTP.
1491 I suspect this new check makes the old local delivery cutoff check
1492 redundant, but I have not verified this so I left the code in place.
1494 TF/02 Correct gecos expansion when From: is a prefix of the username.
1496 Test 0254 submits a message to Exim with the header
1500 When I ran the test suite under the user fanf2, Exim expanded
1501 the header to contain my full name, whereas it should have added
1502 a Resent-Sender: header. It erroneously treats any prefix of the
1503 username as equal to the username.
1505 This change corrects that bug.
1507 GF/01 DCC debug and logging tidyup
1508 Error conditions log to paniclog rather than rejectlog.
1509 Debug lines prefixed by "DCC: " to remove any ambiguity.
1511 TF/03 Avoid unnecessary rebuilds of lookup-related code.
1513 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
1514 Bug spotted by Jeremy Harris; was flawed since initial commit.
1515 Would have resulted in OCSP responses post-SNI triggering an Exim
1516 NULL dereference and crash.
1518 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
1520 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
1521 Bug detection, analysis and fix by Samuel Thibault.
1522 Bugzilla 1331, Debian bug #698092.
1524 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
1526 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
1527 Server implementation by Todd Lyons, client by JH.
1528 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
1529 config variable "prdr_enable" controls whether the server
1530 advertises the facility. If the client requests PRDR a new
1531 acl_data_smtp_prdr ACL is called once for each recipient, after
1532 the body content is received and before the acl_smtp_data ACL.
1533 The client is controlled by both of: a hosts_try_prdr option
1534 on the smtp transport, and the server advertisement.
1535 Default client logging of deliveries and rejections involving
1536 PRDR are flagged with the string "PRDR".
1538 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
1539 fclose(). Diagnosis by Todd Lyons.
1541 PP/17 Update configure.default to handle IPv6 localhost better.
1542 Patch by Alain Williams (plus minor tweaks).
1545 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
1546 This is now consistent with GnuTLS, and is now documented: the
1547 previous undocumented portable approach to treating the option as
1548 unset was to force an expansion failure. That still works, and
1549 an empty string is now equivalent.
1551 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
1552 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
1553 not performing validation itself.
1555 PP/20 Added force_command boolean option to pipe transport.
1556 Patch from Nick Koston, of cPanel Inc.
1558 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
1561 TF/04 Added udpsend ACL modifier and hexquote expansion operator
1563 PP/21 Fix eximon continuous updating with timestamped log-files.
1564 Broken in a format-string cleanup in 4.80, missed when I repaired the
1565 other false fix of the same issue.
1566 Report and fix from Heiko Schlichting.
1569 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
1570 Report from Prashanth Katuri.
1572 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
1573 It's SecureTransport, so affects any MacOS clients which use the
1574 system-integrated TLS libraries, including email clients.
1576 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
1577 using a MIME ACL for non-SMTP local injection.
1578 Report and assistance in diagnosis by Warren Baker.
1580 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
1582 JH/16 Fix comparisons for 64b. Bugzilla 1385.
1584 TL/09 Add expansion variable $authenticated_fail_id to keep track of
1585 last id that failed so it may be referenced in subsequent ACL's.
1587 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
1590 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
1591 ldap library initialization, allowing self-signed CA's to be
1592 used. Also properly sets require_cert option later in code by
1593 using NULL (global ldap config) instead of ldap handle (per
1594 session). Bug diagnosis and testing by alxgomz.
1596 TL/12 Enhanced documentation in the ratelimit.pl script provided in
1597 the src/util/ subdirectory.
1599 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
1600 renamed to Transport Post Delivery Action by Jeremy Harris, as
1603 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
1604 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
1605 redis_servers = needs to be configured which will be used by the redis
1606 lookup. Patch from Warren Baker, of The Packet Hub.
1608 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
1610 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
1611 hostname or reverse DNS when processing a host list. Used suggestions
1612 from multiple comments on this bug.
1614 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
1616 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
1617 Missed a few lines, added it to make the runtest require no keyboard
1620 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
1621 contains upper case chars. Make router use caseful_local_part.
1623 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
1624 support when GnuTLS has been built with p11-kit.
1630 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
1632 This, or similar/improved, will also be change PP/11 of 4.82.
1638 PP/01 Handle short writes when writing local log-files.
1639 In practice, only affects FreeBSD (8 onwards).
1640 Bugzilla 1053, with thanks to Dmitry Isaikin.
1642 NM/01 Bugzilla 949 - Documentation tweak
1644 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
1647 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
1649 PP/02 Implemented gsasl authenticator.
1651 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
1653 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
1654 `pkg-config foo` for cflags/libs.
1656 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
1657 with rest of GSASL and with heimdal_gssapi.
1659 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
1660 `pkg-config foo` for cflags/libs for the TLS implementation.
1662 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
1663 properties get this fed in as external SSF. A number of robustness
1664 and debugging improvements to the cyrus_sasl authenticator.
1666 PP/08 cyrus_sasl server now expands the server_realm option.
1668 PP/09 Bugzilla 1214 - Log authentication information in reject log.
1669 Patch by Jeremy Harris.
1671 PP/10 Added dbmjz lookup type.
1673 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
1675 PP/12 MAIL args handles TAB as well as SP, for better interop with
1676 non-compliant senders.
1677 Analysis and variant patch by Todd Lyons.
1679 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
1680 Bug report from Lars Müller <lars@samba.org> (via SUSE),
1681 Patch from Dirk Mueller <dmueller@suse.com>
1683 PP/13 tls_peerdn now print-escaped for spool files.
1684 Observed some $tls_peerdn in wild which contained \n, which resulted
1685 in spool file corruption.
1687 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
1688 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
1689 or write after TLS renegotiation, which otherwise led to messages
1692 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
1693 as a tracking header (ie: a signed header comes before the signature).
1694 Patch from Wolfgang Breyha.
1696 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
1697 comma-sep list; embedded commas doubled.
1699 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
1701 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
1703 Report and patch from Dmitry Banschikov.
1705 PP/16 Removed "dont_insert_empty_fragments" from "openssl_options".
1706 Removed SSL_clear() after SSL_new() which led to protocol negotiation
1707 failures. We appear to now support TLS1.1+ with Exim.
1709 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
1710 lets Exim select keys and certificates based upon TLS SNI from client.
1711 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
1712 before an outbound SMTP session. New log_selector, +tls_sni.
1714 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
1715 NULL dereference. Report and patch from Alun Jones.
1717 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
1718 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
1719 Not seeing resolver debug output on NetBSD, but suspect this is a
1720 resolver implementation change.
1722 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
1723 Left warnings. Added "eximon gdb" invocation mode.
1725 PP/21 Defaulting "accept_8bitmime" to true, not false.
1727 PP/22 Added -bw for inetd wait mode support.
1729 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
1730 locate the relevant includes and libraries. Made this the default.
1732 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
1733 Bugzilla 1246, report and most of solution from Tomasz Kusy.
1735 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
1736 This may cause build issues on older platforms.
1738 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
1739 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
1740 gnutls_require_mac & gnutls_require_protocols (no longer supported).
1741 Added SNI support via GnuTLS too.
1742 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
1744 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
1746 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
1747 Applied second patch from Janne, implementing suggestion to default
1748 multiple-strings-in-record handling to match SPF spec.
1750 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
1752 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
1753 read-only, out of scope).
1754 Patch from Wolfgang Breyha, report from Stuart Northfield.
1756 PP/29 Fix three issues highlighted by clang analyser static analysis.
1757 Only crash-plausible issue would require the Cambridge-specific
1758 iplookup router and a misconfiguration.
1759 Report from Marcin Mirosław.
1761 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
1763 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
1764 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
1765 As part of this, removing so much warning spew let me fix some minor
1766 real issues in debug logging.
1768 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
1769 assignment on my part. Fixed.
1771 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
1772 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
1773 Janne Snabb (who went above and beyond: thank you).
1775 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
1776 string otherwise requires a connection and a bunch more work and it's
1777 relatively easy to get wrong. Should also expose TLS library linkage
1780 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
1781 64-bit ${eval} (JH/03).
1783 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
1784 GNU libc to support some of the 64-bit stuff, should not lead to
1785 conflicts. Defined before os.h is pulled in, so if a given platform
1786 needs to override this, it can.
1788 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
1789 protection layer was required, which is not implemented.
1790 Bugzilla 1254, patch from Wolfgang Breyha.
1792 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
1793 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
1794 tls_dhparam take prime identifiers. Also unbreak combination of
1795 OpenSSL+DH_params+TLSSNI.
1797 PP/39 Disable SSLv2 by default in OpenSSL support.
1803 PP/01 Solaris build fix for Oracle's LDAP libraries.
1804 Bugzilla 1109, patch from Stephen Usher.
1806 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
1808 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
1811 TF/02 Fix a couple more cases where we did not log the error message
1812 when unlink() failed. See also change 4.74-TF/03.
1814 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
1815 lock up or crash if it happened to be inside a call to libc when it
1816 got a SIGUSR1 from exiwhat.
1818 The SIGUSR1 handler appends the current process status to the process
1819 log which is later printed by exiwhat. It used to use the general
1820 purpose logging code to do this, but several functions it calls are
1821 not safe for signals.
1823 The new output code in the SIGUSR1 handler is specific to the process
1824 log, and simple enough that it's easy to inspect for signal safety.
1825 Removing some special cases also simplifies the general logging code.
1826 Removing the spurious timestamps from the process log simplifies
1829 TF/04 Improved ratelimit ACL condition.
1831 The /noupdate option has been deprecated in favour of /readonly which
1832 has clearer semantics. The /leaky, /strict, and /readonly update modes
1833 are mutually exclusive. The update mode is no longer included in the
1834 database key; it just determines when the database is updated. (This
1835 means that when you upgrade Exim will forget old rate measurements.)
1837 Exim now checks that the per_* options are used with an update mode that
1838 makes sense for the current ACL. For example, when Exim is processing a
1839 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
1840 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
1841 must specify per_mail/readonly. If you omit the update mode it defaults to
1842 /leaky where that makes sense (as before) or /readonly where required.
1844 The /noupdate option is now undocumented but still supported for
1845 backwards compatibility. It is equivalent to /readonly except that in
1846 ACLs where /readonly is required you may specify /leaky/noupdate or
1847 /strict/noupdate which are treated the same as /readonly.
1849 A useful new feature is the /count= option. This is a generalization
1850 of the per_byte option, so that you can measure the throughput of other
1851 aggregate values. For example, the per_byte option is now equivalent
1852 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
1854 The per_rcpt option has been generalized using the /count= mechanism
1855 (though it's more complicated than the per_byte equivalence). When it is
1856 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
1857 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
1858 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
1859 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
1860 Note that using per_rcpt with a non-readonly update mode in more than
1861 one ACL will cause the recipients to be double-counted. (The per_mail
1862 and per_byte options don't have this problem.)
1864 The handling of very low rates has changed slightly. If the computed rate
1865 is less than the event's count (usually one) then this event is the first
1866 after a long gap. In this case the rate is set to the same as this event's
1867 count, so that the first message of a spam run is counted properly.
1869 The major new feature is a mechanism for counting the rate of unique
1870 events. The new per_addr option counts the number of different
1871 recipients that someone has sent messages to in the last time period. It
1872 behaves like per_rcpt if all the recipient addresses are different, but
1873 duplicate recipient addresses do not increase the measured rate. Like
1874 the /count= option this is a general mechanism, so the per_addr option
1875 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
1876 example, measure the rate that a client uses different sender addresses
1877 with the options per_mail/unique=$sender_address. There are further
1878 details in the main documentation.
1880 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
1882 TF/06 Removed a few PCRE remnants.
1884 TF/07 Automatically extract Exim's version number from tags in the git
1885 repository when doing development or release builds.
1887 PP/02 Raise smtp_cmd_buffer_size to 16kB.
1888 Bugzilla 879. Patch from Paul Fisher.
1890 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
1891 Heavily based on revision 40f9a89a from Simon Arlott's tree.
1894 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
1896 PP/05 Variable $av_failed, true if the AV scanner deferred.
1897 Bugzilla 1078. Patch from John Horne.
1899 PP/06 Stop make process more reliably on build failure.
1900 Bugzilla 1087. Patch from Heiko Schlittermann.
1902 PP/07 Make maildir_use_size_file an _expandable_ boolean.
1903 Bugzilla 1089. Patch from Heiko Schlittermann.
1905 PP/08 Handle ${run} returning more data than OS pipe buffer size.
1906 Bugzilla 1131. Patch from Holger Weiß.
1908 PP/09 Handle IPv6 addresses with SPF.
1909 Bugzilla 860. Patch from Wolfgang Breyha.
1911 PP/10 GnuTLS: support TLS 1.2 & 1.1.
1913 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
1916 PP/11 match_* no longer expand right-hand-side by default.
1917 New compile-time build option, EXPAND_LISTMATCH_RHS.
1918 New expansion conditions, "inlist", "inlisti".
1920 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
1922 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
1924 PP/14 fix log_write() format string regression from TF/03.
1925 Bugzilla 1152. Patch from Dmitry Isaikin.
1931 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
1933 PP/02 Harmonised TLS library version reporting; only show if debugging.
1934 Layout now matches that introduced for other libraries in 4.74 PP/03.
1936 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
1938 PP/04 New "dns_use_edns0" global option.
1940 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
1943 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
1944 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
1946 TK/01 Updated PolarSSL code to 0.14.2.
1947 Bugzilla 1097. Patch from Andreas Metzler.
1949 PP/07 Catch divide-by-zero in ${eval:...}.
1950 Fixes bugzilla 1102.
1952 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
1955 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
1956 format-string attack -- SECURITY: remote arbitrary code execution.
1958 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
1959 time unintentionally subject to list matching rules, letting the header
1960 cause arbitrary Exim lookups (of items which can occur in lists, *not*
1961 arbitrary string expansion). This allowed for information disclosure.
1963 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
1964 INT_MIN/-1 -- value coerced to INT_MAX.
1970 NM/01 Workaround for PCRE version dependency in version reporting
1973 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
1974 This fixes portability to compilers other than gcc, notably
1975 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
1977 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
1978 makefiles for portability to HP-UX and POSIX correctness.
1980 PP/01 Permit LOOKUP_foo enabling on the make command-line.
1981 Also via indented variable definition in the Makefile.
1982 (Debugging by Oliver Heesakkers).
1984 PP/02 Restore caching of spamd results with expanded spamd_address.
1985 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
1987 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
1988 Improves build reliability. Fix from: Frank Elsner
1990 NM/02 Fix wide character breakage in the rfc2047 coding
1991 Fixes bug 1064. Patch from Andrey N. Oktyabrski
1993 NM/03 Allow underscore in dnslist lookups
1994 Fixes bug 1026. Patch from Graeme Fowler
1996 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
1997 Code patches from Adam Ciarcinski of NetBSD.
1999 NM/04 Fixed exiqgrep to cope with mailq missing size issue
2002 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
2003 is logged, to avoid truncation. Patch from John Horne.
2005 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
2006 Patch from Jakob Hirsch.
2008 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
2009 SQL string expansion failure details.
2010 Patch from Andrey Oktyabrski.
2012 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
2013 Patch from Simon Arlott.
2015 PP/09 New lookups functionality failed to compile on old gcc which rejects
2016 extern declarations in function scope.
2017 Patch from Oliver Fleischmann
2019 PP/10 Use sig_atomic_t for flags set from signal handlers.
2020 Check getgroups() return and improve debugging.
2021 Fixed developed for diagnosis in bug 927 (which turned out to be
2024 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
2025 Patch from Mark Zealey.
2027 PP/12 Bugzilla 1056: Improved spamd server selection.
2028 Patch from Mark Zealey.
2030 PP/13 Bugzilla 1086: Deal with maildir quota file races.
2031 Based on patch from Heiko Schlittermann.
2033 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
2034 Patch from Uwe Doering, sign-off by Michael Haardt.
2036 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
2037 variable declaration deep within a block. Bug and patch from
2040 PP/15 lookups-Makefile IRIX compatibility coercion.
2042 PP/16 Make DISABLE_DKIM build knob functional.
2044 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
2045 Patch by Simon Arlott
2047 TF/03 Fix valgrind.h portability to C89 compilers that do not support
2048 variable argument macros. Our copy now differs from upstream.
2054 TF/01 Failure to get a lock on a hints database can have serious
2055 consequences so log it to the panic log.
2057 TF/02 Log LMTP confirmation messages in the same way as SMTP,
2058 controlled using the smtp_confirmation log selector.
2060 TF/03 Include the error message when we fail to unlink a spool file.
2062 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
2063 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
2064 for maintaining out-of-tree patches for some time.
2066 PP/01 Bugzilla 139: Documentation and portability issues.
2067 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
2068 Handle per-OS dynamic-module compilation flags.
2070 PP/02 Let /dev/null have normal permissions.
2071 The 4.73 fixes were a little too stringent and complained about the
2072 permissions on /dev/null. Exempt it from some checks.
2073 Reported by Andreas M. Kirchwitz.
2075 PP/03 Report version information for many libraries, including
2076 Exim version information for dynamically loaded libraries. Created
2077 version.h, now support a version extension string for distributors
2078 who patch heavily. Dynamic module ABI change.
2080 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
2081 privilege escalation vulnerability whereby the Exim run-time user
2082 can cause root to append content of the attacker's choosing to
2085 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
2088 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
2089 If dropping privileges for untrusted macros, we disabled normal logging
2090 on the basis that it would fail; for the Exim run-time user, this is not
2091 the case, and it resulted in successful deliveries going unlogged.
2092 Fixed. Reported by Andreas Metzler.
2098 PP/01 Date: & Message-Id: revert to normally being appended to a message,
2099 only prepend for the Resent-* case. Fixes regression introduced in
2100 Exim 4.70 by NM/22 for Bugzilla 607.
2102 PP/02 Include check_rfc2047_length in configure.default because we're seeing
2103 increasing numbers of administrators be bitten by this.
2105 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
2107 PP/03 Bugzilla 994: added openssl_options main configuration option.
2109 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
2111 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
2113 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
2115 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
2116 without a peer certificate, leading to a segfault because of an
2117 assumption that peers always have certificates. Be a little more
2118 paranoid. Problem reported by Martin Tscholak.
2120 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
2121 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
2122 NB: ClamAV planning to remove STREAM in "middle of 2010".
2123 CL also introduces -bmalware, various -d+acl logging additions and
2124 more caution in buffer sizes.
2126 PP/09 Implemented reverse_ip expansion operator.
2128 PP/10 Bugzilla 937: provide a "debug" ACL control.
2130 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
2132 PP/12 Bugzilla 973: Implement --version.
2134 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
2136 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
2138 PP/15 Bugzilla 816: support multiple condition rules on Routers.
2140 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
2141 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
2142 ignore trailing whitespace.
2144 JJ/02 prevent non-panic DKIM error from being sent to paniclog
2146 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
2149 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
2150 Notification from Dr Andrew Aitchison.
2152 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
2153 ExtendedDetectionInfo response format.
2154 Notification from John Horne.
2156 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
2159 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
2160 XSL and documented dependency on system catalogs, with examples of how
2163 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
2166 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
2167 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
2168 configuration file which is writeable by the Exim user or group.
2170 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
2171 of configuration files to cover files specified with the -C option if
2172 they are going to be used with root privileges, not just the default
2175 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
2176 option (effectively making it always true).
2178 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
2179 files to be used while preserving root privileges.
2181 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
2182 that rogue child processes cannot use them.
2184 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
2185 run-time user, instead of root.
2187 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
2188 Exim run-time user without dropping privileges.
2190 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
2191 result string, instead of calling string_vformat() twice with the same
2194 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
2195 for other users. Others should always drop root privileges if they use
2196 -C on the command line, even for a whitelisted configure file.
2198 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
2200 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
2206 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
2207 $data_path, and $header_path variables; fixed documentation bugs and
2210 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
2211 exipick to access non-standard spools, including the "frozen" queue
2214 NM/01 Bugzilla 965: Support mysql stored procedures.
2215 Patch from Alain Williams
2217 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
2219 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
2220 Patch from Andreas Metzler
2222 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
2223 Patch from Kirill Miazine
2225 NM/05 Bugzilla 671: Added umask to procmail example.
2227 JJ/03 installed exipick 20100323.0, fixing doc bug
2229 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
2230 directory. Notification and patch from Dan Rosenberg.
2232 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
2234 TK/02 Improve log output when DKIM signing operation fails.
2236 MH/01 Treat the transport option dkim_domain as a colon separated
2237 list, not as a single string, and sign the message with each element,
2238 omitting multiple occurences of the same signer.
2240 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
2241 Bugzilla 985, 986. Patch by Simon Arlott
2243 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
2244 Patch by Simon Arlott
2246 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
2247 MBX locking. Notification from Dan Rosenberg.
2253 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
2255 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
2257 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
2259 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
2261 NM/04 Bugzilla 915: Flag broken perl installation during build.
2267 TK/01 Added patch by Johannes Berg that expands the main option
2268 "spamd_address" if it starts with a dollar sign.
2270 TK/02 Write list of recipients to X-Envelope-Sender header when building
2271 the mbox-format spool file for content scanning (suggested by Jakob
2274 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
2275 (http://www.dcc-servers.net/) support via dccifd. Activated by
2276 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
2278 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
2279 by Mark Daniel Reidel <mr@df.eu>.
2281 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
2282 When building exim an external PCRE library is now needed -
2283 PCRE is a system library on the majority of modern systems.
2284 See entry on PCRE_LIBS in EDITME file.
2286 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
2287 conversation. Added nologin parameter to request.
2288 Patch contributed by Kirill Miazine.
2290 TF/01 Do not log submission mode rewrites if they do not change the address.
2292 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
2294 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
2295 log files in place. Contributed by Roberto Lima.
2297 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
2299 TF/03 Bugzilla 615: When checking the local_parts router precondition
2300 after a local_part_suffix or local_part_prefix option, Exim now
2301 does not use the address's named list lookup cache, since this
2302 contains cached lookups for the whole local part.
2304 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
2305 Robert Millan. Documentation is in experimental-spec.txt.
2307 TF/04 Bugzilla 668: Fix parallel build (make -j).
2309 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
2311 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
2312 Patch provided by Jan Srzednicki.
2314 TF/05 Leading white space used to be stripped from $spam_report which
2315 wrecked the formatting. Now it is preserved.
2317 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
2318 that they are available at delivery time.
2320 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
2322 TF/08 TLS error reporting now respects the incoming_interface and
2323 incoming_port log selectors.
2325 TF/09 Produce a more useful error message if an SMTP transport's hosts
2326 setting expands to an empty string.
2328 NM/06 Bugzilla 744: EXPN did not work under TLS.
2329 Patch provided by Phil Pennock.
2331 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
2332 Patch provided by Richard Godbee.
2334 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
2335 acl_smtp_notquit, added index entry.
2337 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
2338 Patch provided by Eugene Bujak.
2340 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
2341 accept(). Patch provided by Maxim Dounin.
2343 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
2344 Patch provided by Phil Pennock.
2346 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
2348 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
2349 Patch provided by Brad "anomie" Jorsch.
2351 NM/14 Bugzilla 622: Added timeout setting to transport filter.
2352 Patch provided by Dean Brooks.
2354 TK/05 Add native DKIM support (does not depend on external libraries).
2356 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
2357 Patch provided by Graeme Fowler.
2359 NM/16 Bugzilla 851: Documentation example syntax fix.
2361 NM/17 Changed NOTICE file to remove references to embedded PCRE.
2363 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
2366 NM/19 Bugzilla 745: TLS version reporting.
2367 Patch provided by Phil Pennock.
2369 NM/20 Bugzilla 167: bool: condition support.
2370 Patch provided by Phil Pennock.
2372 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
2373 clients. Patch provided by Phil Pennock.
2375 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
2376 Patch provided by Brad "anomie" Jorsch.
2378 NM/23 Bugzilla 687: Fix misparses in eximstats.
2379 Patch provided by Heiko Schlittermann.
2381 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
2382 Patch provided by Heiko Schlittermann.
2384 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
2385 plus update to original patch.
2387 NM/26 Bugzilla 799: Documentation correction for ratelimit.
2389 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
2390 Patch provided by David Brownlee.
2392 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
2394 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
2396 NM/30 Bugzilla 888: TLS documentation bugfixes.
2398 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
2400 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
2401 Unlike the original bugzilla I have changed all shell scripts in src tree.
2403 NM/33 Bugzilla 898: Transport filter timeout fix.
2404 Patch by Todd Rinaldo.
2406 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
2407 Patch by Serge Demonchaux.
2409 NM/35 Bugzilla 39: Base64 decode bug fixes.
2410 Patch by Jakob Hirsch.
2412 NM/36 Bugzilla 909: Correct connect() call in dcc code.
2414 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
2416 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
2418 NM/39 Bugzilla 911: Fixed MakeLinks build script.
2424 TK/01 Add preliminary DKIM support. Currently requires a forked version of
2425 ALT-N's libdkim that I have put here:
2426 http://duncanthrax.net/exim-experimental/
2428 Note to Michael Haardt: I had to rename some vars in sieve.c. They
2429 were called 'true' and it seems that C99 defines that as a reserved
2430 keyword to be used with 'bool' variable types. That means you could
2431 not include C99-style headers which use bools without triggering
2432 build errors in sieve.c.
2434 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
2435 as mailq or other aliases. Changed the --help handling significantly
2436 to do whats expected. exim_usage() emits usage/help information.
2438 SC/01 Added the -bylocaldomain option to eximstats.
2440 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
2442 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
2444 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
2450 PH/01 Another patch from the Sieve maintainer.
2452 PH/02 When an IPv6 address is converted to a string for single-key lookup
2453 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
2454 dots are used instead of colons so that keys in lsearch files need not
2455 contain colons. This was done some time before quoting was made available
2456 in lsearch files. However, iplsearch files do require colons in IPv6 keys
2457 (notated using the quote facility) so as to distinguish them from IPv4
2458 keys. This meant that lookups for IP addresses in host lists did not work
2459 for iplsearch lookups.
2461 This has been fixed by arranging for IPv6 addresses to be expressed with
2462 colons if the lookup type is iplsearch. This is not incompatible, because
2463 previously such lookups could never work.
2465 The situation is now rather anomalous, since one *can* have colons in
2466 ordinary lsearch keys. However, making the change in all cases is
2467 incompatible and would probably break a number of configurations.
2469 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
2472 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
2473 conversion specification without a maximum field width, thereby enabling
2474 a rogue spamd server to cause a buffer overflow. While nobody in their
2475 right mind would setup Exim to query an untrusted spamd server, an
2476 attacker that gains access to a server running spamd could potentially
2477 exploit this vulnerability to run arbitrary code as the Exim user.
2479 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
2480 $primary_hostname instead of what libspf2 thinks the hosts name is.
2482 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
2483 a directory entry by the name of the lookup key. Previously, if a
2484 symlink pointed to a non-existing file or a file in a directory that
2485 Exim lacked permissions to read, a lookup for a key matching that
2486 symlink would fail. Now it is enough that a matching directory entry
2487 exists, symlink or not. (Bugzilla 503.)
2489 PH/03 The body_linecount and body_zerocount variables are now exported in the
2492 PH/04 Added the $dnslist_matched variable.
2494 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
2495 This means they are set thereafter only if the connection becomes
2498 PH/06 Added the client_condition to authenticators so that some can be skipped
2499 by clients under certain conditions.
2501 PH/07 The error message for a badly-placed control=no_multiline_responses left
2502 "_responses" off the end of the name.
2504 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
2506 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
2507 (without spaces) instead of just copying the configuration text.
2509 PH/10 Added the /noupdate option to the ratelimit ACL condition.
2511 PH/11 Added $max_received_linelength.
2513 PH/12 Added +ignore_defer and +include_defer to host lists.
2515 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
2516 way in which PCRE > 7.0 is built.
2518 PH/14 Implemented queue_only_load_latch.
2520 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
2521 MAIL command. The effect was to mangle the value on 64-bit systems.
2523 PH/16 Another patch from the Sieve maintainer.
2525 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
2527 PH/18 If a system quota error occurred while trying to create the file for
2528 a maildir delivery, the message "Mailbox is full" was not appended to the
2529 bounce if the delivery eventually timed out. Change 4.67/27 below applied
2530 only to a quota excession during the actual writing of the file.
2532 PH/19 It seems that peer DN values may contain newlines (and other non-printing
2533 characters?) which causes problems in log lines. The DN values are now
2534 passed through string_printing() before being added to log lines.
2536 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
2537 and InterBase are left for another time.)
2539 PH/21 Added message_body_newlines option.
2541 PH/22 Guard against possible overflow in moan_check_errorcopy().
2543 PH/23 POSIX allows open() to be a macro; guard against that.
2545 PH/24 If the recipient of an error message contained an @ in the local part
2546 (suitably quoted, of course), incorrect values were put in $domain and
2547 $local_part during the evaluation of errors_copy.
2553 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
2554 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
2557 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
2558 issue a MAIL command.
2560 PH/02 In an ACL statement such as
2562 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
2564 if a client was not listed at all, or was listed with a value other than
2565 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
2566 the condition was not true (as it should be), so access was not denied.
2567 The bug was that the ! inversion was incorrectly passed on to the second
2568 item. This has been fixed.
2570 PH/03 Added additional dnslists conditions == and =& which are different from
2571 = and & when the dns lookup returns more than one IP address.
2573 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
2574 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
2576 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
2577 FSYNC, which compiles an option called disable_fsync that allows for
2578 bypassing fsync(). The documentation is heavily laced with warnings.
2580 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
2582 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
2583 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
2584 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
2585 including adding "make clean"; (3) Added -fPIC when compiling the test
2586 dynamically loaded module, to get rid of a warning.
2588 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
2589 message fails, move_frozen_messages = true and ignore_bounce_errors_after
2590 = 0s. The bug is otherwise harmless.
2592 PH/07 There was a bug in the dovecot authenticator such that the value of
2593 $auth1 could be overwritten, and so not correctly preserved, after a
2594 successful authentication. This usually meant that the value preserved by
2595 the server_setid option was incorrect.
2597 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
2599 PH/09 Installed PCRE release 7.0.
2601 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
2602 run for batched SMTP input. It is now run at the start of every message
2603 in the batch. While fixing this I discovered that the process information
2604 (output by running exiwhat) was not always getting set for -bs and -bS
2605 input. This is fixed, and it now also says "batched" for BSMTP.
2607 PH/11 Added control=no_pipelining.
2609 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
2610 patch, slightly modified), and move the expansion of helo_data till after
2611 the connection is made in the smtp transport (so it can use these
2614 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
2616 PH/14 Added log_selector = +pid.
2618 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
2620 PH/16 Add ${if forany and ${if forall.
2622 PH/17 Added dsn_from option to vary the From: line in DSNs.
2624 PH/18 Flush SMTP output before performing a callout, unless control =
2625 no_callout_flush is set.
2627 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
2628 was true (the default) a successful delivery failed to delete the retry
2629 item, thus causing premature timeout of the address. The bug is now
2632 PH/20 Added hosts_avoid_pipelining to the smtp transport.
2634 PH/21 Long custom messages for fakedefer and fakereject are now split up
2635 into multiline responses in the same way that messages for "deny" and
2636 other ACL rejections are.
2638 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
2639 with slight modification.
2641 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
2642 draft, changing the syntax and factoring some duplicate code".
2644 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
2645 for deliveries of the second and subsequent messages over the same SMTP
2648 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
2649 ${reduce, with only minor "tidies".
2651 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
2653 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
2654 expansion side effects.
2656 PH/27 When a message times out after an over-quota error from an Exim-imposed
2657 quota, the bounce message says "mailbox is full". This message was not
2658 being given when it was a system quota that was exceeded. It now should
2661 MH/03 Made $recipients available in local_scan(). local_scan() already has
2662 better access to the recipient list through recipients_list[], but
2663 $recipients can be useful in postmaster-provided expansion strings.
2665 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
2666 in the case of a MAIL command with additional options following the
2667 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
2668 were accidentally chopped off.
2670 PH/29 SMTP synchronization checks are implemented when a command is read -
2671 there is a check that no more input is waiting when there shouldn't be
2672 any. However, for some commands, a delay in an ACL can mean that it is
2673 some time before the response is written. In this time, more input might
2674 arrive, invalidly. So now there are extra checks after an ACL has run for
2675 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
2676 pipelining has not been advertised.
2678 PH/30 MH's patch to allow iscntrl() characters to be list separators.
2680 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
2681 returned in the SMTP response when smtp_return_error_details was false.
2682 This has been fixed.
2684 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
2685 instead of the C I/O that was originally supplied, because problems were
2686 reported on Solaris.
2688 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
2689 Exim which did not show up earlier: it was assuming that a call to
2690 SSL_CTX_set_info_callback() might give an error value. In fact, there is
2691 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
2692 was a macro that became an assignment, so it seemed to work. This has
2693 changed to a proper function call with a void return, hence the compile
2694 error. Exim's code has been fixed.
2696 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
2699 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
2701 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
2703 PH/37 If a message is not accepted after it has had an id assigned (e.g.
2704 because it turns out to be too big or there is a timeout) there is no
2705 "Completed" line in the log. When some messages of this type were
2706 selected by exigrep, they were listed as "not completed". Others were
2707 picked up by some special patterns. I have improved the selection
2708 criteria to be more general.
2710 PH/38 The host_find_failed option in the manualroute router can now be set
2711 to "ignore", to completely ignore a host whose IP address cannot be
2712 found. If all hosts are ignored, the behaviour is controlled by the new
2713 host_all_ignored option.
2715 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
2716 homing or because of multiple MX records with /mx) generated more than
2717 one IP address, and the following item turned out to be the local host,
2718 all the secondary addresses of the first item were incorrectly removed
2719 from the list, along with the local host and any following hosts (which
2720 is what is supposed to happen).
2722 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
2723 whoever called Exim into the -H file. In the case of the daemon it was
2724 behaving confusingly. When first started, it used values for whoever
2725 started the daemon, but after a SIGHUP it used the Exim user (because it
2726 calls itself on a restart). I have changed the code so that it now always
2729 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
2730 message are rejected with the same error (e.g. no authentication or bad
2731 sender address), and a DATA command is nevertheless sent (as can happen
2732 with PIPELINING or a stupid MUA), the error message that was given to the
2733 RCPT commands is included in the rejection of the DATA command. This is
2734 intended to be helpful for MUAs that show only the final error to their
2737 PH/42 Another patch from the Sieve maintainer.
2739 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
2740 Eximstats - Fixed some broken HTML links and added missing column headers
2742 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
2743 columns for Rejects, Temp Rejects, Ham, and Spam rows.
2745 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
2747 PH/43 Yet another patch from the Sieve maintainer.
2749 PH/44 I found a way to check for a TCP/IP connection going away before sending
2750 the response to the final '.' that terminates a message, but only in the
2751 case where the client has not sent further data following the '.'
2752 (unfortunately, this is allowed). However, in many cases there won't be
2753 any further data because there won't be any more messages to send. A call
2754 to select() can be used: if it shows that the input is "ready", there is
2755 either input waiting, or the socket has been closed. An attempt to read
2756 the next input character can distinguish the two cases. Previously, Exim
2757 would have sent an OK response which the client would never have see.
2758 This could lead to message repetition. This fix should cure that, at
2759 least in a lot of common cases.
2761 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
2762 advertised in response to EHLO.
2768 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
2769 fixed by 4.65/MH/01 (is this a record?) are fixed:
2771 (i) An empty string was always treated as zero by the numeric comparison
2772 operators. This behaviour has been restored.
2774 (ii) It is documented that the numeric comparison operators always treat
2775 their arguments as decimal numbers. This was broken in that numbers
2776 starting with 0 were being interpreted as octal.
2778 While fixing these problems I realized that there was another issue that
2779 hadn't been noticed. Values of message_size_limit (both the global option
2780 and the transport option) were treated as octal if they started with 0.
2781 The documentation was vague. These values are now always treated as
2782 decimal, and I will make that clear in the documentation.
2788 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
2789 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
2792 MH/01 Don't check that the operands of numeric comparison operators are
2793 integers when their expansion is in "skipping" mode (fixes bug
2794 introduced by 4.64-PH/07).
2796 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
2797 child addresses, Exim now panics and dies. Previously, because the count
2798 is held in a short int, deliveries were likely to be lost. As such a
2799 large number of recipients for a single message is ridiculous
2800 (performance will be very, very poor), I have chosen to impose a limit
2801 rather than extend the field.
2807 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
2808 leftover -K file (the existence of which was triggered by #402).
2809 While we were at it, introduced process PID as part of the -K
2810 filename. This should rule out race conditions when creating
2813 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
2814 processing considerably. Previous code took too long for large mails,
2815 triggering a timeout which in turn triggers #401.
2817 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
2818 in the DK code in transports.c. sendfile() is not really portable,
2819 hence the _LINUX specificness.
2821 TF/01 In the add_headers option to the mail command in an Exim filter,
2822 there was a bug that Exim would claim a syntax error in any
2823 header after the first one which had an odd number of characters
2826 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
2827 callout verification, Exim cached a "reject" for the entire domain. This
2828 is correct for most verifications, but it is not correct for a recipient
2829 verification with use_sender or use_postmaster set, because in that case
2830 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
2831 case of MAIL FROM:<> rejection from other early rejections (e.g.
2832 rejection of HELO). When verifying a recipient using a non-null MAIL
2833 address, the cache is ignored if it shows MAIL FROM:<> rejection.
2834 Whatever the result of the callout, the value of the domain cache is
2835 left unchanged (for any other kind of callout, getting as far as trying
2836 RCPT means that the domain itself is ok).
2838 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
2841 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
2842 manifest itself as EPIPE rather than ECONNECT. When tidying away a
2843 session, the daemon ignores ECONNECT errors and logs others; it now
2844 ignores EPIPE as well.
2846 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
2847 (quoted-printable decoding).
2849 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
2850 later the small subsequent patch to fix an introduced bug.
2852 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
2854 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
2856 PH/08 An error is now given if message_size_limit is specified negative.
2858 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
2859 to be given (somewhat) arbitrary names.
2861 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
2864 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
2865 miscellaneous code fixes
2867 PH/10 Added the log_reject_target ACL modifier to specify where to log
2870 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
2871 hostname. This is wrong, because it relates to the incoming message (and
2872 probably the interface on which it is arriving) and not to the outgoing
2873 callout (which could be using a different interface). This has been
2874 changed to use the value of the helo_data option from the smtp transport
2875 instead - this is what is used when a message is actually being sent. If
2876 there is no remote transport (possible with a router that sets up host
2877 addresses), $smtp_active_hostname is used.
2879 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
2880 tweaks were necessary in order to get it to work (see also 21 below):
2881 (a) The code assumed that strncpy() returns a negative number on buffer
2882 overflow, which isn't the case. Replaced with Exim's string_format()
2884 (b) There were several signed/unsigned issues. I just did the minimum
2885 hacking in of casts. There is scope for a larger refactoring.
2886 (c) The code used strcasecmp() which is not a standard C function.
2887 Replaced with Exim's strcmpic() function.
2888 (d) The code set only $1; it now sets $auth1 as well.
2889 (e) A simple test gave the error "authentication client didn't specify
2890 service in request". It would seem that Dovecot has changed its
2891 interface. Fortunately there's a specification; I followed it and
2892 changed what the client sends and it appears to be working now.
2894 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
2897 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
2898 address A is aliased to B and C, where B exists and C does not. Without
2899 -v the output is "A verified" because verification stops after a
2900 successful redirection if more than one address is generated. However,
2901 with -v the child addresses are also verified. Exim was outputting "A
2902 failed to verify" and then showing the successful verification for C,
2903 with its parentage. It now outputs "B failed to verify", showing B's
2904 parentage before showing the successful verification of C.
2906 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
2907 look up a TXT record in a specific list after matching in a combined
2910 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
2911 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
2912 they consult the DNS. I had assumed they would set it the way they
2913 wanted; and indeed my experiments on Linux seem to show that in some
2914 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
2915 To be on the safe side, however, I have now made the interface to
2916 host_find_byname() similar to host_find_bydns(), with an argument
2917 containing the DNS resolver options. The host_find_byname() function now
2918 sets these options at its start, just as host_find_bydns() does. The smtp
2919 transport options dns_qualify_single and dns_search_parents are passed to
2920 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
2921 of host_find_byname() use the default settings of RES_DEFNAMES
2922 (qualify_single) but not RES_DNSRCH (search_parents).
2924 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
2925 spool_read_header() do less string testing, by means of a preliminary
2926 switch on the second character of optional "-foo" lines. (This is
2927 overdue, caused by the large number of possibilities that now exist.
2928 Originally there were few.) While I was there, I also converted the
2929 str(n)cmp tests so they don't re-test the leading "-" and the first
2930 character, in the hope this might squeeze out yet more improvement.
2932 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
2933 flag allowing group syntax was set by the header_syntax check but not
2934 turned off, possible causing trouble later; (2) The flag was not being
2935 set at all for the header_verify test, causing "group"-style headers to
2936 be rejected. I have now set it in this case, and also caused header_
2937 verify to ignore an empty address taken from a group. While doing this, I
2938 came across some other cases where the code for allowing group syntax
2939 while scanning a header line wasn't quite right (mostly, not resetting
2940 the flag correctly in the right place). These bugs could have caused
2941 trouble for malformed header lines. I hope it is now all correct.
2943 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
2944 with the "reply" argument non-NULL. The code, however (which originally
2945 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
2946 but it didn't always do it. This confused somebody who was copying the
2947 code for some other use. I have removed all the tests.
2949 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
2950 feature that was used to support insecure browsers during the U.S. crypto
2951 embargo. It requires special client support, and Exim is probably the
2952 only MTA that supported it -- and would never use it because real RSA is
2953 always available. This code has been removed, because it had the bad
2954 effect of slowing Exim down by computing (never used) parameters for the
2955 RSA_EXPORT functionality.
2957 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
2958 authenticator to fail if there's a tab character in the incoming data
2959 (there should never be unless someone is messing about, as it's supposed
2960 to be base64-encoded). Also added, on Timo's advice, the "secured" option
2961 if the connection is using TLS or if the remote IP is the same as the
2962 local IP, and the "valid-client-cert option" if a client certificate has
2965 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
2966 authenticators. This can be used for authorization after authentication
2967 succeeds. (In the case of plaintext, it servers for both authentication
2970 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
2971 if any retry times were supplied.
2973 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
2974 connection, where there is no client IP address to check. In this
2975 situation, the verify now always succeeds.
2977 PH/25 Applied John Jetmore's -Mset patch.
2979 PH/26 Added -bem to be like -Mset, but loading a message from a file.
2981 PH/27 In a string expansion for a processed (not raw) header when multiple
2982 headers of the same name were present, leading whitespace was being
2983 removed from all of them, but trailing whitespace was being removed only
2984 from the last one. Now trailing whitespace is removed from each header
2985 before concatenation. Completely empty headers in a concatenation (as
2986 before) are ignored.
2988 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
2989 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
2991 PH/29 [Removed. This was a change that I later backed out, and forgot to
2992 correct the ChangeLog entry (that I had efficiently created) before
2993 committing the later change.]
2995 PH/30 Exim was sometimes attempting to deliver messages that had suffered
2996 address errors (4xx response to RCPT) over the same connection as other
2997 messages routed to the same hosts. Such deliveries are always "forced",
2998 so retry times are not inspected. This resulted in far too many retries
2999 for the affected addresses. The effect occurred only when there were more
3000 hosts than the hosts_max_try setting in the smtp transport when it had
3001 the 4xx errors. Those hosts that it had tried were not added to the list
3002 of hosts for which the message was waiting, so if all were tried, there
3003 was no problem. Two fixes have been applied:
3005 (i) If there are any address or message errors in an SMTP delivery, none
3006 of the hosts (tried or untried) are now added to the list of hosts
3007 for which the message is waiting, so the message should not be a
3008 candidate for sending over the same connection that was used for a
3009 successful delivery of some other message. This seems entirely
3010 reasonable: after all the message is NOT "waiting for some host".
3011 This is so "obvious" that I'm not sure why it wasn't done
3012 previously. Hope I haven't missed anything, but it can't do any
3013 harm, as the worst effect is to miss an optimization.
3015 (ii) If, despite (i), such a delivery is accidentally attempted, the
3016 routing retry time is respected, so at least it doesn't keep
3017 hammering the server.
3019 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
3020 in ${readsocket because some servers need this prod.
3022 PH/32 Added some extra debug output when updating a wait-xxx database.
3024 PH/33 The hint "could be header name not terminated by colon", which has been
3025 given for certain expansion errors for a long time, was not being given
3026 for the ${if def:h_colon_omitted{... case.
3028 PH/34 The spec says: "With one important exception, whenever a domain list is
3029 being scanned, $domain contains the subject domain." There was at least
3030 one case where this was not true.
3032 PH/35 The error "getsockname() failed: connection reset by peer" was being
3033 written to the panic log as well as the main log, but it isn't really
3034 panic-worthy as it just means the connection died rather early on. I have
3035 removed the panic log writing for the ECONNRESET error when getsockname()
3038 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
3039 runs only) independently of the message's sender address. This meant
3040 that, if the 4xx error was in fact related to the sender, a different
3041 message to the same recipient with a different sender could confuse
3042 things. In particular, this can happen when sending to a greylisting
3043 server, but other circumstances could also provoke similar problems.
3044 I have changed the default so that the retry time for these errors is now
3045 based a combination of the sender and recipient addresses. This change
3046 can be overridden by setting address_retry_include_sender=false in the
3049 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
3050 remote server are returned as part of bounce messages. This was not
3051 happening for LMTP over a pipe (the lmtp transport), but now it is the
3052 same for both kinds of LMTP.
3054 PH/38 Despite being documented as not happening, Exim was rewriting addresses
3055 in header lines that were in fact CNAMEs. This is no longer the case.
3057 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
3058 and queue runs started by the daemon processed all messages. This has
3059 been fixed so that -R and -S can now usefully be given with -q<time>.
3061 PH/40 Import PCRE release 6.7 (fixes some bugs).
3063 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
3065 PH/42 Give an error if -q is specified more than once.
3067 PH/43 Renamed the variables $interface_address and $interface_port as
3068 $received_ip_address and $received_port, to make it clear that these
3069 values apply to message reception, and not to the outgoing interface when
3070 a message is delivered. (The old names remain recognized, of course.)
3072 PH/44 There was no timeout on the connect() call when using a Unix domain
3073 socket in the ${readsocket expansion. There now is.
3075 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
3076 be meaningful with "accept".
3078 SC/01 Eximstats V1.43
3079 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
3081 SC/02 Eximstats V1.44
3082 Use a glob alias rather than an array ref in the generated
3083 parser. This improves both readability and performance.
3085 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
3086 Collect SpamAssassin and rejection statistics.
3087 Don't display local sender or destination tables unless
3088 there is data to show.
3089 Added average volumes into the top table text output.
3091 SC/04 Eximstats V1.46
3092 Collect data on the number of addresses (recipients)
3093 as well as the number of messages.
3095 SC/05 Eximstats V1.47
3096 Added 'Message too big' to the list of mail rejection
3097 reasons (thanks to Marco Gaiarin).
3099 SC/06 Eximstats V1.48
3100 Mainlog lines which have GMT offsets and are too short to
3101 have a flag are now skipped.
3103 SC/07 Eximstats V1.49 (Alain Williams)
3104 Added the -emptyok flag.
3106 SC/08 Eximstats V1.50
3107 Fixes for obtaining the IP address from reject messages.
3109 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
3110 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
3111 whitespace changes from 4.64-PH/27
3113 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
3116 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
3117 are found, allow negative numbers in numeric criteria)
3119 JJ/06 exipick.20061117.2, added new $message_body_missing variable
3121 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
3122 to match changes made in 4.64-PH/43
3124 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
3126 PH/47 Put in an explicit test for a DNS lookup of an address record where the
3127 "domain" is actually an IP address, and force a failure. This locks out
3128 those revolvers/nameservers that support "A-for-A" lookups, in
3129 contravention of the specifications.
3131 PH/48 When a host name was looked up from an IP address, and the subsequent
3132 forward lookup of the name timed out, the host name was left in
3133 $sender_host_name, contrary to the specification.
3135 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
3136 restricted to single-key lookups, Exim was not diagnosing an error if
3137 * or *@ was used with a query-style lookup.
3139 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
3141 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
3142 long ago, but noone interested enough thought of it. Let's just say that
3143 the "1.1" means that there are some new functions that weren't there at
3144 some point in the past.
3146 PH/51 Error processing for expansion failure of helo_data from an smtp
3147 transport during callout processing was broken.
3149 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
3150 tested/used via the -bh/-bhc/-bs options.
3152 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
3153 bug, fixed in subsequent PCRE releases).
3155 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
3156 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
3158 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
3164 SC/01 Use a glob alias rather than an array ref in eximstats generated
3165 parser. This improves both readability and performance.
3167 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
3168 Don't display local sender or destination tables in eximstats unless
3169 there is data to show.
3170 Added average volumes into the eximstats top table text output.
3172 SC/03 Collect data on the number of addresses (recipients) as well
3173 as the number of messages in eximstats.
3175 TF/01 Correct an error in the documentation for the redirect router. Exim
3176 does (usually) call initgroups() when daemonizing.
3178 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
3179 with consistent privilege compared to when running as a daemon.
3181 TF/03 Note in the spec that $authenticated_id is not set for local
3182 submissions from trusted users.
3184 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
3185 Thanks to Dean Brooks <dean@iglou.com> for the patch.
3187 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
3188 by adding some example configuration directives to the default
3189 configuration file. A little bit of work is required to uncomment the
3190 directives and define how usernames and passwords are checked, but
3191 there is now a framework to start from.
3193 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
3194 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
3195 without this. I don't know how relevant this is to other LDAP libraries.
3197 PH/02 Add the verb name to the "unknown ACL verb" error.
3199 PH/03 Magnus Holmgren's patch for filter_prepend_home.
3201 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
3203 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
3204 directory not expanded when it should be if an expanded home directory
3205 was set for the address (which is overridden by the transport).
3207 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
3210 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
3211 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
3212 because it is too late at that time, and has no effect.
3214 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
3215 security issue with \' (bugzilla #107). I could not use the
3216 PQescapeStringConn() function, because it needs a PGconn value as one of
3219 PH/08 When testing addresses using -bt, indicate those final addresses that
3220 are duplicates that would not cause an additional delivery. At least one
3221 person was confused, thinking that -bt output corresponded to deliveries.
3222 (Suppressing duplicates isn't a good idea as you lose the information
3223 about possibly different redirections that led to the duplicates.)
3225 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
3226 systems where poll() doesn't work, in particular OS X.
3228 PH/10 Added more information to debugging output for retry time not reached.
3230 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
3231 operations in malware.c.
3233 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
3236 PH/13 If write_rejectlog was set false when logging was sent to syslog with
3237 syslog_duplication set false, log lines that would normally be written
3238 both the the main log and to the reject log were not written to syslog at
3241 PH/14 In the default configuration, change the use of "message" in ACL warn
3242 statements to "add_header".
3244 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
3245 not followed by a command (e.g. "seen endif").
3247 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
3248 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
3251 PH/17 Added extra conditions to the default value of delay_warning_condition
3255 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
3256 { match{$h_precedence:}{(?i)bulk|list|junk} } \
3257 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
3260 The Auto-Submitted: and various List- headers are standardised, whereas I
3261 don't think Precedence: ever was.
3263 PH/18 Refactored debugging code in route_finduser() to show more information,
3264 in particular, the error code if getpwnam() issues one.
3266 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
3267 This is apparently needed in addition to the PH/07 change above to avoid
3268 any possible encoding problems.
3270 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
3271 but not after initializing Perl.
3273 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
3274 output them only if debugging. By default they are written stderr,
3275 apparently, which is not desirable.
3277 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
3280 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
3283 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
3285 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
3286 authenticated or an ident call has been made. Suppress the default
3287 values for $authenticated_id and $authenticated_sender (but permit -oMai
3288 and -oMas) when testing with -bh.
3290 PH/24 Re-jigged the order of the tests in the default configuration so that the
3291 tests for valid domains and recipients precede the DNS black list and CSA
3292 tests, on the grounds that those ones are more expensive.
3294 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
3295 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
3296 This bug exists in every version of Exim that I still have, right back to
3299 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
3300 However, an attempt to turn on case-sensitivity in a regex key by
3301 including (?-i) didn't work because the subject string was already
3302 lowercased, and the effects were non-intuitive. It turns out that a
3303 one-line patch can be used to allow (?-i) to work as expected.
3309 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
3310 other effects) broke the use of negated acl sub-conditions.
3312 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
3315 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
3316 "Deny" causes Exim to reject the incoming connection with a 554 error.
3317 Unfortunately, if there is a major crisis, such as a disk failure,
3318 tcp-wrappers gives "deny", whereas what one would like would be some
3319 kind of temporary error. A kludge has been added to help with this.
3320 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
3321 554 error is used if errno is still zero or contains ENOENT (which occurs
3322 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
3325 PH/03 Add -lutil to the default FreeBSD LIBS setting.
3327 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
3328 errors. Otherwise a message that provokes a temporary error (when other
3329 messages do not) can cause a whole host to time out.
3331 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
3332 addresses were routed directly to files or pipes from a redirect router.
3333 File deliveries just didn't batch; pipe deliveries might have suffered
3336 PH/06 A failure to get a lock for a hints database would erroneously always say
3337 "Failed to get write lock", even when it was really a read lock.
3339 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
3340 of 0600. This has been changed to use the value of the lockfile_mode
3341 option (which defaults to 0600).
3343 PH/08 Applied small patch from the Sieve maintainer.
3345 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
3346 folder from quota calculations, a direct delivery into this folder messed
3347 up the contents of the maildirsize file. This was because the regex was
3348 used only to exclude .Trash (or whatever) when the size of the mailbox
3349 was calculated. There was no check that a delivery was happening into an
3350 excluded directory. This bug has been fixed by ignoring all quota
3351 processing for deliveries into excluded directories.
3353 PH/10 Added the maildirfolder_create_regex option to appendfile.
3359 PH/01 The code for finding all the local interface addresses on a FreeBSD
3360 system running IPv6 was broken. This may well have applied to all BSD
3361 systems, as well as to others that have similar system calls. The broken
3362 code found IPv4 interfaces correctly, but gave incorrect values for the
3363 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
3364 that it would not match correctly against @[] and not recognize the IPv6
3367 PH/02 The ipliteral router was not recognizing addresses of the form user@
3368 [ipv6:....] because it didn't know about the "ipv6:" prefix.
3370 PH/03 Added disable_ipv6.
3372 PH/04 Changed $reply_address to use the raw form of the headers instead of the
3373 decoded form, because it is most often used to construct To: headers
3374 lines in autoreplies, and the decoded form may well be syntactically
3375 invalid. However, $reply_address has leading white space removed, and all
3376 newlines turned into spaces so that the autoreply transport does not
3379 PH/05 If group was specified without a user on a router, and no group or user
3380 was specified on a transport, the group from the router was ignored.
3382 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
3383 for visible compile-time settings that can be used to change these
3384 numbers, for those that want even more. Backwards compatibility with old
3385 spool files has been maintained. However, going back to a previous Exim
3386 release will lost any variables that are in spool files.
3388 PH/07 Two small changes when running in the test harness: increase delay when
3389 passing a TCP/IP connection to a new process, in case the original
3390 process has to generate a bounce, and remove special handling of
3391 127.0.0.2 (sic), which is no longer necessary.
3393 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
3394 be the same on different OS.
3396 PH/09 Moved a debug statement in filter processing to avoid a race problem when
3399 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
3400 whether --show-vars was specified or not
3402 JJ/02 exipick: Added support for new ACL variable spool format introduced
3405 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
3406 syntactically invalid From: or Reply-to: line, and a filter used this to
3407 generate an autoreply, and therefore failed to obtain an address for the
3408 autoreply, Exim could try to deliver to a non-existent relative file
3409 name, causing unrelated and misleading errors. What now happens is that
3410 it logs this as a hard delivery error, but does not attempt to create a
3413 PH/11 The exinext utility has a -C option for testing purposes, but although
3414 the given file was scanned by exinext itself; it wasn't being passed on
3415 when Exim was called.
3417 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
3418 an end-of-file indication when reading a command response.
3420 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
3421 compiled. In many other places in Exim, IPv6 addresses are always
3422 recognized, so I have changed this. It also means that IPv4 domain
3423 literals of the form [IPV4:n.n.n.n] are now always recognized.
3425 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
3426 used if the router is not running as root, for example, when verifying at
3427 ACL time, or when using -bh. The debugging output from this situation was
3428 non-existent - all you got was a failure to exec. I have made two
3431 (a) Failures to set uid/gid, the current directory, or a process leader
3432 in a subprocess such as that created by queryprogram now generate
3433 suitable debugging output when -d is set.
3435 (b) The queryprogram router detects when it is not running as root,
3436 outputs suitable debugging information if -d is set, and then runs
3437 the subprocess without attempting to change uid/gid.
3439 PH/15 Minor change to Makefile for building test_host (undocumented testing
3442 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
3443 additional section of a DNS packet that returns MX or SRV records.
3444 Instead, it always explicitly searches for A/AAAA records. This avoids
3445 major problems that occur when a DNS server includes only records of one
3446 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
3447 fixed another bug: if SRV records were looked up and the corresponding
3448 address records were *not* found in the additional section, the port
3449 values from the SRV records were lost.
3451 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
3452 using the correct key (the original address) when searching the retry
3453 rules in order to find which one to use for generating the retry hint.
3455 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
3456 adding the default one. Similarly, if it contains a Reply-To: header, the
3457 errors_reply_to option, if set, is not used.
3459 PH/19 When calculating a retry time, Exim used to measure the "time since
3460 failure" by looking at the "first failed" field in the retry record. Now
3461 it does not use this if it is later than than the arrival time of the
3462 message. Instead it uses the arrival time. This makes for better
3463 behaviour in cases where some deliveries succeed, thus re-setting the
3464 "first failed" field. An example is a quota failure for a huge message
3465 when small messages continue to be delivered. Without this change, the
3466 "time since failure" will always be short, possible causing more frequent
3467 delivery attempts for the huge message than are intended.
3468 [Note: This change was subsequently modified - see PH/04 for 4.62.]
3470 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
3471 $1, $2, $3) because the numerical variables can be reset during some
3472 expansion items (e.g. "match"), thereby losing the authentication data.
3474 PH/21 Make -bV show the size of off_t variables so that the test suite can
3475 decide whether to run tests for quotas > 2G.
3477 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
3478 mailbox_size, and mailbox_filecount in the appendfile transport. If a
3479 filecount value is greater than 2G or if a quota value is greater than 2G
3480 on a system where the size of off_t is not greater than 4, a panic error
3483 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
3484 never match. The debug and -bh output now contains an explicit error
3485 message indicating a malformed IPv4 address or mask.
3487 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
3488 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
3489 PH/23 above applies.
3491 PH/25 Do not write to syslog when running in the test harness. The only
3492 occasion when this arises is a failure to open the main or panic logs
3493 (for which there is an explicit test).
3495 PH/26 Added the /no_tell option to "control=freeze".
3497 PH/27 If a host name lookup failed very early in a connection, for example, if
3498 the IP address matched host_lookup and the reverse lookup yielded a name
3499 that did not have a forward lookup, an error message of the form "no IP
3500 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
3501 could be logged. Now it outputs the IP address instead of "NULL".
3503 PH/28 An enabling patch from MH: add new function child_open_exim2() which
3504 allows the sender and the authenticated sender to be set when
3505 submitting a message from within Exim. Since child_open_exim() is
3506 documented for local_scan(), the new function should be too.
3508 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
3509 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
3510 results in an empty string is now treated as unset.
3512 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
3514 PH/31 Added sender_verify_fail logging option.
3516 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
3517 needed by RFC 822 but not by RFC 2822 was commented out. I have now
3518 tidied the source and removed it altogether.
3520 PH/33 When a queue run was abandoned because the load average was too high, a
3521 log line was always written; now it is written only if the queue_run log
3522 selector is set. In addition, the log line for abandonment now contains
3523 information about the queue run such as the pid. This is always present
3524 in "start" and "stop" lines but was omitted from the "abandon" line.
3526 PH/34 Omit spaces between a header name and the colon in the error message that
3527 is given when verify = headers_syntax fails (if there are lots of them,
3528 the message gets confusing).
3530 PH/35 Change the default for dns_check_names_pattern to allow slashes within
3531 names, as there are now some PTR records that contain slashes. This check
3532 is only to protect against broken name servers that fall over on strange
3533 characters, so the fact that it applies to all lookups doesn't matter.
3535 PH/36 Now that the new test suite is complete, we can remove some of the
3536 special code in Exim that was needed for the old test suite. For example,
3537 sorting DNS records because real resolvers return them in an arbitrary
3538 order. The new test suite's fake resolver always returns records in the
3541 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
3542 bounces) except when queue_only is set, to avoid logging races between
3543 the different processes.
3545 PH/38 Panic-die if .include specifies a non-absolute path.
3547 PH/39 A tweak to the "H" retry rule from its user.
3549 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
3550 a label. They prevented compilation on older perls.
3552 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
3553 a warning to be raised on newish perls.
3555 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
3556 on queue. Changes to match documented behaviour of showing count of
3557 messages matching specified criteria.
3559 PH/40 Changed the default ident timeout from 30s to 5s.
3561 PH/41 Added support for the use of login_cap features, on those BSD systems
3562 that have them, for controlling the resources used by pipe deliveries.
3564 PH/42 The content-scanning code uses fopen() to create files in which to put
3565 message data. Previously it was not paying any attention to the mode of
3566 the files. Exim runs with umask(0) because the rest of the code creates
3567 files with open(), and sets the required mode explicitly. Thus, these
3568 files were ending up world-writeable. This was not a big issue, because,
3569 being within the spool directory, they were not world-accessible. I have
3570 created a function called modefopen, which takes an additional mode
3571 argument. It sets umask(777), creates the file, chmods it to the required
3572 mode, then resets the umask. All the relevant calls to fopen() in the
3573 content scanning code have been changed to use this function.
3575 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
3576 to 24 hours. This avoids potential overflow problems when processing G
3577 and H retry rules. I suspect nobody ever tinkers with this value.
3579 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
3581 PH/45 When the plaintext authenticator is running as a client, the server's
3582 challenges are checked to ensure they are valid base64 strings. By
3583 default, the authentication attempt is cancelled if an invalid string is
3584 received. Setting client_ignore_invalid_base64 true ignores these errors.
3585 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
3586 they are received. Thus, the responses can be made to depend on the
3587 challenges. If an invalid string is ignored, an empty string is placed in
3590 PH/46 Messages that are created by the autoreply transport now contains a
3591 References: header, in accordance with RFCs 2822 and 3834.
3593 PH/47 Added authenticated_sender_force to the smtp transport.
3595 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
3597 PH/49 Installed latest patch from the Sieve maintainer.
3599 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
3600 was also set, the appendfile transport was unnecessarily scanning a
3601 directory of message files (e.g. for maildir delivery) to find the count
3602 of files (along with the size), even though it did not need this
3603 information. It now does the scan only if it needs to find either the
3604 size of the count of files.
3606 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
3608 PH/52 Two bugs concerned with error handling when the smtp transport is
3611 (i) Exim was not creating retry information for temporary errors given
3612 for individual recipients after the DATA command when the smtp transport
3613 was used in LMTP mode. This meant that they could be retried too
3614 frequently, and not timed out correctly.
3616 (ii) Exim was setting the flag that allows error details to be returned
3617 for LMTP errors on RCPT commands, but not for LMTP errors for individual
3618 recipients that were returned after the DATA command.
3620 PH/53 This is related to PH/52, but is more general: for any failing address,
3621 when detailed error information was permitted to be returned to the
3622 sender, but the error was temporary, then after the final timeout, only
3623 "retry timeout exceeded" was returned. Now it returns the full error as
3624 well as "retry timeout exceeded".
3626 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
3627 do this, and (what is worse) MTAs that accept it.
3629 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
3630 will now be deprecated.
3632 PH/56 New os.c-cygwin from the Cygwin maintainer.
3634 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
3635 formats (previously only available in exim formats via -bpr, -bpru,
3636 and -bpra. Now also available in native and exiqgrep formats)
3638 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
3639 with very large, slow to parse queues
3641 JJ/08 exipick: added ! as generic prefix to negate any criteria format
3643 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
3645 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
3646 responses to authentication challenges, though it was showing the
3647 challenges; (ii) I've removed the CR characters from the debug output for
3650 PH/58 Allow for the insertion of a newline as well as a space when a string
3651 is turned into more than one encoded-word during RFC 2047 encoding. The
3652 Sieve code now uses this.
3654 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
3655 data_4xx, lost_connection, tls_required.
3657 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
3658 message was being sent as an SMTP response.
3660 PH/61 Add -l and -k options to exicyclog.
3662 PH/62 When verifying, if an address was redirected to one new address, so that
3663 verification continued, and the new address failed or deferred after
3664 having set something in $address_data, the value of $address_data was not
3665 passed back to the ACL. This was different to the case when no
3666 redirection occurred. The value is now passed back in both cases.
3668 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
3669 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
3670 use login_cap.h, so on its own it isn't the distinguishing feature. The
3671 new name refers directly to the setclassresources() function.
3673 PH/65 Added configuration files for NetBSD3.
3675 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
3677 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
3678 is preferred over IPv4.
3680 PH/68 The bounce_return_message and bounce_return_body options were not being
3681 honoured for bounces generated during the reception of non-SMTP messages.
3682 In particular, this applied to messages rejected by the ACL. This bug has
3683 been fixed. However, if bounce_return_message is true and bounce_return_
3684 body is false, the headers that are returned for a non-SMTP message
3685 include only those that have been read before the error was detected.
3686 (In the case of an ACL rejection, they have all been read.)
3688 PH/69 The HTML version of the specification is now built in a directory called
3689 spec_html instead of spec.html, because the latter looks like a path with
3690 a MIME-type, and this confuses some software.
3692 PH/70 Catch two compiler warnings in sieve.c.
3694 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
3695 function verify_get_ident() calls ip_connect() to connect a socket, but
3696 if the "connect()" function timed out, ip_connect() used to close the
3697 socket. However, verify_get_ident() also closes the socket later, and in
3698 between Exim writes to the log, which may get opened at this point. When
3699 the socket was closed in ip_connect(), the log could get the same file
3700 descriptor number as the socket. This naturally causes chaos. The fix is
3701 not to close the socket in ip_connect(); the socket should be closed by
3702 the function that creates it. There was only one place in the code where
3703 this was missing, in the iplookup router, which I don't think anybody now
3704 uses, but I've fixed it anyway.
3706 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
3707 well as to direct DNS lookups. Otherwise the handling of names in host
3708 lists is inconsistent and therefore confusing.
3714 PH/01 Two changes to the default runtime configuration:
3716 (1) Move the checks for relay_from_hosts and authenticated clients from
3717 after to before the (commented out) DNS black list checks.
3719 (2) Add control=submission to the relay_from_hosts and authenticated
3720 clients checks, on the grounds that messages accepted by these
3721 statements are most likely to be submissions.
3723 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
3725 (1) Generate an error if the third argument for the ${prvs expansion is
3728 (2) Treat a missing third argument of ${prvscheck as if it were an empty
3731 (3) Reset the variables that are obtained from the first argument of
3732 ${prvscheck and used in the second argument before leaving the code,
3733 because their memory is reclaimed, so using them afterwards may do
3736 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
3737 one (it's much easier than Tom thought :-).
3739 (5) Because of (4), we can now allow for the use of $prvscheck_result
3740 inside the third argument.
3742 PH/03 For some reason, the default setting of PATH when running a command from
3743 a pipe transport was just "/usr/bin". I have changed it to
3746 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
3747 anything to be listed in the output from -bV.
3749 PH/05 When a filter generated an autoreply, the entire To: header line was
3750 quoted in the delivery log line, like this:
3752 => >A.N.Other <ano@some.domain> <original@ddress> ...
3754 This has been changed so that it extracts the operative address. There
3755 may be more than one such address. If so, they are comma-separated, like
3758 => >ano@some.domain,ona@other.domain <original@ddress> ...
3760 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
3761 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
3762 not being looked up in the rDNS to get a host name, Exim was showing the
3763 IP address twice in Received: lines, even though the IP addresses were
3764 identical. For example:
3766 Received: from [1.2.3.4] (helo=[1.2.3.4])
3768 However, if the real host name was known, it was omitting the HELO data
3769 if it matched the actual IP address. This has been tidied up so that it
3770 doesn't show the same IP address twice.
3772 PH/07 When both +timestamp and +memory debugging was on, the value given by
3773 $tod_xxx expansions could be wrong, because the tod_stamp() function was
3774 called by the debug printing, thereby overwriting the timestamp buffer.
3775 Debugging no longer uses the tod_stamp() function when +timestamp is set.
3777 PH/08 When the original message was included in an autoreply transport, it
3778 always said "this is a copy of the message, including all the headers",
3779 even if body_only or headers_only was set. It now gives an appropriate
3782 PH/09 Applied a patch from the Sieve maintainer which:
3784 o fixes some comments
3785 o adds the (disabled) notify extension core
3786 o adds some debug output for the result of if/elsif tests
3787 o points to the current vacation draft in the documentation
3788 and documents the missing references header update
3792 o fixes a bug in processing the envelope test (when testing
3793 multiple envelope elements, the last element determined the
3796 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
3797 Electronic Mail") by including:
3799 Auto-submitted: auto-generated
3801 in the messages that it generates (bounce messages and others, such as
3802 warnings). In the case of bounce messages for non-SMTP messages, there was
3803 also a typo: it was using "Auto_submitted" (underscore instead of
3804 hyphen). Since every message generated by Exim is necessarily in response
3805 to another message, thes have all been changed to:
3807 Auto-Submitted: auto-replied
3809 in accordance with these statements in the RFC:
3811 The auto-replied keyword:
3813 - SHOULD be used on messages sent in direct response to another
3814 message by an automatic process,
3816 - MUST NOT be used on manually-generated messages,
3818 - MAY be used on Delivery Status Notifications (DSNs) and Message
3819 Disposition Notifications (MDNs),
3821 - MUST NOT be used on messages generated by automatic or periodic
3822 processes, except for messages which are automatic responses to
3825 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
3826 to the default Received: header definition.
3828 PH/12 Added log selector acl_warn_skipped (default on).
3830 PH/13 After a successful wildlsearch lookup, discard the values of numeric
3831 variables because (a) they are in the wrong storage pool and (b) even if
3832 they were copied, it wouldn't work properly because of the caching.
3834 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
3835 checking when decoding. Apparently there are clients that generate
3836 overlong encoded strings. Why am I not surprised?
3838 PH/15 If the first argument of "${if match_address" was not empty, but did not
3839 contain an "@" character, Exim crashed. Now it writes a panic log message
3840 and treats the condition as false.
3842 PH/16 In autoreply, treat an empty string for "once" the same as unset.
3844 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
3845 extension "envelope-auth". The code is finished and in agreement with
3846 other implementations, but there is no documentation so far and in fact,
3847 nobody wrote the draft yet. This extension is currently #undef'ed, thus
3848 not changing the active code.
3850 Print executed "if" and "elsif" statements when debugging is used. This
3851 helps a great deal to understand what a filter does.
3853 Document more things not specified clearly in RFC3028. I had all this
3854 sorted out, when out of a sudden new issues came to my mind. Oops."
3856 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
3859 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
3860 canonical form (as documented). However, after a host name lookup from
3861 the IP address, check_host() was doing a simple string comparison with
3862 addresses acquired from the DNS when checking that the found name did
3863 have the original IP as one of its addresses. Since any found IPv6
3864 addresses are likely to be in abbreviated form, the comparison could
3865 fail. Luckily, there already exists a function for doing the comparison
3866 by converting both addresses to binary, so now that is used instead of
3867 the text comparison.
3869 PH/20 There was another similar case to PH/19, when a complete host name was
3870 given in a host list; looking up its IP address could give an abbreviated
3871 form, whereas the current host's name might or might not be abbreviated.
3872 The same fix has been applied.
3878 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
3879 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
3882 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
3883 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
3885 PH/03 Typo: missing ".o" in src/pcre/Makefile.
3887 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
3888 header line, restrict the check to what is listed in RFCs 2369 and 2929.
3889 Also, for "Auto-Submitted", treat anything other than "no" as
3890 non-personal, in accordance with RFC 3834. (Previously it treated
3891 anything starting "auto-" as non-personal.)
3893 TF/01 The control=submission/name=... option had a problem with syntax
3894 errors if the name included a slash character. The /name= option
3895 now slurps the rest of the string, so it can include any characters
3896 but it must come last in the list of options (after /sender_retain
3899 PH/05 Some modifications to the interface to the fake nameserver for the new
3907 TK/01 Added the "success_on_redirect" address verification option. See
3908 NewStuff for rationale and an example.
3910 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
3912 PH/02 Patch to exigrep to allow it to work on syslog lines.
3914 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
3915 fread() to skip over the body file's header line, because in Cygwin the
3916 header line is locked and is inaccessible.
3918 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
3919 co-exist for some time) to make it clear that it is the Exim ID that is
3920 referenced, not the Message-ID: header line.
3922 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
3923 string_format() function, because snprintf() does not exist on all
3926 PH/06 The use of forbid_filter_existstest now also locks out the use of the
3927 ${stat: expansion item.
3929 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
3930 protocol synchronization error", to keep the pedants happy.
3932 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
3933 well as for IRIX systems, when gcc is being used. See the host.c source
3936 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
3938 PH/10 Named domain lists were not working if used in a queue_smtp_domains
3941 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
3942 transport and to the smtp transport in LMTP mode.
3944 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
3946 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
3947 run a filter in a subprocess. This could lead to confusion in subsequent
3948 lookups in the parent process. There should also be a search_tidyup() at
3949 the end of the subprocess.
3951 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
3952 only if the host matched helo_try_verify_hosts, which caused the
3953 verification to occur when the EHLO/HELO command was issued. The ACL just
3954 tested the remembered result. Now, if a previous verification attempt has
3955 not happened, "verify = helo" does it there and then.
3957 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
3959 TK/03 Fix log output including CR from clamd.
3961 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
3962 exist provoked a memory error which could cause a segfault.
3964 PH/15 Installed PCRE 6.2
3966 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
3968 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
3969 of the problem. Specifically, suggested +O2 rather than +O1 for the
3972 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
3974 PH/20 If a delivery was routed to a non-standard port by means of an SRV
3975 record, the port was not correctly logged when the outgoing_port log
3976 selector was set (it logged the transort's default port).
3978 PH/21 Added support for host-specific ports to manualroute, queryprogram,
3979 fallback_hosts, and "hosts" in the smtp transport.
3981 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
3982 host errors such as "Connection refused".
3984 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
3985 authentication with radiusclient 0.4.9:
3987 - Error returned from rc_read_config was caught wrongly
3988 - Username/password not passed on to radius server due to wrong length.
3990 The presumption is that some radiusclient API changes for 4.51/PH/17
3991 were not taken care of correctly. The code is still untested by me (my
3992 Linux distribution still has 0.3.2 of radiusclient), but it was
3993 contributed by a Radius user.
3995 PH/24 When doing a callout, the value of $domain wasn't set correctly when
3996 expanding the "port" option of the smtp transport.
3998 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
3999 while reading a MIME header. Thanks to Tom Hughes for a patch.
4001 PH/24 Include config.h inside local_scan.h so that configuration settings are
4004 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
4005 that in an ACL for RCPT (for example), you can examine exactly what was
4008 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
4009 commands, but it was not correctly comparing the address with the actual
4010 client host address. Thus, it would show the EHLO address in Received:
4011 header lines when this was not necessary.
4013 PH/27 Added the % operator to ${eval:}.
4015 PH/28 Exim tries to create and chdir to its spool directory when it starts;
4016 it should be ignoring failures (because with -C, for example, it has lost
4017 privilege). It wasn't ignoring creation failures other than "already
4020 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
4023 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
4024 because an input line was too long, either on its own, or by virtue of
4025 too many continuations, the temporary file was not being removed, and the
4026 return code was incorrect.
4028 PH/31 Missing "BOOL" in function definition in filtertest.c.
4030 PH/32 Applied Sieve patches from the maintainer.
4032 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
4034 PH/33 Added "verify = not_blind".
4036 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
4037 Local/Makefile (with some defaults set). These are used in built scripts
4038 such as exicyclog, but they have never been used in the exim_install
4039 script (though there are many overriding facilities there). I have
4040 arranged that the exim_install script now takes note of these two
4043 PH/35 Installed configuration files for Dragonfly.
4045 PH/36 When a locally submitted message by a trusted user did not contain a
4046 From: header, and the sender address was obtained from -f or from an SMTP
4047 MAIL command, and the trusted user did not use -F to supply a sender
4048 name, $originator_name was incorrectly used when constructing a From:
4049 header. Furthermore, $originator_name was used for submission mode
4050 messages from external hosts without From: headers in a similar way,
4051 which is clearly wrong.
4053 PH/37 Added control=suppress_local_fixups.
4055 PH/38 When log_selector = +received_sender was set, and the addition of the
4056 sender made the log line's construction buffer exactly full, or one byte
4057 less than full, an overflow happened when the terminating "\n" was
4060 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
4061 when the result of a list match is failure because a DNS lookup failed.
4063 PH/40 RM_COMMAND is now used in the building process.
4065 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
4066 the "build-* directories that it finds.
4068 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
4069 address in a domain literal was a prefix of an interface address.
4071 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
4072 when verifying a sender address, unless rewrite_headers is false.
4074 PH/44 Wrote a long comment about why errors_to addresses are verified as
4075 recipients, not senders.
4077 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
4078 the ratelimit ACL was added.
4080 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
4082 PH/46 Added extra information about PostgreSQL errors to the error string.
4084 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
4085 avoiding the need to install special zones in a real server. This is
4086 backwards compatible; if it can't find the fake resolver, it drops back.
4087 Thus, both old and new test suites can be run.
4089 TF/02 Added util/ratelimit.pl
4091 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
4092 clock is set back in time.
4094 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
4095 Candler <B.Candler@pobox.com>.
4097 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
4098 OK for addresses that are the result of redirections.
4100 PH/48 A number of further additions for the benefit of the new test suite,
4101 including a fake gethostbyname() that interfaces to the fake DNS resolver
4104 TF/06 The fix for widen_domains has also been applied to qualify_single and
4105 search_parents which are the other dnslookup options that can cause
4108 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
4111 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
4113 TF/07 Exim produced the error message "an SRV record indicated no SMTP
4114 service" if it encountered an MX record with an empty target hostname.
4115 The message is now "an MX or SRV record indicated no SMTP service".
4117 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
4118 if the DNS of the sending site is misconfigured. This is quite a
4119 common situation. This change restores the behaviour of treating a
4120 helo verification defer as a failure.
4122 PH/51 If self=fail was set on a router, the bounce message did not include the
4123 actual error message.
4129 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
4131 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
4132 command itself ended in error, the underlying message about the transport
4133 filter timeout was being overwritten with the pipe command error. Now the
4134 underlying error message should be appended to the second error message.
4136 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
4138 PH/02 Reduce the amount of output that "make" produces by default. Full output
4139 can still be requested.
4141 PH/03 The warning log line about a condition test deferring for a "warn" verb
4142 was being output only once per connection, rather than after each
4143 occurrence (because it was using the same function as for successful
4144 "warn" verbs). This seems wrong, so I have changed it.
4146 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
4147 it should not have, which might have caused a crash in the right
4148 circumstances, but probably never did.
4150 PH/04 Installed a modified version of Tony Finch's patch to make submission
4151 mode fix the return path as well as the Sender: header line, and to
4152 add a /name= option so that you can make the user's friendly name appear
4155 TF/03 Added the control = fakedefer ACL modifier.
4157 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
4158 Mark Lowes for thorough testing.
4160 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
4162 TK/03 Merged latest SRS patch from Miles Wilton.
4164 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
4165 with the definition in sysexits.h (which is #included earlier).
4166 Fortunately, Exim does not actually use EX_OK. The code used to try to
4167 preserve the sysexits.h value, by assuming that macro definitions were
4168 scanned for macro replacements. I have been disabused of this notion,
4169 so now the code just undefines EX_OK before #including unistd.h.
4171 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
4172 in the smtp transport. When a block could not be written in a single
4173 write() function, the timeout was being re-applied to each part-write.
4174 This seems wrong - if the receiver was accepting one byte at a time it
4175 would take for ever. The timeout is now adjusted when this happens. It
4176 doesn't have to be particularly precise.
4178 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
4179 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
4181 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
4182 without a domain if the check to <postmaster@domain> fails.
4184 SC/01 Eximstats: added -xls and the ability to specify output files
4185 (patch written by Frank Heydlauf).
4187 SC/02 Eximstats: use FileHandles for outputting results.
4189 SC/03 Eximstats: allow any combination of xls, txt, and html output.
4191 SC/04 Eximstats: fixed display of large numbers with -nvr option
4193 SC/05 Eximstats: fixed merging of reports with empty tables.
4195 SC/06 Eximstats: added the -include_original_destination flag
4197 SC/07 Eximstats: removed tabs and trailing whitespace.
4199 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
4201 TK/06 MBOX spool code: Add real "From " MBOX separator line
4202 so the .eml file is really in mbox format (even though
4203 most programs do not really care). Patch from Alex Miller.
4205 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
4206 The latter is generated from $received_to and is only set if the
4207 message has one envelope recipient. SA can use these headers,
4208 obviously out-of-the-box. Patch from Alex Miller.
4210 PH/08 The ${def test on a variable was returning false if the variable's
4211 value was "0", contrary to what the specification has always said!
4212 The result should be true unless the variable is empty.
4214 PH/09 The syntax error of a character other than { following "${if
4215 def:variable_name" (after optional whitespace) was not being diagnosed.
4216 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
4217 accidental colon was present, for example, could give incorrect results.
4219 PH/10 Tidied the code in a number of places where the st_size field of a stat()
4220 result is used (not including appendfile, where other changes are about
4223 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
4224 This involved changing a lot of size variables from int to off_t. It
4225 should work with maildirs and everything.
4227 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
4228 spamd dying while we are connected to it.
4230 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
4233 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
4234 function for BDB 4.3.
4236 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
4238 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
4239 the sources for PCRE have been re-arranged and more files are now
4242 PH/15 The code I had for printing potentially long long variables in PH/11
4243 above was not the best (it lost precision). The length of off_t variables
4244 is now inspected at build time, and an appropriate printing format (%ld
4245 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
4246 to be "long long int" or "long int". This is needed for the internal
4247 formatting function string_vformat().
4249 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
4250 the configuration file to be ":syslog", then the script "guesses" where
4251 the logs files are, rather than using the compiled in default. In our
4252 case the guess is not the same as the compiled default, so the script
4253 suddenly stopped working when I started to use syslog. The patch checks
4254 to see if log_file_path is "". If so, it attempts to read it from exim
4255 with no configuration file to get the compiled in version, before it
4256 falls back to the previous guessing code."
4258 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
4259 implementing BATV in an Exim configuration. See NewStuff for the gory
4262 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
4263 Makefile that are specific to HP-UX.
4265 PH/18 If the "use_postmaster" option was set for a recipient callout together
4266 with the "random" option, the postmaster address was used as the MAIL
4267 FROM address for the random test, but not for the subsequent recipient
4268 test. It is now used for both.
4270 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
4271 patch removes a few documentation additions to RFC 3028, because the
4272 latest draft now contains them. It adds the new en;ascii-case comparator
4273 and a new error check for 8bit text in MIME parts. Comparator and
4274 require names are now matched exactly. I enabled the subaddress
4275 extension, but it is not well tested yet (read: it works for me)."
4277 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
4278 rework some of the code of TK/09 above to avoid the hardwired use of
4279 "%lld" and "long long". Replaced the call to snprintf() with a call to
4282 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
4283 records point to non-existent hosts", "retry timeout exceeded", and
4284 "retry time not reached for any host after a long failure period".
4286 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
4287 experimental DomainKeys support:
4289 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
4290 (2) On an error such as an illegally used "control", the wrong name for
4291 the control was given.
4293 These problems did NOT occur unless DomainKeys support was compiled.
4295 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
4297 PH/24 Added ${if match_ip condition.
4299 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
4300 databases so that it will be absolutely obvious if a crash occurs in the
4301 DB library. This is a regular occurrence (often caused by mis-matched
4304 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
4305 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
4306 picked up on a user's system that detects such things. There doesn't seem
4307 to be a gcc warning option for this - only an attribute that has to be
4308 put on the function's prototype. It seems that in Fedora Core 4 they have
4309 set this on a number of new functions. No doubt there will be more in due
4312 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
4313 specify a transport. However, if an address that was verified by such a
4314 router was the subject of a callout, Exim crashed because it tried to
4315 read the rcpt_include_affixes from the non-existent transport. Now it
4316 just assumes that the setting of that option is false. This bug was
4317 introduced by 4.51/PH/31.
4319 PH/28 Changed -d+all to exclude +memory, because that information is very
4320 rarely of interest, but it makes the output a lot bigger. People tend to
4321 do -d+all out of habit.
4323 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
4324 code in os-type was giving problems when libc.so lives in lib64, like on
4327 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
4328 aren't the modern standard, and it seems that some systems' include files
4329 don't always have them. Exim was already checking for some of the newer
4330 ones like T_AAAA, and defining it itself. I've added checks for all the
4331 record types that Exim uses.
4333 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
4334 not automatically generating a new one, as it is supposed to. This
4335 prevented TLS from working. If the file did exist, but contained invalid
4336 data, a new version was generated, as expected. It was only the case of a
4337 non-existent file that was broken.
4339 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
4340 with a change in libdomainkeys > 0.64.
4342 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
4343 from DNS. If the selector record carries the flag, it now has
4344 precedence over the domain-wide flag.
4346 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
4348 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
4349 the use of a port name that isn't defined in /etc/services) occurred, the
4350 message was deferred as in a normal delivery, and thus remained on the
4351 spool, instead of being failed because of the mua_wrapper setting. This
4352 is now fixed, and I tidied up some of the mua_wrapper messages at the
4355 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
4356 the messages in a hash of arrays rather than using individual hashes.
4357 This is a bit cleaner and results in dramatic memory savings, albeit
4358 at a slight CPU cost.
4360 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
4361 as requested by Marc Sherman.
4363 SC/10 Eximstats: added histograms for user specified patterns as requested
4366 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
4368 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
4369 fopen() in the content-scanning modules that did not already have it.
4375 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
4376 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
4378 TK/02 Fix ACL "control" statement not being available in MIME ACL.
4380 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
4382 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
4383 to test Sieve filters that use "vacation".
4385 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
4386 that changes the way the GnuTLS parameters are stored in the cache file.
4387 The new format can be generated externally. For backward compatibility,
4388 if the data in the cache doesn't make sense, Exim assumes it has read an
4389 old-format file, and it generates new data and writes a new file. This
4390 means that you can't go back to an older release without removing the
4393 PH/03 A redirect router that has both "unseen" and "one_time" set does not
4394 work if there are any delivery delays because "one_time" forces the
4395 parent to be marked "delivered", so its unseen clone is never tried
4396 again. For this reason, Exim now forbids the simultaneous setting of
4399 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
4400 redirected to themselves ("homonym" addresses). Read the long ChangeLog
4401 entry if you want to know the details. The fix, however, neglected to
4402 consider the case when local delivery batching is involved. The test for
4403 "previously delivered" was not happening when checking to see if an
4404 address could be batched with a previous (undelivered) one; under
4405 certain circumstances this could lead to multiple deliveries to the same
4408 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
4409 in its include files, and this causes problems building Exim.
4411 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
4412 header_syntax) but Exim was just ignoring anything given after a slash.
4413 In particular, this caused confusion with an attempt to use "verify =
4414 reverse_host_lookup/defer_ok". An error is now given when options are
4415 supplied for verify items that do not have them. (Maybe reverse_host_
4416 lookup should have a defer_ok option, but that's a different point.)
4418 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
4419 defined by RFC 821) to 2048, because there were problems with some AUTH
4420 commands, and RFC 1869 says the size should be increased for extended
4421 SMTP commands that take arguments.
4423 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
4426 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
4427 "unknown" error; now it says that the functionality isn't in the binary.
4429 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
4430 an address' error message when a string expansion fails (syntax or
4431 whatever). Otherwise the password may appear in the log. Following change
4432 PH/42 below, there is no longer a chance of it appearing in a bounce
4435 PH/11 Installed exipick version 20050225.0 from John Jetmore.
4437 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
4438 of its addresses was ever tried. (Bugzilla bug #2.)
4440 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
4441 the result incorrectly in the debug output. (It correctly added a newline
4442 to what was transported.)
4444 TF/01 Added $received_time.
4446 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
4447 commented out examples of how to interface to a virus scanner and to
4448 SpamAssassin. Also added commented examples of av_scanner and
4449 spamd_address settings.
4451 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
4452 and controls are allowed in which ACLs. There were a couple of minor
4453 errors. Some of the entries in the conditions table (which is a table of
4454 where they are NOT allowed) were getting very unwieldy; rewrote them as a
4455 negation of where the condition IS allowed.
4457 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
4459 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
4460 header file does not have a version number, so I've had to invent a new
4461 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
4462 API. The code is untested by me (my Linux distribution still has 0.3.2 of
4463 radiusclient), but it was contributed by a Radius user.
4465 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
4466 files or directories, for OpenSSL.
4468 PH/19 When an Exim process that is running as root has to create an Exim log
4469 file, it does so in a subprocess that runs as exim:exim so as to get the
4470 ownership right at creation (otherwise, other Exim processes might see
4471 the file with the wrong ownership). There was no test for failure of this
4472 fork() call, which would lead to the process getting stuck as it waited
4473 for a non-existent subprocess. Forks do occasionally fail when resources
4474 run out. I reviewed all the other calls to fork(); they all seem to check
4477 PH/20 When checking for unexpected SMTP input at connect time (before writing
4478 the banner), Exim was not dealing correctly with a non-positive return
4479 from the read() function. If the client had disconnected by this time,
4480 the result was a log entry for a synchronization error with an empty
4481 string after "input=" when read() returned zero. If read() returned -1
4482 (an event I could not check), uninitialized data bytes were printed.
4483 There were reports of junk text (parts of files, etc) appearing after
4486 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
4488 PH/22 Added support for macro redefinition, and (re)definition in between
4489 driver and ACL definitions.
4491 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
4492 forgetting to use the resulting value; it was using the unexpanded value.
4494 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
4495 hadn't been configured. The fix is from Juergen Kreileder, who
4496 understands it better than I do:
4498 "Here's what I see happening with three configured cyrus_sasl
4499 authenticators configured (plain, login, cram-md5):
4501 On startup auth_cyrus_sasl_init() gets called for each of these.
4502 This means three calls to sasl_listmech() without a specified mech_list.
4503 => SASL tests which mechs of all available mechs actually work
4504 => three warnings about OTP not working
4505 => the returned list contains: plain, login, cram-md5, digest-md5, ...
4507 With the patch, sasl_listmech() also gets called three times. But now
4508 SASL's mech_list option is set to the server_mech specified in the the
4509 authenticator. Or in other words, the answer from sasl_listmech()
4510 gets limited to just the mech you're testing for (which is different
4512 => the return list contains just 'plain' or 'login', 'cram-md5' or
4513 nothing depending on the value of ob->server_mech.
4515 I've just tested the patch: Authentication still works fine,
4516 unavailable mechs specified in the exim configuration are still
4517 caught, and the auth.log warnings about OTP are gone."
4519 PH/25 When debugging is enabled, the contents of the command line are added
4520 to the debugging output, even when log_selector=+arguments is not
4523 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
4524 answer is "GNU", and only if the return is "GNU/something" is the answer
4527 PH/27 $acl_verify_message is now set immediately after the failure of a
4528 verification in an ACL, and so is available in subsequent modifiers. In
4529 particular, the message can be preserved by coding like this:
4531 warn !verify = sender
4532 set acl_m0 = $acl_verify_message
4534 Previously, $acl_verify_message was set only while expanding "message"
4535 and "log_message" when a very denied access.
4537 PH/28 Modified OS/os.c-Linux with
4539 -#ifndef OS_LOAD_AVERAGE
4540 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
4542 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
4543 nomenclature these days.)
4545 PH/29 Installed patch from the Sieve maintainer that adds the options
4546 sieve_useraddress and sieve_subaddress to the redirect router.
4548 PH/30 In these circumstances:
4549 . Two addresses routed to the same list of hosts;
4550 . First host does not offer TLS;
4551 . First host accepts first address;
4552 . First host gives temporary error to second address;
4553 . Second host offers TLS and a TLS session is established;
4554 . Second host accepts second address.
4555 Exim incorrectly logged both deliveries with the TLS parameters (cipher
4556 and peerdn, if requested) that were in fact used only for the second
4559 PH/31 When doing a callout as part of verifying an address, Exim was not paying
4560 attention to any local part prefix or suffix that was matched by the
4561 router that accepted the address. It now behaves in the same way as it
4562 does for delivery: the affixes are removed from the local part unless
4563 rcpt_include_affixes is set on the transport.
4565 PH/32 Add the sender address, as F=<...>, to the log line when logging a
4566 timeout during the DATA phase of an incoming message.
4568 PH/33 Sieve envelope tests were broken for match types other than :is. I have
4569 applied a patch sanctioned by the Sieve maintainer.
4571 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
4572 the uid or gid is negative. A case of a negative gid caused this to be
4573 noticed. The fix allows for either to be negative.
4575 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
4576 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
4579 PH/36 The change PH/12 above was broken. Fixed it.
4581 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
4582 the grounds that routing the same address twice would always produce the
4583 same answer. This might have been true once, but it is certainly no
4584 longer true now. Routing a child address may depend on the previous
4585 routing that produced that child. Some complicated redirection strategies
4586 went wrong when messages had multiple recipients, and made Exim's
4587 behaviour dependent on the order in which the addresses were given.
4589 I have moved the duplicate checking until after the routing is complete.
4590 Exim scans the addresses that are assigned to local and remote
4591 transports, and removes any duplicates. This means that more work will be
4592 done, as duplicates will always all be routed, but duplicates are
4593 presumably rare, so I don't expect this is of any significance.
4595 For deliveries to pipes, files, and autoreplies, the duplicate checking
4596 still happens during the routing process, since they are not going to be
4599 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
4600 It corrects a timeout issue with spamd. This is Ian's comment: "The
4601 background is that sometimes spamd either never reads data from a
4602 connection it has accepted, or it never writes response data. The exiscan
4603 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
4604 blindly assumes that writes won't block so it may never time out."
4606 PH/39 Allow G after quota size as well as K and M.
4608 PH/40 The value set for $authenticated_id in an authenticator may not contain
4609 binary zeroes or newlines because the value is written to log lines and
4610 to spool files. There was no check on this. Now the value is run through
4611 the string_printing() function so that such characters are converted to
4612 printable escape sequences.
4614 PH/41 $message_linecount is a new variable that contains the total number of
4615 lines in the message. Compare $body_linecount, which is the count for the
4618 PH/42 Exim no longer gives details of delivery errors for specific addresses in
4619 bounce and delay warning messages, except in certain special cases, which
4622 (a) An SMTP error message from a remote host;
4623 (b) A message specified in a :fail: redirection;
4624 (c) A message specified in a "fail" command in a system filter;
4625 (d) A message specified in a FAIL return from the queryprogram router;
4626 (e) A message specified by the cannot_route_message router option.
4628 In these cases only, Exim does include the error details in bounce and
4629 warning messages. There are also a few cases where bland messages such
4630 as "unrouteable address" or "local delivery error" are given.
4632 PH/43 $value is now also set for the "else" part of a ${run expansion.
4634 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
4635 being worked on, but at least Exim now implements the latest version to
4638 PH/45 In a pipe transport, although a timeout while waiting for the pipe
4639 process to complete was treated as a delivery failure, a timeout while
4640 writing the message to the pipe was logged, but erroneously treated as a
4641 successful delivery. Such timeouts include transport filter timeouts. For
4642 consistency with the overall process timeout, these timeouts are now
4643 treated as errors, giving rise to delivery failures by default. However,
4644 there is now a new Boolean option for the pipe transport called
4645 timeout_defer, which, if set TRUE, converts the failures into defers for
4646 both kinds of timeout. A transport filter timeout is now identified in
4649 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
4650 systems where "make" and "gmake" are different, calling "gmake" at top
4651 level broke things. I've arranged for the value of $(MAKE) to be passed
4652 from the Makefile to this script so that it can call the same version of
4656 A note about Exim versions 4.44 and 4.50
4657 ----------------------------------------
4659 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
4660 changes of various kinds. As a consequence, a big documentation update was
4661 needed. This delayed the release for rather longer than seemed good, especially
4662 in the light of a couple of (minor) security issues. Therefore, the changes
4663 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
4664 release. So 4.44 and 4.50 are in effect two different branches that both start
4667 I have left the 4.50 change log unchanged; it contains all the changes since
4668 4.43. The change log for 4.44 is below; many of its items are identical to
4669 those for 4.50. This seems to be the most sensible way to preserve the
4670 historical information.
4676 1. Minor wording change to the doc/README.SIEVE file.
4678 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
4679 computation of the current number of files was incorrect.
4681 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4682 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4685 4. Give more explanation in the error message when the command for a transport
4686 filter fails to execute.
4688 5. There are several places where Exim runs a non-Exim command in a
4689 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4690 was being done only for the command run by the queryprogram router. It is
4691 now done for all such subprocesses. The other cases are: ${run, transport
4692 filters, and the commands run by the lmtp and pipe transports.
4694 6. Added CONFIGURE_GROUP build-time option.
4696 7. Some older OS have a limit of 256 on the maximum number of file
4697 descriptors. Exim was using setrlimit() to set 1000 as a large value
4698 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4699 systems. I've change it so that if it can't get 1000, it tries for 256.
4701 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4702 was an oversight, and furthermore, ever since the addition of extra
4703 controls (e.g. 4.43/32), the checks on when to allow different forms of
4704 "control" were broken. There should now be diagnostics for all cases when a
4705 control that does not make sense is encountered.
4707 9. Added the /retain_sender option to "control=submission".
4709 10. $recipients is now available in the predata ACL (oversight).
4711 11. Tidy the search cache before the fork to do a delivery from a message
4712 received from the command line. Otherwise the child will trigger a lookup
4713 failure and thereby defer the delivery if it tries to use (for example) a
4714 cached ldap connection that the parent has called unbind on.
4716 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4717 of $address_data from the recipient verification was clobbered by the
4718 sender verification.
4720 13. The value of address_data from a sender verification is now available in
4721 $sender_address_data in subsequent conditions in the ACL statement.
4723 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
4725 15. Added a new option "connect=<time>" to callout options, to set a different
4728 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4729 was its contents. (It was OK if the option was not defined at all.)
4731 17. A "Completed" log line is now written for messages that are removed from
4732 the spool by the -Mrm option.
4734 18. New variables $sender_verify_failure and $recipient_verify_failure contain
4735 information about exactly what failed.
4737 19. Added -dd to debug only the daemon process.
4739 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
4740 handles timeouts, both on the server side and network timeouts. Renamed the
4741 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
4743 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
4744 It is now set to "smtps".
4746 22. $host_address is now set to the target address during the checking of
4747 ignore_target_hosts.
4749 23. When checking ignore_target_hosts for an ipliteral router, no host name was
4750 being passed; this would have caused $sender_host_name to have been used if
4751 matching the list had actually called for a host name (not very likely,
4752 since this list is usually IP addresses). A host name is now passed as
4755 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4756 code that specifies a non-restarting handler (typically sigaction() in
4757 modern systems) in an attempt to fix a rare and obscure crash bug.
4759 25. Narrowed the window for a race in the daemon that could cause it to ignore
4760 SIGCHLD signals. This is not a major problem, because they are used only to
4761 wake it up if nothing else does.
4763 26. A malformed maildirsize file could cause Exim to calculate negative values
4764 for the mailbox size or file count. Odd effects could occur as a result.
4765 The maildirsize information is now recalculated if the size or filecount
4768 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4769 support for a long time. Removed HAVE_SYS_VFS_H.
4771 28. Installed the latest version of exipick from John Jetmore.
4773 29. In an address list, if the pattern was not a regular expression, an empty
4774 subject address (from a bounce message) matched only if the pattern was an
4775 empty string. Non-empty patterns were not even tested. This was the wrong
4776 because it is perfectly reasonable to use an empty address as part of a
4777 database query. An empty address is now tested by patterns that are
4778 lookups. However, all the other forms of pattern expect the subject to
4779 contain a local part and a domain, and therefore, for them, an empty
4780 address still always fails if the pattern is not itself empty.
4782 30. Exim went into a mad DNS loop when attempting to do a callout where the
4783 host was specified on an smtp transport, and looking it up yielded more
4784 than one IP address.
4786 31. Re-factored the code for checking spool and log partition space into a
4787 function that finds that data and another that does the check. The former
4788 is then used to implement four new variables: $spool_space, $log_space,
4789 $spool_inodes, and $log_inodes.
4791 32. The RFC2047 encoding function was originally intended for short strings
4792 such as real names; it was not keeping to the 75-character limit for
4793 encoded words that the RFC imposes. It now respects the limit, and
4794 generates multiple encoded words if necessary. To be on the safe side, I
4795 have increased the buffer size for the ${rfc2047: expansion operator from
4798 33. It is now permitted to omit both strings after an "if" condition; if the
4799 condition is true, the result is "true". As before, when the second string
4800 is omitted, a false condition yields an empty string. This makes it less
4801 cumbersome to write custom ACL and router conditions.
4803 34. Failure to deliver a bounce message always caused it to be frozen, even if
4804 there was an errors_to setting on the router. The errors_to setting is now
4807 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4808 canonical form (fully expanded) before being placed in
4809 $sender_host_address.
4811 36. The table in the code that translates DNS record types into text (T_A to
4812 "A" for instance) was missing entries for NS and CNAME. It is just possible
4813 that this could have caused confusion if both these types were looked up
4814 for the same domain, because the text type is used as part of Exim's
4815 per-process caching. But the chance of anyone hitting this buglet seems
4818 37. The dnsdb lookup has been extended in a number of ways.
4820 (1) There is a new type, "zns", which walks up the domain tree until it
4821 finds some nameserver records. It should be used with care.
4823 (2) There is a new type, "mxh", which is like "mx" except that it returns
4824 just the host names, not the priorities.
4826 (3) It is now possible to give a list of domains (or IP addresses) to be
4827 looked up. The behaviour when one of the lookups defers can be
4828 controlled by a keyword.
4830 (4) It is now possible to specify the separator character for use when
4831 multiple records are returned.
4833 38. The dnslists ACL condition has been extended: it is now possible to supply
4834 a list of IP addresses and/or domains to be looked up in a particular DNS
4837 39. Added log_selector=+queue_time_overall.
4839 40. When running the queue in the test harness, wait just a tad after forking a
4840 delivery process, to get repeatability of debugging output.
4842 41. Include certificate and key file names in error message when GnuTLS fails
4843 to set them up, because the GnuTLS error message doesn't include the name
4844 of the failing file when there is a problem reading it.
4846 42. Allow both -bf and -bF in the same test run.
4848 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
4850 44. The "Exiscan patch" is now merged into the mainline Exim source.
4852 45. Sometimes the final signoff response after QUIT could fail to get
4853 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4854 < 0 before doing a fflush(). This bug looks as though it goes back to the
4855 introduction of TLS in release 3.20, but "sometimes" must have been rare
4856 because the tests only now provoked it.
4858 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
4859 (this can affect the format of dates).
4861 47. exim_tidydb, when checking for the continued existence of a message for
4862 which it has found a message-specific retry record, was not finding
4863 messages that were in split spool directories. Consequently, it was
4864 deleting retry records that should have stayed in existence.
4866 48. Steve fixed some bugs in eximstats.
4868 49. The SPA authentication driver was not abandoning authentication and moving
4869 on to the next authenticator when an expansion was forced to fail,
4870 contradicting the general specification for all authenticators. Instead it
4871 was generating a temporary error. It now behaves as specified.
4873 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
4874 (the order specifies the preference for clients). The order is now AES256,
4875 AES128, 3DES, ARCFOUR128.
4877 51. Small patch to Sieve code - explicitly set From: when generating an
4880 52. Exim crashed if a remote delivery caused a very long error message to be
4881 recorded - for instance if somebody sent an entire SpamAssassin report back
4882 as a large number of 550 error lines. This bug was coincidentally fixed by
4883 increasing the size of one of Exim's internal buffers (big_buffer) that
4884 happened as part of the Exiscan merge. However, to be on the safe side, I
4885 have made the code more robust (and fixed the comments that describe what
4888 53. Now that there can be additional text after "Completed" in log lines (if
4889 the queue_time_overall log selector is set), a one-byte patch to exigrep
4890 was needed to allow it to recognize "Completed" as not the last thing in
4893 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
4894 patch that reportedly fixes this has been added. I am not expert enough to
4895 create a test for it. This is what the patch creator wrote:
4897 "I found a little strange behaviour of ldap code when working with
4898 Windows 2003 AD Domain, where users was placed in more than one
4899 Organization Units. When I tried to give exim partial DN, the exit code
4900 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
4901 But simultaneously result of request was absolutely normal ldap result,
4902 so I produce this patch..."
4904 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
4905 REFERENCE, so I have modified the code to exclude the patch when that macro
4908 55. Some experimental protocols are using DNS PTR records for new purposes. The
4909 keys for these records are domain names, not reversed IP addresses. The
4910 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4911 leaves it alone. Component reversal etc. now happens only for IP addresses.
4914 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
4916 57. Double the size of the debug message buffer (to 2048) so that more of very
4917 long debug lines gets shown.
4919 58. The exicyclog utility now does better if the number of log files to keep
4920 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
4922 59. Two changes related to the smtp_active_hostname option:
4924 (1) $smtp_active_hostname is now available as a variable.
4925 (2) The default for smtp_banner uses $smtp_active_hostname instead
4926 of $primary_hostname.
4928 60. The host_aton() function is supposed to be passed a string that is known
4929 to be a valid IP address. However, in the case of IPv6 addresses, it was
4930 not checking this. This is a hostage to fortune. Exim now panics and dies
4931 if the condition is not met. A case was found where this could be provoked
4932 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4933 components; fortuitously, this particular loophole had already been fixed
4934 by change 4.50/55 above.
4936 If there are any other similar loopholes, the new check in host_aton()
4937 itself should stop them being exploited. The report I received stated that
4938 data on the command line could provoke the exploit when Exim was running as
4939 exim, but did not say which command line option was involved. All I could
4940 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4941 running as the user.
4944 61. There was a buffer overflow vulnerability in the SPA authentication code
4945 (which came originally from the Samba project). I have added a test to the
4946 spa_base64_to_bits() function which I hope fixes it.
4949 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
4950 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
4952 63. The daemon start-up calls getloadavg() while still root for those OS that
4953 need the first call to be done as root, but it missed one case: when
4954 deliver_queue_load_max is set with deliver_drop_privilege. This is
4955 necessary for the benefit of the queue runner, because there is no re-exec
4956 when deliver_drop_privilege is set.
4958 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
4959 This has been fixed.
4961 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4962 was in use, was not putting the data itself into the right store pool;
4963 consequently, it could be overwritten for a subsequent message in the same
4964 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4967 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
4969 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
4970 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
4971 the function were treating the return as a boolean value, which happened to
4972 work because 0=false and not-0=true, but is not correct code.
4974 68. The host_aton() function was not handling scoped IPv6 addresses (those
4975 with, for example, "%eth0" on the end) correctly.
4977 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
4978 negated items (that is, ~something) in unsigned ints. Some compilers
4979 apparently mutter when there is no cast.
4981 70. If an address verification called from an ACL failed, and did not produce a
4982 user-specific message (i.e. there was only a "system" message), nothing was
4983 put in $acl_verify_message. In this situation, it now puts the system
4986 71. Change 4.23/11 added synchronization checking at the start of an SMTP
4987 session; change 4.31/43 added the unwanted input to the log line - except
4988 that it did not do this in the start of session case. It now does.
4990 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
4991 This is wrong and can cause the other end to generate a synchronization
4992 error if it is another Exim or anything else that does the synchronization
4993 check. A QUIT command is no longer sent after a timeout.
4995 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
4996 during host lookups.
4998 74. The defer_ok option of callout verification was not working if it was used
4999 when verifying addresses in header lines, that is, for this case:
5001 verify = header_sender/callout=defer_ok
5003 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
5004 those file descriptors could be used for SMTP connections. If anything
5005 wrote to stderr (the example that came up was "warn" in embedded Perl), it
5006 could be sent to the SMTP client, causing chaos. The daemon now opens
5007 stdin, stdout, and stderr to /dev/null when it puts itself into the
5010 76. Arrange for output from Perl's "warn" command to be written to Exim's main
5011 log by default. The user can override this with suitable Perl magic.
5013 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
5014 the log message when discard triggers, was not working for the DATA ACL or
5015 for the non-SMTP ACL.
5017 78. Error message wording change in sieve.c.
5019 79. If smtp_accept_max_per_host was set, the number of connections could be
5020 restricted to fewer than expected, because the daemon was trying to set up
5021 a new connection before checking whether the processes handling previous
5022 connections had finished. The check for completed processes is now done
5023 earlier. On busy systems, this bug wouldn't be noticed because something
5024 else would have woken the daemon, and it would have reaped the completed
5027 80. If a message was submitted locally by a user whose login name contained one
5028 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
5029 It caused a spool format error. I have fixed the spool reading code. A
5030 related problem was that the "from" clause in the Received: line became
5031 illegal because of the space(s). It is now covered by ${quote_local_part.
5033 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
5036 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
5038 83. Patch from Sieve maintainer to fix unterminated string problem in
5039 "vacation" handling.
5041 84. Some minor changes to the Linux configuration files to help with other
5042 OS variants using glibc.
5044 85. One more patch for Sieve to update vacation handling to latest spec.
5047 ----------------------------------------------------
5048 See the note above about the 4.44 and 4.50 releases.
5049 ----------------------------------------------------
5055 1. Change 4.43/35 introduced a bug that caused file counts to be
5056 incorrectly computed when quota_filecount was set in an appendfile
5059 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
5060 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
5063 3. Give more explanation in the error message when the command for a transport
5064 filter fails to execute.
5066 4. There are several places where Exim runs a non-Exim command in a
5067 subprocess. The SIGUSR1 signal should be disabled for these processes. This
5068 was being done only for the command run by the queryprogram router. It is
5069 now done for all such subprocesses. The other cases are: ${run, transport
5070 filters, and the commands run by the lmtp and pipe transports.
5072 5. Some older OS have a limit of 256 on the maximum number of file
5073 descriptors. Exim was using setrlimit() to set 1000 as a large value
5074 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
5075 systems. I've change it so that if it can't get 1000, it tries for 256.
5077 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
5078 was an oversight, and furthermore, ever since the addition of extra
5079 controls (e.g. 4.43/32), the checks on when to allow different forms of
5080 "control" were broken. There should now be diagnostics for all cases when a
5081 control that does not make sense is encountered.
5083 7. $recipients is now available in the predata ACL (oversight).
5085 8. Tidy the search cache before the fork to do a delivery from a message
5086 received from the command line. Otherwise the child will trigger a lookup
5087 failure and thereby defer the delivery if it tries to use (for example) a
5088 cached ldap connection that the parent has called unbind on.
5090 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
5091 of $address_data from the recipient verification was clobbered by the
5092 sender verification.
5094 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
5095 was its contents. (It was OK if the option was not defined at all.)
5097 11. A "Completed" log line is now written for messages that are removed from
5098 the spool by the -Mrm option.
5100 12. $host_address is now set to the target address during the checking of
5101 ignore_target_hosts.
5103 13. When checking ignore_target_hosts for an ipliteral router, no host name was
5104 being passed; this would have caused $sender_host_name to have been used if
5105 matching the list had actually called for a host name (not very likely,
5106 since this list is usually IP addresses). A host name is now passed as
5109 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
5110 code that specifies a non-restarting handler (typically sigaction() in
5111 modern systems) in an attempt to fix a rare and obscure crash bug.
5113 15. Narrowed the window for a race in the daemon that could cause it to ignore
5114 SIGCHLD signals. This is not a major problem, because they are used only to
5115 wake it up if nothing else does.
5117 16. A malformed maildirsize file could cause Exim to calculate negative values
5118 for the mailbox size or file count. Odd effects could occur as a result.
5119 The maildirsize information is now recalculated if the size or filecount
5122 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
5123 support for a long time. Removed HAVE_SYS_VFS_H.
5125 18. Updated exipick to current release from John Jetmore.
5127 19. Allow an empty sender to be matched against a lookup in an address list.
5128 Previously the only cases considered were a regular expression, or an
5131 20. Exim went into a mad DNS lookup loop when doing a callout where the
5132 host was specified on the transport, if the DNS lookup yielded more than
5135 21. The RFC2047 encoding function was originally intended for short strings
5136 such as real names; it was not keeping to the 75-character limit for
5137 encoded words that the RFC imposes. It now respects the limit, and
5138 generates multiple encoded words if necessary. To be on the safe side, I
5139 have increased the buffer size for the ${rfc2047: expansion operator from
5142 22. Failure to deliver a bounce message always caused it to be frozen, even if
5143 there was an errors_to setting on the router. The errors_to setting is now
5146 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
5147 canonical form (fully expanded) before being placed in
5148 $sender_host_address.
5150 24. Updated eximstats to version 1.33
5152 25. Include certificate and key file names in error message when GnuTLS fails
5153 to set them up, because the GnuTLS error message doesn't include the name
5154 of the failing file when there is a problem reading it.
5156 26. Expand error message when OpenSSL has problems setting up cert/key files.
5159 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
5160 (this can affect the format of dates).
5162 28. exim_tidydb, when checking for the continued existence of a message for
5163 which it has found a message-specific retry record, was not finding
5164 messages that were in split spool directories. Consequently, it was
5165 deleting retry records that should have stayed in existence.
5167 29. eximstats updated to version 1.35
5168 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
5169 1.35 - bugfix such that pie charts by volume are generated correctly
5171 30. The SPA authentication driver was not abandoning authentication and moving
5172 on to the next authenticator when an expansion was forced to fail,
5173 contradicting the general specification for all authenticators. Instead it
5174 was generating a temporary error. It now behaves as specified.
5176 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
5177 (the order specifies the preference for clients). The order is now AES256,
5178 AES128, 3DES, ARCFOUR128.
5180 31. Small patch to Sieve code - explicitly set From: when generating an
5183 32. Exim crashed if a remote delivery caused a very long error message to be
5184 recorded - for instance if somebody sent an entire SpamAssassin report back
5185 as a large number of 550 error lines. This bug was coincidentally fixed by
5186 increasing the size of one of Exim's internal buffers (big_buffer) that
5187 happened as part of the Exiscan merge. However, to be on the safe side, I
5188 have made the code more robust (and fixed the comments that describe what
5191 33. Some experimental protocols are using DNS PTR records for new purposes. The
5192 keys for these records are domain names, not reversed IP addresses. The
5193 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
5194 leaves it alone. Component reversal etc. now happens only for IP addresses.
5197 34. The host_aton() function is supposed to be passed a string that is known
5198 to be a valid IP address. However, in the case of IPv6 addresses, it was
5199 not checking this. This is a hostage to fortune. Exim now panics and dies
5200 if the condition is not met. A case was found where this could be provoked
5201 from a dnsdb PTR lookup with an IPv6 address that had more than 8
5202 components; fortuitously, this particular loophole had already been fixed
5203 by change 4.50/55 or 4.44/33 above.
5205 If there are any other similar loopholes, the new check in host_aton()
5206 itself should stop them being exploited. The report I received stated that
5207 data on the command line could provoke the exploit when Exim was running as
5208 exim, but did not say which command line option was involved. All I could
5209 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
5210 running as the user.
5213 35. There was a buffer overflow vulnerability in the SPA authentication code
5214 (which came originally from the Samba project). I have added a test to the
5215 spa_base64_to_bits() function which I hope fixes it.
5218 36. The daemon start-up calls getloadavg() while still root for those OS that
5219 need the first call to be done as root, but it missed one case: when
5220 deliver_queue_load_max is set with deliver_drop_privilege. This is
5221 necessary for the benefit of the queue runner, because there is no re-exec
5222 when deliver_drop_privilege is set.
5224 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
5225 was in use, was not putting the data itself into the right store pool;
5226 consequently, it could be overwritten for a subsequent message in the same
5227 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
5230 38. Sometimes the final signoff response after QUIT could fail to get
5231 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
5232 < 0 before doing a fflush(). This bug looks as though it goes back to the
5233 introduction of TLS in release 3.20, but "sometimes" must have been rare
5234 because the tests only now provoked it.
5240 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
5241 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
5242 Now it returns an integer. A number of calls were still expecting a T/F
5243 return. Fortuitously, in all cases, the tests worked in OK situations,
5244 which is the norm. However, things would have gone wrong on any write
5245 failures on the smtp file descriptor. This function is used when sending
5246 messages over SMTP and also when doing verify callouts.
5248 2. When Exim is called to do synchronous delivery of a locally submitted
5249 message (the -odf or -odi options), it no longer closes stderr before doing
5252 3. Implemented the mua_wrapper option.
5254 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
5256 5. Implemented the functions header_remove(), header_testname(),
5257 header_add_at_position(), and receive_remove_recipient(), and exported them
5260 6. If an ACL "warn" statement specified the addition of headers, Exim already
5261 inserted X-ACL-Warn: at the start if there was no header name. However, it
5262 was not making this test for the second and subsequent header lines if
5263 there were newlines in the string. This meant that an invalid header could
5264 be inserted if Exim was badly configured.
5266 7. Allow an ACL "warn" statement to add header lines at the start or after all
5267 the Received: headers, as well as at the end.
5269 8. Added the rcpt_4xx retry error code.
5271 9. Added postmaster_mailfrom=xxx to callout verification option.
5273 10. Added mailfrom=xxxx to the callout verification option, for verify=
5276 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
5277 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
5279 12. Inserted some casts to stop certain compilers warning when using pointer
5280 differences as field lengths or precisions in printf-type calls (mostly
5281 affecting debugging statements).
5283 13. Added optional readline() support for -be (dynamically loaded).
5285 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
5286 same clock tick as a message's arrival, so that its received time was the
5287 same as the "first fail" time on the retry record, and that message
5288 remained on the queue past the ultimate address timeout, every queue runner
5289 would try a delivery (because it was past the ultimate address timeout) but
5290 after another failure, the ultimate address timeout, which should have then
5291 bounced the address, did not kick in. This was a "< instead of <=" error;
5292 in most cases the first failure would have been in the next clock tick
5293 after the received time, and all would be well.
5295 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
5296 being recognized when the domain list was tested by the match_domain
5297 condition in an expansion string.
5299 16. Added the ${str2b64: operator.
5301 17. Exim was always calling setrlimit() to set a large limit for the number of
5302 processes, without checking whether the existing limit was already
5303 adequate. (It did check for the limit on file descriptors.) Furthermore,
5304 errors from getrlimit() and setrlimit() were being ignored. Now they are
5305 logged to the main and panic logs, but Exim does carry on, to try to do its
5306 job under whatever limits there are.
5308 18. Imported PCRE 5.0.
5310 19. Trivial typo in log message " temporarily refused connection" (the leading
5313 20. If the log selector return_path_on_delivery was set and an address was
5314 redirected to /dev/null, the delivery process crashed because it assumed
5315 that a return path would always be set for a "successful" delivery. In this
5316 case, the whole delivery is bypassed as an optimization, and therefore no
5319 21. Internal re-arrangement: the function for sending a challenge and reading
5320 a response while authentication was assuming a zero-terminated challenge
5321 string. It's now changed to take a pointer and a length, to allow for
5322 binary data in such strings.
5324 22. Added the cyrus_sasl authenticator (code supplied by MBM).
5326 23. Exim was not respecting finduser_retries when seeking the login of the
5327 uid under which it was called; it was always trying 10 times. (The default
5328 setting of finduser_retries is zero.) Also, it was sleeping after the final
5329 failure, which is pointless.
5331 24. Implemented tls_on_connect_ports.
5333 25. Implemented acl_smtp_predata.
5335 26. If the domain in control=submission is set empty, Exim assumes that the
5336 authenticated id is a complete email address when it generates From: or
5337 Sender: header lines.
5339 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
5340 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
5341 chown and chgrp in /bin and hostname in /usr/bin.
5343 28. Exim was keeping the "process log" file open after each use, just as it
5344 does for the main log. This opens the possibility of it remaining open for
5345 long periods when the USR1 signal hits a daemon. Occasional processlog
5346 errors were reported, that could have been caused by this. Anyway, it seems
5347 much more sensible not to leave this file open at all, so that is what now
5350 29. The long-running daemon process does not normally write to the log once it
5351 has entered its main loop, and it closes the log before doing so. This is
5352 so that log files can straightforwardly be renamed and moved. However,
5353 there are a couple of unusual error situations where the daemon does write
5354 log entries, and I had neglected to close the log afterwards.
5356 30. The text of an SMTP error response that was received during a remote
5357 delivery was being truncated at 512 bytes. This is too short for some of
5358 the long messages that one sometimes sees. I've increased the limit to
5361 31. It is now possible to make retry rules that apply only when a message has a
5362 specific sender, in particular, an empty sender.
5364 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
5365 it possible to be selective about when SMTP synchronization is enforced.
5367 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
5369 32. Implemented hosts_connection_nolog.
5371 33. Added an ACL for QUIT.
5373 34. Setting "delay_warning=" to disable warnings was not working; it gave a
5376 35. Added mailbox_size and mailbox_filecount to appendfile.
5378 36. Added control = no_multiline_responses to ACLs.
5380 37. There was a bug in the logic of the code that waits for the clock to tick
5381 in the case where the clock went backwards by a substantial amount such
5382 that the microsecond fraction of "now" was more than the microsecond
5383 fraction of "then" (but the whole seconds number was less).
5385 38. Added support for the libradius Radius client library this is found on
5386 FreeBSD (previously only the radiusclient library was supported).
5392 1. When certain lookups returned multiple values in the form name=value, the
5393 quoting of the values was not always being done properly. Specifically:
5394 (a) If the value started with a double quote, but contained no whitespace,
5396 (b) If the value contained whitespace other than a space character (i.e.
5397 tabs or newlines or carriage returns) it was not quoted.
5398 This fix has been applied to the mysql and pgsql lookups by writing a
5399 separate quoting function and calling it from the lookup code. The fix
5400 should probably also be applied to nisplus, ibase and oracle lookups, but
5401 since I cannot test any of those, I have not disturbed their existing code.
5403 2. A hit in the callout cache for a specific address caused a log line with no
5404 reason for rejecting RCPT. Now it says "Previous (cached) callout
5405 verification failure".
5407 3. There was an off-by-one bug in the queryprogram router. An over-long
5408 return line was truncated at 256 instead of 255 characters, thereby
5409 overflowing its buffer with the terminating zero. As well as fixing this, I
5410 have increased the buffer size to 1024 (and made a note to document this).
5412 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
5413 when Exim is waiting for an SMTP response from a remote server, Exim
5414 restarts its select() call on the socket, thereby resetting its timeout.
5415 This is not a problem when such interrupts are rare. Somebody set up a cron
5416 job to run exiwhat every 2 minutes, which is less than the normal select()
5417 timeout (5 or 10 minutes). This meant that the select() timeout never
5418 kicked in because it was always reset. I have fixed this by comparing the
5419 time when an interrupt arrives with the time at the start of the first call
5420 to select(). If more time than the timeout has elapsed, the interrupt is
5421 treated as a timeout.
5423 5. Some internal re-factoring in preparation for the addition of Sieve
5424 extensions (by MH). In particular, the "personal" test is moved to a
5425 separate function, and given an option for scanning Cc: and Bcc: (which is
5426 not set for Exim filters).
5428 6. When Exim created an email address using the login of the caller as the
5429 local part (e.g. when creating a From: or Sender: header line), it was not
5430 quoting the local part when it contained special characters such as @.
5432 7. Installed new OpenBSD configuration files.
5434 8. Reworded some messages for syntax errors in "and" and "or" conditions to
5435 try to make them clearer.
5437 9. Callout options, other than the timeout value, were being ignored when
5438 verifying sender addresses in header lines. For example, when using
5440 verify = header_sender/callout=no_cache
5442 the cache was (incorrectly) being used.
5444 10. Added a missing instance of ${EXE} to the exim_install script; this affects
5445 only the Cygwin environment.
5447 11. When return_path_on_delivery was set as a log selector, if different remote
5448 addresses in the same message used different return paths and parallel
5449 remote delivery occurred, the wrong values would sometimes be logged.
5450 (Whenever a remote delivery process finished, the return path value from
5451 the most recently started remote delivery process was logged.)
5453 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
5454 lines when AUTH and/or TLS are in use. This is the "received protocol"
5455 field. Exim used to use "asmtp" for authenticated SMTP, without any
5456 indication (in the protocol name) for TLS use. Now it follows the RFC and
5457 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
5458 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
5459 names appear in log lines as well as in Received: header lines.
5461 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
5462 extensions, and comparison tests, and to fix some bugs.
5464 14. Changes to the "personal" filter test:
5466 (1) The test was buggy in that it was just doing the equivalent of
5467 "contains" tests on header lines. For example, if a user's address was
5468 anne@some.where, the "personal" test would incorrectly be true for
5470 To: susanne@some.where
5472 This test is now done by extracting each address from the header in turn,
5473 and checking the entire address. Other tests that are part of "personal"
5474 are now done using regular expressions (for example, to check local parts
5475 of addresses in From: header lines).
5477 (2) The list of non-personal local parts in From: addresses has been
5478 extended to include "listserv", "majordomo", "*-request", and "owner-*",
5479 taken from the Sieve specification recommendations.
5481 (3) If the message contains any header line starting with "List-" it is
5482 treated as non-personal.
5484 (4) The test for "circular" in the Subject: header line has been removed
5485 because it now seems ill-conceived.
5487 15. Minor typos in src/EDITME comments corrected.
5489 16. Installed latest exipick from John Jetmore.
5491 17. If headers_add on a router specified a text string that was too long for
5492 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
5493 of string_sprintf() is now avoided.
5495 18. $message_body_size was not set (it was always zero) when running the DATA
5496 ACL and the local_scan() function.
5498 19. For the "mail" command in an Exim filter, no default was being set for
5499 the once_repeat time, causing a random time value to be used if "once" was
5500 specified. (If the value happened to be <= 0, no repeat happened.) The
5501 default is now 0s, meaning "never repeat". The "vacation" command was OK
5502 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
5503 (I found it when inspecting the code).
5505 20. There is now an overall timeout for performing a callout verification. It
5506 defaults to 4 times the callout timeout, which applies to individual SMTP
5507 commands during the callout. The overall timeout applies when there is more
5508 than one host that can be tried. The timeout is checked before trying the
5509 next host. This prevents very long delays if there are a large number of
5510 hosts and all are timing out (e.g. when the network connections are timing
5511 out). The value of the overall timeout can be changed by specifying an
5512 additional sub-option for "callout", called "maxwait". For example:
5514 verify = sender/callout=5s,maxwait=20s
5516 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
5517 to the end before writing, but this should make it even safer).
5519 22. Exim was forgetting that it had advertised PIPELINING for the second and
5520 subsequent messages on an SMTP connection. It was also not resetting its
5521 memory on STARTTLS and an internal HELO.
5523 23. When Exim logs an SMTP synchronization error within a session, it now
5524 records whether PIPELINING has been advertised or not.
5526 24. Added 3 instances of "(long int)" casts to time_t variables that were being
5527 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
5528 rather than long int.
5530 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
5532 26. Added the never_mail option to autoreply.
5538 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
5539 crash if the getsockname() call failed; this can happen if a connection is
5540 closed very soon after it is established. The problem was simply in the
5541 order in which certain operations were done, causing Exim to try to write
5542 to the SMTP stream before it had set up the file descriptor. The bug has
5543 been fixed by making things happen in the correct order.
5549 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
5550 before the connection was closed, thus losing the rejection response.
5552 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
5553 some early Solaris releases, but causes trouble in current releases where
5554 socklen_t is defined.
5556 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
5559 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
5562 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
5563 that releasing the top of it at the end releases what was used for sub-
5564 expansions (unless the block got too big). However, discard this block if
5565 the first thing is a variable or header, so that we can use its block when
5566 it is dynamic (useful for very large $message_headers, for example).
5568 6. Lookups now cache *every* query, not just the most recent. A new, separate
5569 store pool is used for this. It can be recovered when all lookup caches are
5570 flushed. Lookups now release memory at the end of their result strings.
5571 This has involved some general refactoring of the lookup sources.
5573 7. Some code has been added to the store_xxx() functions to reduce the amount
5574 of flapping under certain conditions.
5576 8. log_incoming_interface used to affect only the <= reception log lines. Now
5577 it causes the local interface and port to be added to several more SMTP log
5578 lines, for example "SMTP connection from", and rejection lines.
5580 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
5582 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
5584 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
5585 could be overwritten at the end of the current message (or the start of a
5586 new message if it was set in a HELO ACL). The value is now preserved for
5587 the duration of the SMTP connection.
5589 12. If a transport had a headers_rewrite setting, and a matching header line
5590 contained an unqualified address, that address was qualified, even if it
5591 did not match any rewriting rules. The underlying bug was that the values
5592 of the flags that permit the existence of unqualified sender and recipient
5593 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
5594 non-local messages, and by -bnq for local messages) were not being
5595 preserved with the message after it was received.
5597 13. When Exim was logging an SMTP synchronization error, it could sometimes log
5598 "next input=" as part of the text comprising the host identity instead of
5599 the correct text. The code was using the same buffer for two different
5600 strings. However, depending on which order the printing function evaluated
5601 its arguments, the bug did not always show up. Under Linux, for example, my
5602 test suite worked just fine.
5604 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
5605 doesn't work with some older versions of Perl. It has been changed to "my",
5606 which in any case is probably the better facility to use.
5608 15. A really picky compiler found some instances of statements for creating
5609 error messages that either had too many or two few arguments for the format
5612 16. The size of the buffer for calls to the DNS resolver has been increased
5613 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
5614 for addresses that have a lot of PTR records. This alleviates a problem; it
5615 does not fully solve it.
5617 17. A dnsdb lookup for PTR records that receives more data than will fit in the
5618 buffer now truncates the list and logs the incident, which is the same
5619 action as happens when Exim is looking up a host name and its aliases.
5620 Previously in this situation something unpredictable would happen;
5621 sometimes it was "internal error: store_reset failed".
5623 18. If a server dropped the connection unexpectedly when an Exim client was
5624 using GnuTLS and trying to read a response, the client delivery process
5625 crashed while trying to generate an error log message.
5627 19. If a "warn" verb in an ACL added multiple headers to a message in a single
5628 string, for example:
5630 warn message = H1: something\nH2: something
5632 the text was added as a single header line from Exim's point of view
5633 though it ended up OK in the delivered message. However, searching for the
5634 second and subsequent header lines using $h_h2: did not work. This has been
5635 fixed. Similarly, if a system filter added multiple headers in this way,
5636 the routers could not see them.
5638 20. Expanded the error message when iplsearch is called with an invalid key to
5639 suggest using net-iplsearch in a host list.
5641 21. When running tests using -bh, any delays imposed by "delay" modifiers in
5642 ACLs are no longer actually imposed (and a message to that effect is
5645 22. If a "gecos" field in a passwd entry contained escaped characters, in
5646 particular, if it contained a \" sequence, Exim got it wrong when building
5647 a From: or a Sender: header from that name. A second bug also caused
5648 incorrect handling when an unquoted " was present following a character
5649 that needed quoting.
5651 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
5652 was not being matched caselessly.
5654 24. Arranged for all hyphens in the exim.8 source to be escaped with
5657 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
5658 properly. Recipient callout cache records were still being keyed to include
5659 the sender, even when use_sender was set false. This led to far more
5660 callouts that were necessary. The sender is no longer included in the key
5661 when use_sender is false.
5663 26. Added "control = submission" modifier to ACLs.
5665 27. Added the ${base62d: operator to decode base 62 numbers.
5667 28. dnsdb lookups can now access SRV records.
5669 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
5670 the configuration file.
5672 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
5673 (-v) mode. This makes the output for a verbose queue run more intelligible.
5675 31. Added a use_postmaster feature to recipient callouts.
5677 32. Added the $body_zerocount variable, containing the number of binary zero
5678 bytes in the message body.
5680 33. The time of last modification of the "new" subdirectory is now used as the
5681 "mailbox time last read" when there is a quota error for a maildir
5684 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
5686 35. Added +ignore_unknown as a special item in host lists.
5688 36. Code for decoding IPv6 addresses in host lists is now included, even if
5689 IPv6 support is not being compiled. This fixes a bug in which an IPv6
5690 address was recognized as an IP address, but was then not correctly decoded
5691 into binary, causing unexpected and incorrect effects when compared with
5698 1. Very minor rewording of debugging text in manualroute to say "list of
5699 hosts" instead of "hostlist".
5701 2. If verify=header_syntax was set, and a header line with an unqualified
5702 address (no domain) and a large number of spaces between the end of the
5703 name and the colon was received, the reception process suffered a buffer
5704 overflow, and (when I tested it) crashed. This was caused by some obsolete
5705 code that should have been removed. The fix is to remove it!
5707 3. When running in the test harness, delay a bit after writing a bounce
5708 message to get a bit more predictability in the log output.
5710 4. Added a call to search_tidyup() just before forking a reception process. In
5711 theory, someone could use a lookup in the expansion of smtp_accept_max_
5712 per_host which, without the tidyup, could leave open a database connection.
5714 5. Added the variables $recipient_data and $sender_data which get set from a
5715 lookup success in an ACL "recipients" or "senders" condition, or a router
5716 "senders" option, similar to $domain_data and $local_part_data.
5718 6. Moved the writing of debug_print from before to after the "senders" test
5721 7. Change 4.31/66 (moving the time when the Received: is generated) caused
5722 problems for message scanning, either using a data ACL, or using
5723 local_scan() because the Received: header was not generated till after they
5724 were called (in order to set the time as the time of reception completion).
5725 I have revised the way this works. The header is now generated after the
5726 body is received, but before the ACL or local_scan() are called. After they
5727 are run, the timestamp in the header is updated.
5733 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
5734 before starting a queue runner without re-exec. This happened only when
5735 deliver_drop_privilege was set or when the Exim user was set to root. The
5736 effect of the bug was that timeouts during subsequent deliveries caused
5737 crashes instead of being properly handled. The handler is now left at its
5738 default (and expected) setting.
5740 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
5741 message, again when deliver_drop_privilege is set or Exim is run as root.
5742 The bug described in (1) was not present in this case, but the tidying up
5743 of the other signals was missing. I have made the two cases consistent.
5745 3. The ignore_target_hosts setting on a manualroute router was being ignored
5746 for hosts that were looked up using the /MX notation.
5748 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
5751 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
5752 operated on the sender address. After changing the $sender_address to <>
5753 for the sender address verify, Exim was re-instated it as the original
5754 (before rewriting) address, but remembering that it had rewritten it, so it
5755 wasn't rewriting it again. This bug also had the effect of breaking the
5756 sender address verification caching when the sender address was rewritten.
5758 6. The ignore_target_hosts option was being ignored by the ipliteral router.
5759 This has been changed so that if the ip literal address matches
5760 ignore_target_hosts, the router declines.
5762 7. Added expansion conditions match_domain, match_address, and match_local_
5763 part (NOT match_host).
5765 8. The placeholder for the Received: header didn't have a length field set.
5767 9. Added code to Exim itself and to exim_lock to test for a specific race
5768 condition that could lead to file corruption when using MBX delivery. The
5769 issue is with the lockfile that is created in /tmp. If this file is removed
5770 after a process has opened it but before that process has acquired a lock,
5771 there is the potential for a second process to recreate the file and also
5772 acquire a lock. This could lead to two Exim processes writing to the file
5773 at the same time. The added code performs the same test as UW imapd; it
5774 checks after acquiring the lock that its file descriptor still refers to
5775 the same named file.
5777 10. The buffer for building added header lines was of fixed size, 8192 bytes.
5778 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
5781 11. Added the smtp_active_hostname option. If used, this will typically be made
5782 to depend on the incoming interface address. Because $interface_address is
5783 not set up until the daemon has forked a reception process, error responses
5784 that can happen earlier (such as "too many connections") no longer contain
5787 12. If an expansion in a condition on a "warn" statement fails because a lookup
5788 defers, the "warn" statement is abandoned, and the next ACL statement is
5789 processed. Previously this caused the whole ACL to be aborted.
5791 13. Added the iplsearch lookup type.
5793 14. Added ident_timeout as a log selector.
5795 15. Added tls_certificate_verified as a log selector.
5797 16. Added a global option tls_require_ciphers (compare the smtp transport
5798 option of the same name). This controls incoming TLS connections.
5800 17. I finally figured out how to make tls_require_ciphers do a similar thing
5801 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
5802 before starting the TLS session.
5804 18. Tabs are now shown as \t in -bP output.
5806 19. If the log selector return_path_on_delivery was set, Exim crashed when
5807 bouncing a message because it had too many Received: header lines.
5809 20. If two routers both had headers_remove settings, and the first one included
5810 a superfluous trailing colon, the final name in the first list and the
5811 first name in the second list were incorrectly joined into one item (with a
5812 colon in the middle).
5818 1. Added -C and -D options to the exinext utility, mainly to make it easier
5819 to include in the automated testing, but these could be helpful when
5820 multiple configurations are in use.
5822 2. The exinext utility was not formatting the output nicely when there was
5823 an alternate port involved in the retry record key, nor when there was a
5824 message id as well (for retries that were specific to a specific message
5825 and a specific host). It was also confused by IPv6 addresses, because of
5826 the additional colons they contain. I have fixed the IPv4 problem, and
5827 patched it up to do a reasonable job for IPv6.
5829 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
5830 delivery, the log line now contains "pipelined" if PIPELINING was used.
5832 4. An SMTP transport process used to panic and die if the bind() call to set
5833 an explicit outgoing interface failed. This has been changed; it is now
5834 treated in the same way as a connect() failure.
5836 5. A reference to $sender_host_name in the part of a conditional expansion
5837 that was being skipped was still causing a DNS lookup. This no longer
5840 6. The def: expansion condition was not recognizing references to header lines
5841 that used bh_ and bheader_.
5843 7. Added the _cache feature to named lists.
5845 8. The code for checking quota_filecount in the appendfile transport was
5846 allowing one more file than it should have been.
5848 9. For compatibility with Sendmail, the command line option
5856 and sets the incoming protocol and host name (for trusted callers). The
5857 host name and its colon can be omitted when only the protocol is to be set.
5858 Note the Exim already has two private options, -pd and -ps, that refer to
5859 embedded Perl. It is therefore impossible to set a protocol value of "d" or
5860 "s", but I don't think that's a major issue.
5862 10. A number of refactoring changes to the code, none of which should affect
5865 (a) The number of logging options was getting close to filling up the
5866 32-bit word that was used as a bit map. I have split them into two classes:
5867 those that are passed in the argument to log_write(), and those that are
5868 only ever tested independently outside of that function. These are now in
5869 separate 32-bit words, so there is plenty of room for expansion again.
5870 There is no change in the user interface or the logging behaviour.
5872 (b) When building, for example, log lines, the code previously used a
5873 macro that called string_cat() twice, in order to add two strings. This is
5874 not really sufficiently general. Furthermore, there was one instance where
5875 it was actually wrong because one of the argument was used twice, and in
5876 one call a function was used. (As it happened, calling the function twice
5877 did not affect the overall behaviour.) The macro has been replaced by a
5878 function that can join an arbitrary number of extra strings onto a growing
5881 (c) The code for expansion conditions now uses a table and a binary chop
5882 instead of a serial search (which was left over from when there were very
5883 few conditions). Also, it now recognizes conditions like "pam" even when
5884 the relevant support is not compiled in: a suitably worded error message is
5885 given if an attempt is made to use such a condition.
5887 11. Added ${time_interval:xxxxx}.
5889 12. A bug was causing one of the ddress fields not to be passed back correctly
5890 from remote delivery subprocesses. The field in question was not being
5891 subsequently used, so this caused to problems in practice.
5893 13. Added new log selectors queue_time and deliver_time.
5895 14. Might have fixed a bug in maildirsizefile handling that threw up
5896 "unexpected character" debug warnings, and recalculated the data
5897 unnecessarily. In any case, I expanded the warning message to give more
5900 15. Added the message "Restricted characters in address" to the statements in
5901 the default ACL that block characters like @ and % in local parts.
5903 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
5904 Three changes have been made:
5906 (a) There was a serious bug; a negative response to MAIL caused the whole
5907 recipient domain to be cached as invalid, thereby blocking all messages
5908 to all local parts at the same domain, from all senders. This bug has
5909 been fixed. The domain is no longer cached after a negative response to
5910 MAIL if the sender used is not empty.
5912 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
5915 (c) A new callout option, "use_sender" has been added for people who want
5916 the modified behaviour.
5922 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
5925 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
5926 indeed breaks things for older releases.
5928 3. Added additional logging to the case where there is a problem reading data
5929 from a filter that is running in a subprocess using a pipe, in order to
5930 try to track down a specific problem.
5932 4. Testing facility fudge: when running in the test harness and attempting
5933 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
5934 getting "No route to host". Convert this to a timeout.
5936 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
5939 6. Some OS don't have socklen_t but use size_t instead. This affects the
5940 fifth argument of getsockopt() amongst other things. This is now
5941 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
5942 can be set for individual OS. I have set it for SunOS5, OSF1, and
5943 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
5944 some earlier ones do not.
5946 7. Change 4.30/15 was not doing the test caselessly.
5948 8. The standard form for an IPv6 address literal was being rejected by address
5949 parsing in, for example, MAIL and RCPT commands. An example of this kind of
5950 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5951 this, as well as the form without the "IPv6" on the front (but only when
5952 address literals are enabled, of course).
5954 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
5956 10. Exim crashed if a message with an empty sender address specified by -f
5957 encountered a router with an errors_to setting. This could be provoked only
5958 by a command such as
5962 where an empty string was supplied; "<>" did not hit this bug.
5964 11. Installed PCRE release 4.5.
5966 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
5967 remained set. It is now erased.
5969 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
5970 times from message ids (which are base 36 rather than the normal 62).
5972 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
5973 were being counted as actual protocol errors, and logged if the log
5974 selector +smtp_protocol_error was set. One cannot be perfect in this test,
5975 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
5976 and DATA following a set of rejected RCPTs do not count as protocol errors.
5977 In other words, Exim assumes they were pipelined, though this may not
5978 actually be the case. Of course, in all cases the client gets an
5979 appropriate error code.
5981 15. If a lookup fails in an ACL condition, a message about the failure may
5982 be available; it is used if testing the ACL cannot continue, because most
5983 such messages specify what the cause of the deferral is. However, some
5984 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
5985 that caused an old message to be retained and used if a later statement
5986 caused a defer, replacing the real cause of the deferral.
5988 16. If an IP address had so many PTR records that the DNS lookup buffer
5989 was not large enough to hold them, Exim could crash while trying to process
5990 the truncated data. It now detects and logs this case.
5992 17. Further to 4.21/58, another change has been made: if (and only if) the
5993 first line of a message (the first header line) ends with CRLF, a bare LF
5994 in a subsequent header line has a space inserted after it, so as not to
5995 terminate the header.
5997 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
5998 unnecessarily, used atoi() instead of strtol(), and didn't check the
5999 termination when getting file sizes from file names by regex.
6001 19. Completely re-implemented the support for maildirsize files, in the light
6002 of a number of problems with the previous contributed implementation
6003 (4.30/29). In particular:
6005 . If the quota is zero, the maildirsize file is maintained, but no quota is
6008 . If the maildir directory does not exist, it is created before any attempt
6009 to write a maildirsize file.
6011 . The quota value in the file is just a cache; if the quota is changed in
6012 the transport, the new value overrides.
6014 . A regular expression is available for excluding directories from the
6017 20. The autoreply transport checks the characters in options that define the
6018 message's headers; it allows continued headers, but it was checking with
6019 isspace() after an embedded newline instead of explicitly looking for a
6022 21. If all the "regular" hosts to which an address was routed had passed their
6023 expiry times, and had not reached their retry times, the address was
6024 bounced, even if fallback hosts were defined. Now Exim should go on to try
6027 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
6028 equivalent code in the SMTP transport. Some hosts send humungous responses
6029 to HELO/EHLO, more than 1024 it seems.
6031 23. Refactoring: code in filter.c used (void *) for "any old type" but this
6032 gives compiler warnings in some environments. I've now done it "properly",
6035 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
6036 (because of problems with the built-in one) was declared to return uschar *
6037 instead of char *, causing compiler failure.
6039 25. Fixed a file descriptor leak when processing alias/forward files.
6041 26. Fixed a minor format string issue in dbfn.c.
6043 27. Typo in exim.c: ("dmbnz" for "dbmnz").
6045 28. If a filter file refered to $h_xxx or $message_headers, and the headers
6046 contained RFC 2047 "words", Exim's memory could, under certain conditions,
6049 29. When a sender address is verified, it is cached, to save repeating the test
6050 when there is more than one recipient in a message. However, when the
6051 verification involves a callout, it is possible for different callout
6052 options to be set for different recipients. It is too complicated to keep
6053 track of this in the cache, so now Exim always runs a verification when a
6054 callout is required, relying on the callout cache for the optimization.
6055 The overhead is duplication of the address routing, but this should not be
6058 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
6059 to be verified with callout=postmaster, and the main callout worked but the
6060 postmaster check failed, the verification correctly failed. However, if a
6061 subsequent RCPT command asked for sender verification *without* the
6062 postmaster check, incorrect caching caused this verification also to fail,
6065 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
6066 it was not caching the DNS options (qualify_single, search_parents) that
6067 were used when the lookup failed. A subsequent lookup with different
6068 options therefore always gave the same answer, though there were cases
6069 where it should not have. (Example: a "domains = !$mx_any" option on a
6070 dnslookup router: the "domains" option is always processed without any
6071 widening, but the router might have qualify_single set.) Now Exim uses the
6072 cached value only when the same options are set.
6074 32. Added John Jetmore's "exipick" utility to the distribution.
6076 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
6077 than a timeout (e.g. a certificate is required, and is not provided), an
6078 Exim server now closes the connection immediately. Previously it waited for
6079 the client to close - but if the client is SSL, it seems that they each
6080 wait for each other, leading to a delay before one of them times out.
6082 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
6083 maintained 0.8.x compatibility because I don't think many are using it, and
6084 it is clearly obsolete.
6086 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
6089 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
6090 client certificate was expired. A simple patch fixes this, though I don't
6091 understand the full logic of why the verify callback is called multiple
6094 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
6095 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
6096 which causes problems with some clients (such as the Certicom SSL Plus
6097 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
6098 disables the coutermeasure allowing Eudora to connect."
6100 38. Exim was not checking that a write() to a log file succeeded. This could
6101 lead to Bad Things if a log got too big, in particular if it hit a file
6102 size limit. Exim now panics and dies if it cannot write to a log file, just
6103 as it does if it cannot open a log file.
6105 39. Modified OS/Makefile-Linux so that it now contains
6107 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
6109 The two -D definitions ensure that Exim is compiled with large file
6110 support, which makes it possible to handle log files that are bigger than
6113 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
6114 instance) a domain was checked against a named list that involved a lookup,
6115 causing $domain_data to be set, then another domain was checked against the
6116 same list, then the first domain was re-checked, the value of $domain_data
6117 after the final check could be wrong. In particular, if the second check
6118 failed, it could be set empty. This bug probably also applied to
6121 41. The strip_trailing_dot option was not being applied to the address given
6122 with the -f command-line option.
6124 42. The code for reading a message's header from the spool was incrementing
6125 $received_count, but never initializing it. This meant that the value was
6126 incorrect (doubled) while delivering a message in the same process in which
6127 it was received. In the most common configuration of Exim, this never
6128 happens - a fresh exec is done - but it can happen when
6129 deliver_drop_privilege is set.
6131 43. When Exim logs an SMTP synchronization error - client data sent too soon -
6132 it now includes up to 150 characters of the unexpected data in the log
6135 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
6136 and building data strings. The size of both of these buffers was 10 000
6137 bytes - far larger than anybody would *ever* want, thought I. Needless to
6138 say, somebody hit the limit. I have increased the maximum line length to
6139 20 000 and the maximum data length of concatenated lines to 100 000. I have
6140 also fixed two bugs, because there was no checking on these buffers. Tsk,
6141 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
6142 buffer is too small.
6144 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
6145 lsearch lookups. Now it does.
6147 46. When parsing a route_list item in a manualroute router, a fixed-length
6148 buffer was used for the list of hosts. I made this 1024 bytes long,
6149 thinking that nobody would ever have a list of hosts that long. Wrong.
6150 Somebody had a whole pile of complicated expansion conditions, and the
6151 string was silently truncated, leading to an expansion error. It turns out
6152 that it is easier to change to an unlimited length (owing to other changes
6153 that have happened since this code was originally written) than to build
6154 structure for giving a limitation error. The length of the item that
6155 expands into the list of hosts is now unlimited.
6157 47. The lsearch lookup could not handle data where the length of text line was
6158 more than 4095 characters. Such lines were truncated, leading to shortened
6159 data being returned. It should now handle lines of any length.
6161 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
6162 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
6165 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
6166 replacing strings such as BIN_DIRECTORY in a source file: the replacement
6167 no longer happens in comment lines. A list of replacements is now placed
6168 at the head of all of the source files, except those whose only change is
6169 to replace PERL_COMMAND in the very first #! line.
6171 50. Replaced the slow insertion sort in queue.c, for sorting the list of
6172 messages on the queue, with a bottom-up merge sort, using code contributed
6173 by Michael Haardt. This should make operations like -bp somewhat faster on
6174 large queues. It won't affect queue runners, except when queue_run_in_order
6177 51. Installed eximstats 1.31 in the distribution.
6179 52. Added support for SRV lookups to the dnslookup router.
6181 53. If an ACL referred to $message_body or $message_body_end, the value was not
6182 reset for any messages that followed in the same SMTP session.
6184 54. The store-handling optimization for building very long strings was not
6185 differentiating between the different store pools. I don't think this
6186 actually made any difference in practice, but I've tidied it.
6188 55. While running the routers to verify a sender address, $sender_address
6189 was still set to the sender address. This is wrong, because when routing to
6190 send a bounce to the sender, it would be empty. Therefore, I have changed
6191 it so that, while verifying a sender address, $sender_address is set to <>.
6192 (There is no change to what happens when verifying a recipient address.)
6194 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
6195 target A or AAAA records (if not already returned) without resetting the
6196 qualify_single or search_parents options of the DNS resolver. These are
6197 inappropriate in this case because the targets of MX and SRV records must
6198 be FQDNs. A broken DNS record could cause trouble if it happened to have a
6199 target that, when qualified, matched something in the local domain. These
6200 two options are now turned off when doing these lookups.
6202 57. It seems that at least some releases of Reiserfs (which does not have the
6203 concept of a fixed number of inodes) returns zero and not -1 for the
6204 number of available inodes. This interacted badly with check_spool_inodes,
6205 which assumed that -1 was the "no such thing" setting. What I have done is
6206 to check that the total number of inodes is greater than zero before doing
6207 the test of how many are available.
6209 58. When a "warn" ACL statement has a log_message modifier, the message is
6210 remembered, and not repeated. This is to avoid a lot of repetition when a
6211 message has many recipients that cause the same warning to be written.
6212 However, Exim was preserving the list of already written lines for an
6213 entire SMTP session, which doesn't seem right. The memory is now reset if a
6214 new message is started.
6216 59. The "rewrite" debugging flag was not showing the result of rewriting in the
6217 debugging output unless log_rewrite was also set.
6219 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
6220 of (int)(handle) when we know that handle contains (void *)(-1).
6222 61. The Exim daemon panic-logs an error return when it closes the incoming
6223 connection. However "connection reset by peer" seems to be common, and
6224 isn't really an error worthy of noting specially, so that particular error
6227 62. When Exim is trying to find all the local interfaces, it used to panic and
6228 die if the ioctl to get the interface flags failed. However, it seems that
6229 on at least one OS (Solaris 9) it is possible to have an interface that is
6230 included in the list of interfaces, but for which you get a failure error
6231 for this call. This happens when the interface is not "plumbed" into a
6232 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
6233 failure of the "get flags" call assumes that the interface is down.
6235 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
6236 makes life easier for people who are doing arithmetic on fields extracted
6237 from dates, where you often get leading zeros that should not be
6238 interpreted as octal.
6240 64. Added qualify_domain to the redirect router, to override the global
6243 65. If a pathologically long header line contained very many addresses (the
6244 report of this problem mentioned 10 000) and each of them was rewritten,
6245 Exim could use up a very large amount of memory. (It kept on making new
6246 copies of the header line as it rewrote, and never released the old ones.)
6247 At the expense of a bit more processing, the header rewriting function has
6248 been changed so that it no longer eats memory in this way.
6250 66. The generation of the Received: header has been moved from the time that a
6251 message starts to be received, to the time that it finishes. The timestamp
6252 in the Received: header should now be very close to that of the <= log
6253 line. There are two side-effects of this change:
6255 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
6256 logged header lines no longer include the local Received: line, because
6257 it has not yet been created. The same applies to a copy of the message
6258 that is returned to a non-SMTP sender when a message is rejected.
6260 (b) When a filter file is tested using -bf, no additional Received: header
6261 is added to the test message. After some thought, I decided that this
6264 This change does not affect the value of $received_for. It is still set
6265 after address rewriting, but before local_scan() is called.
6267 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
6269 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
6270 gave an unhelpful panic error message, and a defer error. I have managed to
6271 change this behaviour so that it now rejects any supplied certificate,
6272 which seems right, as the list of acceptable certificates is empty.
6274 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
6275 gave an unhelpful defer error. I have not managed to make this reject any
6276 supplied certificates, but the error message it gives is "no certificate
6277 supplied", which is not helpful.
6279 70. exigrep's output now also includes lines that are not associated with any
6280 message, but which match the given pattern. Implemented by a patch from
6281 Martin Sluka, which also tidied up the Perl a bit.
6283 71. Recipient callout verification, like sender verification, was using <> in
6284 the MAIL FROM command. This isn't really the right thing, since the actual
6285 sender may affect whether the remote host accepts the recipient or not. I
6286 have changed it to use the actual sender in the callout; this means that
6287 the cache record is now keyed on a recipient/sender pair, not just the
6288 recipient address. There doesn't seem to be a real danger of callout loops,
6289 since a callout by the remote host to check the sender would use <>.
6290 [SEE ABOVE: changed after hitting problems.]
6292 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
6293 temporary errors. However, in the case of such a code being given after
6294 the end of a data transmission (i.e. after ".") Exim was failing to write
6295 a retry record for the message. (Yes, there was some broken host that was
6296 actually sending 8xx at this point.)
6298 73. An unknown lookup type in a host list could cause Exim to panic-die when
6299 the list was checked. (An example that provoked this was putting <; in the
6300 middle of a list instead of at the start.) If this happened during a DATA
6301 ACL check, a -D file could be left lying around. This kind of configuration
6302 error no longer causes Exim to die; instead it causes a defer error. The
6303 incident is still logged to the main and panic logs.
6305 74. Buglet left over from Exim 3 conversion. The message "too many messages
6306 in one connection" was written to the rejectlog but not the mainlog, except
6307 when address rewriting (yes!) was being logged.
6309 75. Added write_rejectlog option.
6311 76. When a system filter was run not as root (that is, when system_filter_user
6312 was set), the values of the $n variables were not being returned to the
6313 main process; thus, they were not subsequently available in the $sn
6316 77. Added +return_path_on_delivery log selector.
6318 78. A connection timeout was being treated differently from recipients deferred
6319 when testing hosts_max_try with a message that was older than the host's
6320 retry timeout. (The host should not be counted, thus allowing all hosts to
6321 be tried at least once before bouncing.) This may have been the cause of an
6322 occasionally reported bug whereby a message would remain on the queue
6323 longer than the retry timeout, but would be bounced if a delivery was
6324 forced. I say "may" because I never totally pinned down the problem;
6325 setting up timeout/retry tests is difficult. See also the next item.
6327 79. The ultimate address timeout was not being applied to errors that involved
6328 a combination of host plus message (for example, a timeout on a MAIL
6329 command). When an address resolved to a number of possible hosts, and they
6330 were not all tried for each delivery (e.g. because of hosts_max_try), a
6331 message could remain on the queue longer than the retry timeout.
6333 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
6336 81. Fixed an obscure SMTP outgoing bug which required at least the following
6337 conditions: (a) there was another message waiting for the same server;
6338 (b) the server returned 5xx to all RCPT commands in the first message so
6339 that the message was not completed; (c) the server dropped the connection
6340 or gave a negative response to the RSET that Exim sends to abort the
6341 transaction. The observed case was a dropped connection after DATA that had
6342 been sent in pipelining mode. That is, the server had advertised PIPELINING
6343 but was not implementing it correctly. The effect of the bug was incorrect
6344 behaviour, such as trying another host, and this could lead to a crash.
6350 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
6351 and daemon.c were passed as pointers to ints; they should have been
6352 pointers to socklen_t variables (which are typically unsigned ints).
6354 2. Some signed/unsigned type warnings in the os.c file for Linux have been
6357 3. Fixed a really odd bug that affected only the testing scheme; patching a
6358 certain fixed string in the binary changed the value of another string that
6359 happened to be identical to the end of the original first string.
6361 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
6362 name", it returns that address as the IP address. On some operating
6363 systems (e.g. Solaris), it also passes back the IP address string as the
6364 "host name". However, on others (e.g. Linux), it passes back an empty
6365 string. Exim wasn't checking for this, and was changing the host name to an
6366 empty string, assuming it had been canonicalized.
6368 5. Although rare, it is permitted to have more than one PTR record for a given
6369 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
6370 all the names associated with an address, because they do in Solaris.
6371 However, it seems that they do not in Linux for data that comes from the
6372 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
6373 I found this out when I moved to a new Linux workstation and tried to run
6374 the Exim test suite.
6376 To get round this problem I have changed the code so that it now does its
6377 own call to the DNS to look up PTR records when searching for a host name.
6378 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
6379 addresses that are only in /etc/hosts are still found.
6381 This behaviour is, however, controlled by an option called host_lookup_
6382 order, which defaults to "bydns:byaddr". If people want to use the other
6383 order, or indeed, just use one or the other means of lookup, they can
6384 specify it in this variable.
6386 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
6387 some operating systems, this comes back from gethostbyaddr() as an empty
6388 string, and this is what Exim used to test for. However, it seems that in
6389 other systems, "." is yielded. Exim now tests for this case too.
6391 7. The values of check_spool_space and check_log_space are now held internally
6392 as a number of kilobytes instead of an absolute number of bytes. If a
6393 numbers is specified without 'K' or 'M', it is rounded up to the nearest
6394 kilobyte. This means that much larger values can be stored.
6396 8. Exim monitor: an attempt to get the action menu when not actually pointing
6397 at a message produces an empty menu entitled "No message selected". This
6398 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
6399 no entries in it ("Shell widget menu has zero width and/or height"). So I
6400 have added a single, blank menu entry in this case.
6402 9. Added ${quote_local_part.
6404 10. MIME decoding is now applied to the contents of Subject: header lines when
6407 11. Now that a reference to $sender_host_address automatically causes a reverse
6408 lookup to occur if necessary (4.13/18), there is no need to arrange for a
6409 host lookup before query-style lookups in lists that might use this
6410 variable. This has therefore been abolished, and the "net-" prefix is no
6411 longer necessary for query-style lookups.
6413 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
6414 have been a typo for LFLAGS, so it has been changed.
6416 13. The install script calls Exim with "-C /dev/null" in order to find the
6417 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
6418 to be output. However, since Exim outputs its version number before the
6419 error, it didn't break the script. It just looked ugly. I fixed this by
6420 always allowing "-C /dev/null" if the caller is root.
6422 14. Ignore overlarge ACL variable number when reading spool file - insurance
6423 against a later release with more variables having written the file.
6425 15. The standard form for an IPv6 address literal was being rejected by EHLO.
6426 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
6427 this, as well as the form without the "IPv6" on the front.
6429 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
6430 OS/Makefile-Darwin file.
6432 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
6433 only to LDAP libraries that do not have LDAP_OPT_DEREF.
6435 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
6436 variable. However, some OS use ints for inodes. I've added cast to long int
6437 to get rid of the compiler warning.
6439 19. I had forgotten to lock out "/../" in configuration file names when
6440 ALT_CONFIG_PREFIX was set.
6442 20. Routers used for verification do not need to specify transports. However,
6443 if such a router generated a host list, and callout was configured, Exim
6444 crashed, because it could not find a port number from the (non-existent)
6445 transport. It now assumes port 25 in this circumstance.
6447 21. Added the -t option to exigrep.
6449 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
6450 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
6451 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
6452 them is set, LOOKUP_LSEARCH is forced.
6454 23. "exim -bV" now outputs a list of lookups that are included in the binary.
6456 24. Added sender and host information to the "rejected by local_scan()" log
6457 line; previously there was no indication of these.
6459 25. Added .include_if_exists.
6461 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
6462 Linux. It turns out that not all file systems support this. Apparently some
6463 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
6464 do it.) To cope with this, the error EINVAL, which means that sync-ing is
6465 not supported on the file descriptor, is now ignored when Exim is trying to
6466 sync a directory. This applies only to Linux.
6468 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
6470 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
6471 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
6474 29. Added support for maildirsize files from supplied patch (modified a bit).
6476 30. The use of :fail: followed by an empty string could lead Exim to respond to
6477 sender verification failures with (e.g.):
6479 550 Verification failed for <xxx>
6480 550 Sender verify failed
6482 where the first response line was missing the '-' that indicates it is not
6483 the final line of the response.
6485 31. The loop for finding the name of the user that called Exim had a hardwired
6486 limit of 10; it now uses the value of finduser_retries, which is used for
6487 all other user lookups.
6489 32. Added $received_count variable, available in data and not_smtp ACLs, and at
6492 33. Exim was neglecting to zero errno before one call of strtol() when
6493 expanding a string and expecting an integer value. On some systems this
6494 resulted in spurious "integer overflow" errors. Also, it was casting the
6495 result into an int without checking.
6497 34. Testing for a connection timeout using "timeout_connect" in the retry rules
6498 did not work. The code looks as if it has *never* worked, though it appears
6499 to have been documented since at least release 1.62. I have made it work.
6501 35. The "timeout_DNS" error in retry rules, also documented since at least
6502 1.62, also never worked. As it isn't clear exactly what this means, and
6503 clearly it isn't a major issue, I have abolished the feature by treating it
6504 as "timeout", and writing a warning to the main and panic logs.
6506 36. The display of retry rules for -brt wasn't always showing the error code
6509 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
6510 timeout_connect_A, timeout_connect_MX.
6512 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
6513 to the empty sender.
6515 39. The daemon was not analysing the content of -oX till after it had closed
6516 stderr and disconnected from the controlling terminal. This meant that any
6517 syntax errors were only noted on the panic log, and the return code from
6518 the command was 0. By re-arranging the code a little, I've made the
6519 decoding happen first, so such errors now appear on stderr, and the return
6520 code is 1. However, the actual setting up of the sockets still happens in
6521 the disconnected process, so errors there are still only recorded on the
6524 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
6525 connections (as happens on some IP stacks) was logged at start up time as
6526 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
6527 it from "IPv6 and IPv4", which means that two separate sockets are being
6530 41. The debug output for gethostbyname2() or getipnodebyname() failures now
6531 says whether AF_INET or AF_INET6 was passed as an argument.
6533 42. Exiwhat output was messed up when time zones were included in log
6536 43. Exiwhat now gives more information about the daemon's listening ports,
6537 and whether -tls-on-connect was used.
6539 44. The "port" option of the smtp transport is now expanded.
6541 45. A "message" modifier in a "warn" statement in a non-message ACL was being
6542 silently ignored. Now an error message is written to the main and panic
6545 46. There's a new ACL modifier called "logwrite" which writes to a log file
6546 as soon as it is encountered.
6548 47. Added $local_user_uid and $local_user_gid at routing time.
6550 48. Exim crashed when trying to verify a sender address that was being
6553 49. Exim was recognizing only a space character after ".include". It now also
6554 recognizes a tab character.
6556 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
6557 extracting the relevant information from the specification. The man page no
6558 longer contains scrambled data for the -d option, and I've added a section
6559 at the front about calling Exim under different names.
6561 51. Added "extra_headers" argument to the "mail" command in filter files.
6563 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
6566 53. Installed eximstats 1.29.
6568 54. Added transport_filter_timeout as a generic transport option.
6570 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
6571 Cc: header lines. This was required by RFC 822, but it not required by RFC
6574 56. Exim used to add From:, Date:, and Message-Id: header lines to any
6575 incoming messages that did not have them. Now it does so only if the
6576 message originates locally, that is, if there is no associated remote host
6577 address. When Resent- header lines are present, this applies to the Resent-
6578 lines rather than the non-Resent- lines.
6580 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
6581 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
6583 58. Messages for configuration errors now include the name of the main
6584 configuration file - useful now that there may be more than one file in a
6585 list (.included file names were always shown).
6587 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
6588 for those rare installations that do not start the daemon as root or run it
6589 setuid root. I've cut out the call to initgroups() if the daemon is not
6592 60. The Exim user and group can now be bound into the binary as text strings
6593 that are looked up at the start of Exim's processing.
6595 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
6597 62. Added $mailstore_basename variable.
6599 63. Installed patch to sieve.c from Michael Haardt.
6601 64. When Exim failed to open the panic log after failing to open the main log,
6602 the original message it was trying to log was written to stderr and debug
6603 output, but if they were not available (the usual case in production), it
6604 was lost. Now it is written to syslog before the two lines that record the
6605 failures to open the logs.
6607 65. Users' Exim filters run in subprocesses under the user's uid. It is
6608 possible for a "deliver" command or an alias in a "personal" command to
6609 provoke an address rewrite. If logging of address rewriting is configured,
6610 this fails because the process is not running as root or exim. There may be
6611 a better way of dealing with this, but for the moment (because 4.30 needs
6612 to be released), I have disabled address rewrite logging when running a
6613 filter in a non-root, non-exim process.
6619 1. The buildconfig auxiliary program wasn't quoting the value set for
6620 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
6621 not defined. This bug was masked in 4.22 by the effect that was fixed in
6624 2. Some messages that were rejected after a message id was allocated were
6625 shown as "incomplete" by exigrep. It no longer does this for messages that
6626 are rejected by local_scan() or the DATA or non-SMTP ACLs.
6628 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
6629 have allow_domain_literals set, the ID did not get logged in the <= line.
6630 Domain literals are now always recognized in Message-ID: header lines.
6632 4. The first argument for a ${extract expansion item is the key name or field
6633 number. Leading and trailing spaces in this item were not being ignored,
6634 causing some misleading effects.
6636 5. When deliver_drop_privilege was set, single queue runner processes started
6637 manually (i.e. by the command "exim -q") or by the daemon (which uses the
6638 same command in the process it spins off) were not dropping privilege.
6640 6. When the daemon running as "exim" started a queue runner, it always
6641 re-executed Exim in the spun-off process. This is a waste of effort when
6642 deliver_drop_privilege is set. The new process now just calls the
6643 queue-runner function directly.
6649 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
6652 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
6653 ignored. Though the use of -oP was forcing the writing of a pid file, it
6654 was always written to the default place.
6656 3. If the message "no IP address found for host xxxx" is generated during
6657 incoming verification, it is now followed by identification of the incoming
6658 connection (so you can more easily find what provoked it).
6660 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
6662 5. Added some features to "harden" Exim a bit more against certain attacks:
6664 (a) There is now a build-time option called FIXED_NEVER_USERS that can
6665 be put in Local/Makefile. This is like the never_users runtime option,
6666 but it cannot be overridden. The default setting is "root".
6668 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
6669 prefix string with which any file named in a -C command line option
6672 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
6673 is retained for -C and -D only if the caller of Exim is root. Without
6674 it, the exim user may also use -C and -D and retain privilege.
6676 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
6677 command line option is disabled.
6679 6. Macro names set by the -D option must start with an upper case letter, just
6680 like macro names defined in the configuration file.
6682 7. Added "dereference=" facility to LDAP.
6684 8. Two instances of the typo "uknown" in the source files are fixed.
6686 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
6687 the Configure-Makefile script screwed up while processing it.
6689 10. Incorporated PCRE 4.4.
6691 11. The SMTP synchronization check was not operating right at the start of an
6692 SMTP session. For example, it could not catch a HELO sent before the client
6693 waited for the greeting. There is now a check for outstanding input at the
6694 point when the greeting is written. Because of the duplex, asynchronous
6695 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
6696 way, but not yet received, when the check is performed.
6698 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
6699 on TCP/IP sockets, because this apparently causes some broken clients to
6702 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
6703 unchanged) from the Cygwin maintainer.
6705 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
6706 was supported instead of testing for mbx. Effectively a typo.
6708 15. The spa authenticator server code was not checking that the input it
6709 received was valid base64.
6711 16. The debug output line for the "set" modifier in ACLs was not showing the
6712 name of the variable that was being set.
6714 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
6716 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
6717 lookup on its own. Something else was needed to trigger the lookup. For
6718 example, a match in host_lookup or the need for a host name in a host list.
6719 Now, if $sender_host_name is referenced and the host name has not yet been
6720 looked up, a lookup is performed. If the lookup fails, the variable remains
6721 empty, and $host_lookup_failed is set to "1".
6723 19. Added "eqi" as a case-independent comparison operator.
6725 20. The saslauthd authentication condition could segfault if neither service
6726 nor realm was specified.
6728 21. If an overflowing value such as "2048M" was set for message_size_limit, the
6729 error message that was logged was misleading, and incoming SMTP
6730 connections were dropped. The message is now more accurate, and temporary
6731 errors are given to SMTP connections.
6733 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
6734 (except RSET) with a 421 error, until QUIT is received. However, it was
6735 failing to send a response to QUIT.
6737 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
6738 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
6739 ACL is now run after the helo_try_verify_hosts code.
6741 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
6742 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
6743 case-independent, so other case variants are also recognized). Apparently
6744 some systems use these upper case variants.
6746 25. If more than two messages were waiting for the same host, and a transport
6747 filter was specified for the transport, Exim sent two messages over the
6748 same TCP/IP connection, and then failed with "socket operation on non-
6749 socket" when it tried to send the third.
6751 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
6753 27. The extern definition of crypt16() in expand.c was not being excluded when
6754 the OS had its own crypt16() function.
6756 28. Added bounce_return_body as a new option, and bounce_return_size_limit
6757 as a preferred synonym for return_size_limit, both as an option and as an
6760 29. Added LIBS=-liconv to OS/Makefile-OSF1.
6762 30. Changed the default configuration ACL to relax the local part checking rule
6763 for addresses that are not in any local domains. For these addresses,
6764 slashes and pipe symbols are allowed within local parts, but the sequence
6765 /../ is explicitly forbidden.
6767 31. SPA server authentication was not clearing the challenge buffer before
6770 32. log_message in a "warn" ACL statement was writing to the reject log as
6771 well as to the main log, which contradicts the documentation and doesn't
6772 seem right (because no rejection is happening). So I have stopped it.
6774 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
6775 However, I am unable to do any testing of this.
6777 34. Fixed an infelicity in the appendfile transport. When checking directories
6778 for a mailbox, to see if any needed to be created, it was accidentally
6779 using path names with one or more superfluous leading slashes; tracing
6780 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
6782 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
6783 contents are added to the log line that is written for every discarded
6784 recipient. (Previously a log_message setting was ignored.)
6786 36. The ${quote: operator now quotes the string if it is empty.
6788 37. The install script runs exim in order to find its version number. If for
6789 some reason other than non-existence or emptiness, which it checks, it
6790 could not run './exim', it was installing it with an empty version number,
6791 i.e. as "exim-". This error state is now caught, and the installation is
6794 38. An argument was missing from the function that creates an error message
6795 when Exim fails to connect to the socket for saslauthd authentication.
6796 This could cause Exim to crash, or give a corrupted message.
6798 39. Added isip, isip4, and isip6 to ${if conditions.
6800 40. The ACL variables $acl_xx are now saved with the message, and can be
6801 accessed later in routers, transports, and filters.
6803 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
6804 strings in the file are not string-expanded.
6806 42. If a MAIL command specified a SIZE value that was too large to fit into an
6807 int variable, the check against message_size_limit failed. Such values are
6808 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
6809 day this will have to be increased, but I don't think I want to be around
6810 when emails are that large.
6817 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
6818 iconv() is not standard in FreeBSD.
6820 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
6821 IPv6 enabled. The observed symptom was a segmentation fault on return from
6822 the function os_common_find_running_interfaces() in src/os.c.
6824 3. In the check_special_case() function in daemon.c I had used "errno" as an
6825 argument name, which causes warnings on some systems. This was basically a
6826 typo, since it was named "eno" in the comments!
6828 4. The code that waits for the clock to tick (at a resolution of some fraction
6829 of a second) so as to ensure message-id uniqueness was always waiting for
6830 at least one whole tick, when it could have waited for less. [This is
6831 almost certainly not relevant at current processor speeds, where it is
6832 unlikely to ever wait at all. But we try to future-proof.]
6834 5. The function that sleeps for a time interval that includes fractions of a
6835 second contained a race. It did not block SIGALRM between setting the
6836 timer, and suspending (a couple of lines later). If the interval was short
6837 and the sigsuspend() was delayed until after it had expired, the suspension
6838 never ended. On busy systems this could lead to processes getting stuck for
6841 6. Some uncommon configurations may cause a lookup to happen in a queue runner
6842 process, before it forks any delivery processes. The open lookup caching
6843 mechanism meant that the open file or database connection was passed into
6844 the delivery process. The problem was that delivery processes always tidy
6845 up cached lookup data. This could cause a problem for the next delivery
6846 process started by the queue runner, because the external queue runner
6847 process does not know about the closure. So the next delivery process
6848 still has data in the lookup cache. In the case of a file lookup, there was
6849 no problem because closing a file descriptor in a subprocess doesn't affect
6850 the parent. However, if the lookup was caching a connection to a database,
6851 the connection was closed, and the second delivery process was likely to
6852 see errors such as "PGSQL: query failed: server closed the connection
6853 unexpectedly". The problem has been fixed by closing all cached lookups
6854 in a queue runner before running a delivery process.
6856 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
6857 seem to have the "const" qualifier which it has on other OS. I've
6860 8. Change 4.21/2 was too strict. It is only if there are two authenticators
6861 *of the same type* (client or server) with the same public name that an
6862 error should be diagnosed.
6864 9. When Exim looked up a host name for an IP address, but failed to find the
6865 original IP address when looking up the host name (a safety check), it
6866 output the message "<ip address> does not match any IP for NULL", which was
6867 confusing, to say the least. The bug was that the host name should have
6868 appeared instead of "NULL".
6870 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
6871 user that is built into the binary, and the -C or -D options is used, root
6872 privilege is dropped before the configuration file is read. In addition,
6873 logging is switched to stderr instead of the normal log files. If the
6874 configuration then re-defines the Exim user, the unprivileged environment
6875 is probably not what is expected, so Exim logs a panic warning message (but
6878 However, if deliver_drop_privilege is set, the unprivileged state may well
6879 be exactly what is intended, so the warning has been cut out in that case,
6880 and Exim is allowed to try to write to its normal log files.
6886 1. smtp_return_error_details was not giving details for temporary sender
6887 or receiver verification errors.
6889 2. Diagnose a configuration error if two authenticators have the same public
6892 3. Exim used not to create the message log file for a message until the first
6893 delivery attempt. This could be confusing when incoming messages were held
6894 for policy or load reasons. The message log file is now created at the time
6895 the message is received, and an initial "Received" line is written to it.
6897 4. The automatically generated man page for command line options had a minor
6898 bug that caused no ill effects; however, a more serious problem was that
6899 the procedure for building the man page automatically didn't always
6900 operate. Consequently, release 4.20 contains an out-of-date version. This
6901 shouldn't happen again.
6903 5. When building Exim with embedded Perl support, the script that builds the
6904 Makefile was calling 'perl' to find its compile-time parameters, ignoring
6905 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
6907 6. The freeze_tell option was not being used for messages that were frozen on
6908 arrival, either by an ACL or by local_scan().
6910 7. Added the smtp_incomplete_transaction log selector.
6912 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
6913 was accepting AUTH without a new EHLO.
6915 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
6916 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
6919 10. Logging of TCP/IP connections (when configured) now happens in the main
6920 daemon process instead of the child process, so that the TCP/IP connection
6921 count is more accurate (but it can never be perfect).
6923 11. The use of "drop" in a nested ACL was not being handled correctly in the
6924 outer ACL. Now, if condition failure induced by the nested "drop" causes
6925 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
6926 or "require"), the connection is dropped.
6928 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
6929 that yield "discard" can now be used with an "accept" or a "discard" verb,
6930 but an error is generated for any others (because I can't see a useful way
6931 to define what should happen).
6933 13. When an ACL is read dynamically from a file (or anywhere else), the lines
6934 are now processed in the same way as lines in the Exim configuration file.
6935 In particular, continuation lines are supported.
6937 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
6939 15. Added -ti meaning -t -i.
6941 16. Check for letters, digits, hyphens, and dots in the names of dnslist
6942 domains, and warn by logging if others are found.
6944 17. At least on BSD, alignment is not guaranteed for the array of ifreq's
6945 returned from GIFCONF when Exim is trying to find the list of interfaces on
6946 a host. The code in os.c has been modified to copy each ifreq to an aligned
6947 structure in all cases.
6949 Also, in some cases, the returned ifreq's were being copied to a 'struct
6950 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
6951 means the last couple of bytes of an IPv6 address could be chopped if the
6952 ifreq contained only a normal sockaddr (14 bytes storage).
6954 18. Named domain lists were not supported in the hosts_treat_as_local option.
6955 An entry such as +xxxx was not recognized, and was treated as a literal
6958 19. Ensure that header lines added by a DATA ACL are included in the reject log
6959 if the ACL subsequently rejects the message.
6961 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
6962 MD5 (which is deprecated).
6964 21. When testing a filter file using -bf, Exim was writing a message when it
6965 took the sender from a "From " line in the message, but it was not doing so
6966 when it took $return_path from a Return-Path: header line. It now does.
6968 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
6969 with a valid header line field name (a series of printing characters
6970 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
6972 23. Changed "disc" in the source to "disk" to conform to the documentation and
6973 the book and for uniformity.
6975 24. Ignore Sendmail's -Ooption=value command line item.
6977 25. When execve() failed while trying to run a command in a pipe transport,
6978 Exim was returning EX_UNAVAILABLE (69) from the subprocess. However, this
6979 could be confused with a return value of 69 from the command itself. This
6980 has been changed to 127, the value the shell returns if it is asked to run
6981 a non-existent command. The wording for the related log line suggests a
6982 non-existent command as the problem.
6984 26. If received_header_text expands to an empty string, do not add a Received:
6985 header line to the message. (Well, it adds a token one on the spool, but
6986 marks it "old" so that it doesn't get used or transmitted.)
6988 27. Installed eximstats 1.28 (addition of -nt option).
6990 28. There was no check for failure on the call to getsockname() in the daemon
6991 code. This can fail if there is a shortage of resources on the system, with
6992 ENOMEM, for example. A temporary error is now given on failure.
6994 29. Contrary to the C standard, it seems that in some environments, the
6995 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
6996 program. Exim now does this explicitly; it affects the formatting of
6997 timestamps using strftime().
6999 30. If exiqsumm was given junk data, it threw up some uninitialized variable
7000 complaints. I've now initialized all the variables, to avoid this.
7002 32. Header lines added by a system filter were not being "seen" during
7003 transport-time rewrites.
7005 33. The info_callback() function passed to OpenSSL is set up with type void
7006 (*)(SSL *, int, int), as described somewhere. However, when calling the
7007 function (actually a macro) that sets it up, the type void(*)() is
7008 expected. I've put in a cast to prevent warnings from picky compilers.
7010 34. If a DNS black list lookup found a CNAME record, but there were no A
7011 records associated with the domain it pointed at, Exim crashed.
7013 35. If a DNS black list lookup returned more than one A record, Exim ignored
7014 all but the first. It now scans all returned addresses if a particular IP
7015 value is being sought. In this situation, the contents of the
7016 $dnslist_value variable are a list of all the addresses, separated by a
7019 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
7020 to accept a host name and all its aliases if the forward lookup for any of
7021 them yielded the IP address of the incoming connection. Now it accepts only
7022 those names whose forward lookup yields the correct IP address. Any other
7023 names are discarded. This closes a loophole whereby a rogue DNS
7024 administrator could create reverse DNS records to break through a
7025 wildcarded host restriction in an ACL.
7027 37. If a user filter or a system filter that ran in a subprocess used any of
7028 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
7029 the wrong values were passed to the pipe command ($thisaddress had the
7030 value of $0, $0 had the value of $1, etc). This bug was introduced by
7031 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
7033 38. Improved the line breaking for long SMTP error messages from ACLs.
7034 Previously, if there was no break point between 40 and 75 characters, Exim
7035 left the rest of the message alone. Two changes have been made: (a) I've
7036 reduced the minimum length to 35 characters; (b) if it can't find a break
7037 point between 35 and 75 characters, it looks ahead and uses the first one
7038 that it finds. This may give the occasional overlong line, but at least the
7039 remaining text gets split now.
7041 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
7042 file descriptors might be low, and that setting 1000 would always raise it.
7043 It turns out that in some environments, the limit is already over 1000 and
7044 that lowering it causes trouble. So now Exim takes care not to decrease it.
7046 40. When delivering a message, the value of $return_path is set to $sender_
7047 address at the start of routing (routers may change the value). By an
7048 oversight, this default was not being set up when an address was tested by
7049 -bt or -bv, which affected the outcome if any router or filter referred to
7052 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
7053 log only when "message" or "log_message" are set. However, if one of the
7054 conditions was an address verification, or a call to a nested ACL, the
7055 messages generated by the underlying test were being passed through. This
7056 no longer happens. The underlying message is available in $acl_verify_
7057 message for both "message" and "log_message" expansions, so it can be
7058 passed through if needed.
7060 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
7061 new expansion $bh_ to give the encoded byte string without charset
7062 translation. Translation happens only if iconv() is available; HAVE_ICONV
7063 indicates this at build time. HEADERS_CHARSET gives the charset to
7064 translate to; headers_charset can change it in the configuration, and
7065 "headers charset" can change it in an individual filter file.
7067 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
7068 that creates RFC 2047 encoded "words" labels them as that charset instead
7069 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
7070 expansion operator; (ii) when Exim creates a From: line for a local
7071 message; (iii) when a header line is rewritten to include a "phrase" part.
7073 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
7074 buggy, causing it to skip the first lines of messages whose message ID
7075 ended in 'D'. This would not have bitten before Exim release 4.14, because
7076 message IDs were unlikely to end in 'D' before then. The effect was to have
7077 incorrect size information for certain domains.
7079 45. #include "config.h" was missing at the start of the crypt16.c module. This
7080 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
7083 46. If there was a timeout during a "random" callout check, Exim treated it as
7084 a failure of the random address, and carried on sending RSET and the real
7085 address. If the delay was just some slowness somewhere, the response to the
7086 original RCPT would be taken as a response to RSET and so on, causing
7087 mayhem of various kinds.
7089 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
7090 when I implemented it. It didn't allow for the fact that some option values
7091 may legitimately be negative (e.g. size_addition), and it didn't even do
7092 the right test for positive values.
7094 48. Domain names in DNS records are case-independent. Exim always looks them up
7095 in lower case. Some resolvers return domain names in exactly the case they
7096 appear in the zone file, that is, they may contain uppercase letters. Not
7097 all resolvers do this - some return always lower case. Exim was treating a
7098 change of case by a resolver as a change of domain, similar to a widening
7099 of a domain abbreviation. This triggered its re-routing code and so it was
7100 trying to route what was effectively the same domain again. This normally
7101 caused routing to fail (because the router wouldn't handle the domain
7102 twice). Now Exim checks for this case specially, and just changes the
7103 casing of the domain that it ultimately uses when it transmits the message
7106 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
7109 50. If a filter generated a file delivery with a non-absolute name (possible if
7110 no home directory exists for the router), the forbid_file option was not
7113 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
7114 the existing equality matching.
7116 52. Exim was using ints instead of ino_t variables in some places where it was
7117 dealing with inode numbers.
7119 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
7120 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
7121 called TMPDIR, and if it finds it is different, it changes its value.
7123 54. The smtp_printf() function is now made available to local_scan() so
7124 additional output lines can be written before returning. There is also an
7125 smtp_fflush() function to enable the detection of a dropped connection.
7126 The variables smtp_input and smtp_batched_input are exported to
7129 55. Changed the default runtime configuration: the message "Unknown user"
7130 has been removed from the ACL, and instead placed on the localuser router,
7131 using the cannot_route_message feature. This means that any verification
7132 failures that generate their own messages won't get overridden. Similarly,
7133 the "Unrouteable address" message that was in the ACL for unverifiable
7134 relay addresses has also been removed.
7136 56. Added hosts_avoid_esmtp to the smtp transport.
7138 57. The exicyclog script was not checking for the esoteric option
7139 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
7140 will work only if exicyclog is run under the appropriate euid.
7142 58. Following a discussion on the list, the rules by which Exim recognises line
7143 endings on incoming messages have been changed. The -dropcr and drop_cr
7144 options are now no-ops, retained only for backwards compatibility. The
7145 following line terminators are recognized: LF CRLF CR. However, special
7146 processing applies to CR:
7148 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
7149 nor a local message in the state where . is a terminator.
7151 (ii) If a bare CR is encountered in a header line, an extra space is added
7152 after the line terminator so as not to end the header. The reasoning
7153 behind this is that bare CRs in header lines are most likely either
7154 to be mistakes, or people trying to play silly games.
7156 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
7157 was being incorrectly given as 18 bytes larger than it should have been.
7158 This is a VOB (very old bug).
7160 60. This may never have affected anything current, but just in case it has:
7161 When the local host is found other than at the start of a list of hosts,
7162 the local host, those with the same MX, and any that follow, are discarded.
7163 When the list in question was part of a longer list of hosts, the following
7164 hosts (not currently being processed) were also being discarded. This no
7165 longer happens. I'm not sure if this situation could ever has previously
7168 61. Added the "/MX" feature to lists of hosts in the manualroute and query
7171 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
7172 header. This is something that is recommended in a new Internet Draft, and
7173 is something that is documented as being done by Sendmail. There are two
7174 possible values. For messages generated by the autoreply transport, Exim
7177 Auto-Submitted: auto-replied
7179 whereas for all other generated messages (e.g. bounces) it adds
7181 Auto-Submitted: auto-generated
7183 63. The "personal" condition in filters now includes a test for the
7184 Auto-Submitted: header. If it contains the string "auto-" the message it
7185 not considered personal.
7187 64. Added rcpt_include_affixes as a generic transport option.
7189 65. Added queue_only_override (default true).
7191 66. Added the syslog_duplication option.
7193 67. If what should have been the first header line of a message consisted of
7194 a space followed by a colon, Exim was mis-interpreting it as a header line.
7195 It isn't of course - it is syntactically invalid and should therefore be
7196 treated as the start of the message body. The misbehaviour could have
7197 caused a number of strange effects, including loss of data in subsequent
7198 header lines, and spool format errors.
7200 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
7201 client host had authenticated. This control can now be exercised by an ACL
7202 for more flexibility.
7204 69. By default, callouts do not happen when testing with -bh. There is now a
7205 variant, -bhc, which does actually run the callout code, including
7206 consulting and updating the callout cache.
7208 70. Added support for saslauthd authentication, courtesy of Alexander
7211 71. If statvfs() failed on the spool or log directories while checking their
7212 size for availability, Exim confusingly gave the error "space shortage".
7213 Furthermore, in debugging mode it crashed with a floating point exception.
7214 These checks are done if check_{spool,log}_{space,inodes} are set, and when
7215 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
7216 serious problem, Exim now writes to the main and panic logs when this
7217 happens, with details of the failure. It then refuses to accept the
7218 incoming message, giving the message "spool directory problem" or "log
7219 directory problem" with a 421 code for SMTP messages.
7221 72. When Exim is about to re-exec itself, it ensures that the file descriptors
7222 0, 1, and 2 exist, because some OS complain for execs without them (see
7223 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
7224 descriptors. However, the code omitted to check that the open succeeded,
7225 causing mysterious errors if for some reason the permissions on /dev/null
7226 got screwed. Now Exim writes a message to the main and panic logs, and
7227 bombs out if it can't open /dev/null.
7229 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
7230 interact so that it is all more flexible. It is supposed to remain
7231 backwards compatible. Also added extra_local_interfaces.
7233 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
7234 to bomb out with an assertion failure - to the client this appears as a
7235 connection drop. This problem occurs in the part of the code that was taken
7236 from the Samba project. Fortunately, the assertion is in a very simple
7237 function, so I have fixed this by reproducing the function inline in the
7238 one place where it is called, and arranging for authentication to fail
7239 instead of killing the process with assert().
7241 75. The SPA client code was not working when the server requested OEM rather
7242 than Unicode encoding.
7244 76. Added code to make require_files with a specific uid setting more usable in
7245 the case where statting the file as root fails - usually a non-root-mounted
7246 NFS file system. When this happens and the failure is EACCES, Exim now
7247 forks a subprocess and does the per-uid checking as the relevant uid.
7249 77. Added process_log_path.
7251 78. If log_file_path was not explicitly set, a setting of check_log_space or
7252 check_log_inodes was ignored.
7254 79. If a space check for the spool or log partitions fails, the incident is now
7255 logged. Of course, in the latter case the data may get lost...
7257 80. Added the %p formatting code to string_format() so that it can be used to
7258 print addresses in debug_print(). Adjusted all the address printing in the
7259 debugging in store.c to use %p rather than %d.
7261 81. There was a concern that a line of code in smtp_in.c could overflow a
7262 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
7263 initially expressed, the concern was not well-founded, because trailing
7264 spaces are removed early. However, if the trailing spaces were followed by
7265 a NULL, they did not get removed, so the overflow was possible. Two fixes
7268 (a) I re-wrote the offending code in a cleaner fashion.
7269 (b) If an incoming SMTP command contains a NULL character, it is rejected
7272 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
7273 runs initgroups(), so that if the Exim user is in any additional groups,
7274 they will be used during message reception.
7280 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / exim.git / blob