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.
66 GF/01 DEFER rather than ERROR on redis cluster MOVED response.
67 When redis_servers is set to a list of > 1 element, and the Redis servers
68 in that list are in cluster configuration, convert the REDIS_REPLY_ERROR
69 case of MOVED into a DEFER case instead, thus moving the query onto the
70 next server in the list. For a cluster of N elements, all N servers must
71 be defined in redis_servers.
73 GF/02 Catch and remove uninitialized value warning in exiqsumm
74 Check for existence of @ARGV before looking at $ARGV[0]
76 JH/01 Replace the store_release() internal interface with store_newblock(),
77 which internalises the check required to safely use the old one, plus
78 the allocate and data copy operations duplicated in both (!) of the
81 JH/02 Disallow '/' characters in queue names specified for the "queue=" ACL
82 modifier. This matches the restriction on the commandline.
84 JH/03 Fix pgsql lookup for multiple result-tuples with a single column.
85 Previously only the last row was returned.
87 JH/04 Bug 2217: Tighten up the parsing of DKIM signature headers. Previously
88 we assumed that tags in the header were well-formed, and parsed the
89 element content after inspecting only the first char of the tag.
90 Assumptions at that stage could crash the receive process on malformed
93 JH/05 Bug 2215: Fix crash associated with dnsdb lookup done from DKIM ACL.
94 While running the DKIM ACL we operate on the Permanent memory pool so that
95 variables created with "set" persist to the DATA ACL. Also (at any time)
96 DNS lookups that fail create cache records using the Permanent pool. But
97 expansions release any allocations made on the current pool - so a dnsdb
98 lookup expansion done in the DKIM ACL releases the memory used for the
99 DNS negative-cache, and bad things result. Solution is to switch to the
100 Main pool for expansions.
101 While we're in that code, add checks on the DNS cache during store_reset,
102 active in the testsuite.
103 Problem spotted, and debugging aided, by Wolfgang Breyha.
105 JH/06 Fix issue with continued-connections when the DNS shifts unreliably.
106 When none of the hosts presented to a transport match an already-open
107 connection, close it and proceed with the list. Previously we would
108 queue the message. Spotted by Lena with Yahoo, probably involving
111 JH/07 Bug 2214: Fix SMTP responses resulting from non-accept result of MIME ACL.
112 Previously a spurious "250 OK id=" response was appended to the proper
115 JH/08 The "support for" informational output now, which built with Content
116 Scanning support, has a line for the malware scanner interfaces compiled
117 in. Interface can be individually included or not at build time.
119 JH/09 The "aveserver", "kavdaemon" and "mksd" interfaces are now not included
120 by the template makefile "src/EDITME". The "STREAM" support for an older
121 ClamAV interface method is removed.
123 JH/10 Bug 2223: Fix mysql lookup returns for the no-data case (when the number of
124 rows affected is given instead).
126 JH/11 The runtime Berkeley DB library version is now additionally output by
127 "exim -d -bV". Previously only the compile-time version was shown.
129 JH/12 Bug 2230: Fix cutthrough routing for nonfirst messages in an initiating
130 SMTP connection. Previously, when one had more receipients than the
131 first, an abortive onward connection was made. Move to full support for
132 multiple onward connections in sequence, handling cutthrough connection
133 for all multi-message initiating connections.
135 JH/13 Bug 2229: Fix cutthrough routing for nonstandard port numbers defined by
136 routers. Previously, a multi-recipient message would fail to match the
137 onward-connection opened for the first recipient, and cause its closure.
139 JH/14 Bug 2174: A timeout on connect for a callout was also erroneously seen as
140 a timeout on read on a GnuTLS initiating connection, resulting in the
141 initiating connection being dropped. This mattered most when the callout
142 was marked defer_ok. Fix to keep the two timeout-detection methods
145 JH/15 Relax results from ACL control request to enable cutthrough, in
146 unsupported situations, from error to silently (except under debug)
147 ignoring. This covers use with PRDR, frozen messages, queue-only and
150 HS/01 Fix Buffer overflow in base64d() (CVE-2018-6789)
152 JH/16 Fix bug in DKIM verify: a buffer overflow could corrupt the malloc
153 metadata, resulting in a crash in free().
155 PP/01 Fix broken Heimdal GSSAPI authenticator integration.
156 Broken in f2ed27cf5, missing an equals sign for specified-initialisers.
157 Broken also in d185889f4, with init system revamp.
159 JH/17 Bug 2113: Fix conversation closedown with the Avast malware scanner.
160 Previously we abruptly closed the connection after reading a malware-
161 found indication; now we go on to read the "scan ok" response line,
164 JH/18 Bug 2239: Enforce non-usability of control=utf8_downconvert in the mail
165 ACL. Previously, a crash would result.
167 JH/19 Speed up macro lookups during configuration file read, by skipping non-
168 macro text after a replacement (previously it was only once per line) and
169 by skipping builtin macros when searching for an uppercase lead character.
171 JH/20 DANE support moved from Experimental to mainline. The Makefile control
172 for the build is renamed.
174 JH/21 Fix memory leak during multi-message connections using STARTTLS. A buffer
175 was allocated for every new TLS startup, meaning one per message. Fix
176 by only allocating once (OpenSSL) or freeing on TLS-close (GnuTLS).
178 JH/22 Bug 2236: When a DKIM verification result is overridden by ACL, DMARC
179 reported the original. Fix to report (as far as possible) the ACL
180 result replacing the original.
182 JH/23 Fix memory leak during multi-message connections using STARTTLS under
183 OpenSSL. Certificate information is loaded for every new TLS startup,
184 and the resources needed to be freed.
186 JH/24 Bug 2242: Fix exim_dbmbuild to permit directoryless filenames.
188 JH/25 Fix utf8_downconvert propagation through a redirect router. Previously it
191 JH/26 Bug 2253: For logging delivery lines under PRDR, append the overall
192 DATA response info to the (existing) per-recipient response info for
193 the "C=" log element. It can have useful tracking info from the
194 destination system. Patch from Simon Arlott.
196 JH/27 Bug 2251: Fix ldap lookups that return a single attribute having zero-
197 length value. Previously this would segfault.
199 HS/02 Support Avast multiline protoocol, this allows passing flags to
200 newer versions of the scanner.
202 JH/28 Ensure that variables possibly set during message acceptance are marked
203 dead before release of memory in the daemon loop. This stops complaints
204 about them when the debug_store option is enabled. Discovered specifically
205 for sender_rate_period, but applies to a whole set of variables.
206 Do the same for the queue-runner and queue-list loops, for variables set
207 from spool message files. Do the same for the SMTP per-message loop, for
208 certain variables indirectly set in ACL operations.
210 JH/29 Bug 2250: Fix a longstanding bug in heavily-pipelined SMTP input (such
211 as a multi-recipient message from a mailinglist manager). The coding had
212 an arbitrary cutoff number of characters while checking for more input;
213 enforced by writing a NUL into the buffer. This corrupted long / fast
214 input. The problem was exposed more widely when more pipelineing of SMTP
215 responses was introduced, and one Exim system was feeding another.
216 The symptom is log complaints of SMTP syntax error (NUL chars) on the
217 receiving system, and refused recipients seen by the sending system
218 (propating to people being dropped from mailing lists).
219 Discovered and pinpointed by David Carter.
221 JH/30 The (EXPERIMENTAL_DMARC) variable $dmarc_ar_header is withdrawn, being
222 replaced by the ${authresults } expansion.
224 JH/31 Bug 2257: Fix pipe transport to not use a socket-only syscall.
226 HS/03 Set a handler for SIGTERM and call exit(3) if running as PID 1. This
227 allows proper process termination in container environments.
229 JH/32 Bug 2258: Fix spool_wireformat in combination with LMTP transport.
230 Previously the "final dot" had a newline after it; ensure it is CR,LF.
232 JH/33 SPF: remove support for the "spf" ACL condition outcome values "err_temp"
233 and "err_perm", deprecated since 4.83 when the RFC-defined words
234 "temperror" and "permerror" were introduced.
236 JH/34 Re-introduce enforcement of no cutthrough delivery on transports having
237 transport-filters or DKIM-signing. The restriction was lost in the
238 consolidation of verify-callout and delivery SMTP handling.
239 Extend the restriction to also cover ARC-signing.
241 JH/35 Cutthrough: for a final-dot response timeout (and nonunderstood responses)
242 in defer=pass mode supply a 450 to the initiator. Previously the message
245 PP/02 DANE: add dane_require_tls_ciphers SMTP Transport option; if unset,
246 tls_require_ciphers is used as before.
248 HS/03 Malware Avast: Better match the Avast multiline protocol. Add
249 "pass_unscanned". Only tmpfails from the scanner are written to
250 the paniclog, as they may require admin intervention (permission
251 denied, license issues). Other scanner errors (like decompression
252 bombs) do not cause a paniclog entry.
254 JH/36 Fix reinitialisation of DKIM logging variable between messages.
255 Previously it was possible to log spurious information in receive log
258 JH/37 Bug 2255: Revert the disable of the OpenSSL session caching. This
259 triggered odd behaviour from Outlook Express clients.
261 PP/03 Add util/renew-opendmarc-tlds.sh script for safe renewal of public
264 JH/38 DKIM: accept Ed25519 pubkeys in SubjectPublicKeyInfo-wrapped form,
265 since the IETF WG has not yet settled on that versus the original
266 "bare" representation.
268 JH/39 Fix syslog logging for syslog_timestamp=no and log_selector +millisec.
269 Previously the millisecond value corrupted the output.
270 Fix also for syslog_pid=no and log_selector +pid, for which the pid
271 corrupted the output.
277 JH/01 Rework error string handling in TLS interface so that the caller in
278 more cases is responsible for logging. This permits library-sourced
279 string to be attached to addresses during delivery, and collapses
280 pairs of long lines into single ones.
282 PP/01 Allow PKG_CONFIG_PATH to be set in Local/Makefile and use it correctly
283 during configuration. Wildcards are allowed and expanded.
285 JH/02 Rework error string handling in DKIM to pass more info back to callers.
286 This permits better logging.
288 JH/03 Rework the transport continued-connection mechanism: when TLS is active,
289 do not close it down and have the child transport start it up again on
290 the passed-on TCP connection. Instead, proxy the child (and any
291 subsequent ones) for TLS via a unix-domain socket channel. Logging is
292 affected: the continued delivery log lines do not have any DNSSEC, TLS
293 Certificate or OCSP information. TLS cipher information is still logged.
295 JH/04 Shorten the log line for daemon startup by collapsing adjacent sets of
296 identical IP addresses on different listening ports. Will also affect
299 PP/02 Bug 2070: uClibc defines __GLIBC__ without providing glibc headers;
300 add noisy ifdef guards to special-case this sillyness.
301 Patch from Bernd Kuhls.
303 JH/05 Tighten up the checking in isip4 (et al): dotted-quad components larger
304 than 255 are no longer allowed.
306 JH/06 Default openssl_options to include +no_ticket, to reduce load on peers.
307 Disable the session-cache too, which might reduce our load. Since we
308 currrectly use a new context for every connection, both as server and
309 client, there is no benefit for these.
310 GnuTLS appears to not support tickets server-side by default (we don't
311 call gnutls_session_ticket_enable_server()) but client side is enabled
312 by default on recent versions (3.1.3 +) unless the PFS priority string
315 PP/03 Add $SOURCE_DATE_EPOCH support for reproducible builds, per spec at
316 <https://reproducible-builds.org/specs/source-date-epoch/>.
318 JH/07 Fix smtp transport use of limited max_rcpt under mua_wrapper. Previously
319 the check for any unsuccessful recipients did not notice the limit, and
320 erroneously found still-pending ones.
322 JH/08 Pipeline CHUNKING command and data together, on kernels that support
323 MSG_MORE. Only in-clear (not on TLS connections).
325 JH/09 Avoid using a temporary file during transport using dkim. Unless a
326 transport-filter is involved we can buffer the headers in memory for
327 creating the signature, and read the spool data file once for the
328 signature and again for transmission.
330 JH/10 Enable use of sendfile in Linux builds as default. It was disabled in
331 4.77 as the kernel support then wasn't solid, having issues in 64bit
332 mode. Now, it's been long enough. Add support for FreeBSD also.
334 JH/11 Bug 2104: Fix continued use of a transport connection with TLS. In the
335 case where the routing stage had gathered several addresses to send to
336 a host before calling the transport for the first, we previously failed
337 to close down TLS in the old transport process before passing the TCP
338 connection to the new process. The new one sent a STARTTLS command
339 which naturally failed, giving a failed delivery and bloating the retry
340 database. Investigation and fix prototype from Wolfgang Breyha.
342 JH/12 Fix check on SMTP command input synchronisation. Previously there were
343 false-negatives in the check that the sender had not preempted a response
344 or prompt from Exim (running as a server), due to that code's lack of
345 awareness of the SMTP input buffering.
347 PP/04 Add commandline_checks_require_admin option.
348 Exim drops privileges sanely, various checks such as -be aren't a
349 security problem, as long as you trust local users with access to their
350 own account. When invoked by services which pass untrusted data to
351 Exim, this might be an issue. Set this option in main configuration
352 AND make fixes to the calling application, such as using `--` to stop
355 JH/13 Do pipelining under TLS. Previously, although safe, no advantage was
356 taken. Now take care to pack both (client) MAIL,RCPT,DATA, and (server)
357 responses to those, into a single TLS record each way (this usually means
358 a single packet). As a side issue, smtp_enforce_sync now works on TLS
361 PP/05 OpenSSL/1.1: use DH_bits() for more accurate DH param sizes. This
362 affects you only if you're dancing at the edge of the param size limits.
363 If you are, and this message makes sense to you, then: raise the
364 configured limit or use OpenSSL 1.1. Nothing we can do for older
367 JH/14 For the "sock" variant of the malware scanner interface, accept an empty
368 cmdline element to get the documented default one. Previously it was
371 JH/15 Fix a crash in the smtp transport caused when two hosts in succession
372 are unsuable for non-message-specific reasons - eg. connection timeout,
373 banner-time rejection.
375 JH/16 Fix logging of delivery remote port, when specified by router, under
378 PP/06 Repair manualroute's ability to take options in any order, even if one
379 is the name of a transport.
382 HS/01 Cleanup, prevent repeated use of -p/-oMr (CVE-2017-1000369)
384 JH/17 Change the list-building routines interface to use the expanding-string
385 triplet model, for better allocation and copying behaviour.
387 JH/18 Prebuild the data-structure for "builtin" macros, for faster startup.
388 Previously it was constructed the first time a possibly-matching string
389 was met in the configuration file input during startup; now it is done
392 JH/19 Bug 2141: Use the full-complex API for Berkeley DB rather than the legacy-
393 compatible one, to avoid the (poorly documented) possibility of a config
394 file in the working directory redirecting the DB files, possibly correpting
395 some existing file. CVE-2017-10140 assigned for BDB.
397 JH/20 Bug 2147: Do not defer for a verify-with-callout-and-random which is not
398 cache-hot. Previously, although the result was properly cached, the
399 initial verify call returned a defer.
401 JH/21 Bug 2151: Avoid using SIZE on the MAIL for a callout verify, on any but
402 the main verify for receipient in uncached-mode.
404 JH/22 Retire historical build files to an "unsupported" subdir. These are
405 defined as "ones for which we have no current evidence of testing".
407 JH/23 DKIM: enforce the DNS pubkey record "h" permitted-hashes optional field,
408 if present. Previously it was ignored.
410 JH/24 Start using specified-initialisers in C structure init coding. This is
411 a C99 feature (it's 2017, so now considered safe).
413 JH/25 Use one-bit bitfields for flags in the "addr" data structure. Previously
414 if was a fixed-sized field and bitmask ops via macros; it is now more
417 PP/07 GitHub PR 56: Apply MariaDB build fix.
418 Patch provided by Jaroslav Škarvada.
420 PP/08 Bug 2161: Fix regression in sieve quoted-printable handling introduced
421 during Coverity cleanups [4.87 JH/47]
422 Diagnosis and fix provided by Michael Fischer v. Mollard.
424 JH/26 Fix DKIM bug: when the pseudoheader generated for signing was exactly
425 the right size to place the terminating semicolon on its own folded
426 line, the header hash was calculated to an incorrect value thanks to
427 the (relaxed) space the fold became.
429 HS/02 Fix Bug 2130: large writes from the transport subprocess where chunked
430 and confused the parent.
432 JH/27 Fix SOCKS bug: an unitialized pointer was deref'd by the transport process
433 which could crash as a result. This could lead to undeliverable messages.
435 JH/28 Logging: "next input sent too soon" now shows where input was truncated
438 JH/29 Fix queue_run_in_order to ignore the PID portion of the message ID. This
439 matters on fast-turnover and PID-randomising systems, which were getting
440 out-of-order delivery.
442 JH/30 Fix a logging bug on aarch64: an unsafe routine was previously used for
443 a possibly-overlapping copy. The symptom was that "Remote host closed
444 connection in response to HELO" was logged instead of the actual 4xx
447 JH/31 Fix CHUNKING code to properly flush the unwanted chunk after an error.
448 Previously only that bufferd was discarded, resulting in SYMTP command
451 JH/32 DKIM: when a message has multiple signatures matching an identity given
452 in dkim_verify_signers, run the dkim acl once for each. Previously only
453 one run was done. Bug 2189.
455 JH/33 Downgrade an unfound-list name (usually a typo in the config file) from
456 "panic the current process" to "deliberately defer". The panic log is
457 still written with the problem list name; the mail and reject logs now
458 get a temp-reject line for the message that was being handled, saying
459 something like "domains check lookup or other defer". The SMTP 451
460 message is still "Temporary local problem".
462 JH/34 Bug 2199: Fix a use-after-free while reading smtp input for header lines.
463 A crafted sequence of BDAT commands could result in in-use memory beeing
464 freed. CVE-2017-16943.
466 HS/03 Bug 2201: Fix checking for leading-dot on a line during headers reading
467 from SMTP input. Previously it was always done; now only done for DATA
468 and not BDAT commands. CVE-2017-16944.
470 JH/35 Bug 2201: Flush received data in BDAT mode after detecting an error fatal
471 to the message (such as an overlong header line). Previously this was
472 not done and we did not exit BDAT mode. Followon from the previous item
473 though a different problem.
479 JH/01 Bug 1922: Support IDNA2008. This has slightly different conversion rules
480 than -2003 did; needs libidn2 in addition to libidn.
482 JH/02 The path option on a pipe transport is now expanded before use.
484 PP/01 GitHub PR 50: Do not call ldap_start_tls_s on ldapi:// connections.
485 Patch provided by "Björn", documentation fix added too.
487 JH/03 Bug 2003: fix Proxy Protocol v2 handling: the address size field was
488 missing a wire-to-host endian conversion.
490 JH/04 Bug 2004: fix CHUNKING in non-PIPELINEING mode. Chunk data following
491 close after a BDAT command line could be taken as a following command,
492 giving a synch failure. Fix by only checking for synch immediately
493 before acknowledging the chunk.
495 PP/02 GitHub PR 52: many spelling fixes, which include fixing parsing of
496 no_require_dnssec option and creation of _HAVE_TRANSPORT_APPEND_MAILDIR
497 macro. Patches provided by Josh Soref.
499 JH/05 Have the EHLO response advertise VRFY, if there is a vrfy ACL defined.
500 Previously we did not; the RFC seems ambiguous and VRFY is not listed
501 by IANA as a service extension. However, John Klensin suggests that we
504 JH/06 Bug 2017: Fix DKIM verification in -bh test mode. The data feed into
505 the dkim code may be unix-mode line endings rather than smtp wire-format
506 CRLF, so prepend a CR to any bare LF.
508 JH/07 Rationalise the coding for callout smtp conversations and transport ones.
509 As a side-benfit, callouts can now use PIPELINING hence fewer round-trips.
511 JH/08 Bug 2016: Fix DKIM verification vs. CHUNKING. Any BDAT commands after
512 the first were themselves being wrongly included in the feed into dkim
513 processing; with most chunk sizes in use this resulted in an incorrect
514 body hash calculated value.
516 JH/09 Bug 2014: permit inclusion of a DKIM-Signature header in a received
517 DKIM signature block, for verification. Although advised against by
518 standards it is specifically not ruled illegal.
520 JH/10 Bug 2025: Fix reception of (quoted) local-parts with embedded spaces.
522 JH/11 Bug 2029: Fix crash in DKIM verification when a message signature block is
523 missing a body hash (the bh= tag).
525 JH/12 Bug 2018: Re-order Proxy Protocol startup versus TLS-on-connect startup.
526 It seems that HAProxy sends the Proxy Protocol information in clear and
527 only then does a TLS startup, so do the same.
529 JH/13 Bug 2027: Avoid attempting to use TCP Fast Open for non-transport client
530 TCP connections (such as for Spamd) unless the daemon successfully set
531 Fast Open mode on its listening sockets. This fixes breakage seen on
532 too-old kernels or those not configured for Fast Open, at the cost of
533 requiring both directions being enabled for TFO, and TFO never being used
534 by non-daemon-related Exim processes.
536 JH/14 Bug 2000: Reject messages recieved with CHUNKING but with malformed line
537 endings, at least on the first header line. Try to canonify any that get
538 past that check, despite the cost.
540 JH/15 Angle-bracket nesting (an error inserted by broken sendmails) levels are
541 now limited to an arbitrary five deep, while parsing addresses with the
542 strip_excess_angle_brackets option enabled.
544 PP/03 Bug 2018: For Proxy Protocol and TLS-on-connect, do not over-read and
545 instead leave the unprompted TLS handshake in socket buffer for the
546 TLS library to consume.
548 PP/04 Bug 2018: Also handle Proxy Protocol v2 safely.
550 PP/05 FreeBSD compat: handle that Ports no longer create /usr/bin/perl
552 JH/16 Drop variables when they go out of scope. Memory management drops a whole
553 region in one operation, for speed, and this leaves assigned pointers
554 dangling. Add checks run only under the testsuite which checks all
555 variables at a store-reset and panics on a dangling pointer; add code
556 explicitly nulling out all the variables discovered. Fixes one known
557 bug: a transport crash, where a dangling pointer for $sending_ip_address
558 originally assigned in a verify callout, is re-used.
560 PP/06 Drop '.' from @INC in various Perl scripts.
562 PP/07 Switch FreeBSD iconv to always use the base-system libc functions.
564 PP/08 Reduce a number of compilation warnings under clang; building with
565 CC=clang CFLAGS+=-Wno-dangling-else -Wno-logical-op-parentheses
566 should be warning-free.
568 JH/17 Fix inbound CHUNKING when DKIM disabled at runtime.
570 HS/01 Fix portability problems introduced by PP/08 for platforms where
571 realloc(NULL) is not equivalent to malloc() [SunOS et al].
573 HS/02 Bug 1974: Fix missing line terminator on the last received BDAT
574 chunk. This allows us to accept broken chunked messages. We need a more
575 general solution here.
577 PP/09 Wrote util/chunking_fixqueue_finalnewlines.pl to help recover
578 already-broken messages in the queue.
580 JH/18 Bug 2061: Fix ${extract } corrupting an enclosing ${reduce } $value.
582 JH/19 Fix reference counting bug in routing-generated-address tracking.
588 JH/01 Use SIZE on MAIL FROM in a cutthrough connection, if the destination
589 supports it and a size is available (ie. the sending peer gave us one).
591 JH/02 The obsolete acl condition "demime" is removed (finally, after ten
592 years of being deprecated). The replacements are the ACLs
593 acl_smtp_mime and acl_not_smtp_mime.
595 JH/03 Upgrade security requirements imposed for hosts_try_dane: previously
596 a downgraded non-dane trust-anchor for the TLS connection (CA-style)
597 or even an in-clear connection were permitted. Now, if the host lookup
598 was dnssec and dane was requested then the host is only used if the
599 TLSA lookup succeeds and is dnssec. Further hosts (eg. lower priority
600 MXs) will be tried (for hosts_try_dane though not for hosts_require_dane)
601 if one fails this test.
602 This means that a poorly-configured remote DNS will make it incommunicado;
603 but it protects against a DNS-interception attack on it.
605 JH/04 Bug 1810: make continued-use of an open smtp transport connection
606 non-noisy when a race steals the message being considered.
608 JH/05 If main configuration option tls_certificate is unset, generate a
609 self-signed certificate for inbound TLS connections.
611 JH/06 Bug 165: hide more cases of password exposure - this time in expansions
612 in rewrites and routers.
614 JH/07 Retire gnutls_require_mac et.al. These were nonfunctional since 4.80
615 and logged a warning sing 4.83; now they are a configuration file error.
617 JH/08 Bug 1836: Fix crash in VRFY handling when handed an unqualified name
618 (lacking @domain). Apply the same qualification processing as RCPT.
620 JH/09 Bug 1804: Avoid writing msglog files when in -bh or -bhc mode.
622 JH/10 Support ${sha256:} applied to a string (as well as the previous
625 JH/11 Cutthrough: avoid using the callout hints db on a verify callout when
626 a cutthrough deliver is pending, as we always want to make a connection.
627 This also avoids re-routing the message when later placing the cutthrough
628 connection after a verify cache hit.
629 Do not update it with the verify result either.
631 JH/12 Cutthrough: disable when verify option success_on_redirect is used, and
632 when routing results in more than one destination address.
634 JH/13 Cutthrough: expand transport dkim_domain option when testing for dkim
635 signing (which inhibits the cutthrough capability). Previously only
636 the presence of an option was tested; now an expansion evaluating as
637 empty is permissible (obviously it should depend only on data available
638 when the cutthrough connection is made).
640 JH/14 Fix logging of errors under PIPELINING. Previously the log line giving
641 the relevant preceding SMTP command did not note the pipelining mode.
643 JH/15 Fix counting of empty lines in $body_linecount and $message_linecount.
644 Previously they were not counted.
646 JH/16 DANE: treat a TLSA lookup response having all non-TLSA RRs, the same
647 as one having no matching records. Previously we deferred the message
648 that needed the lookup.
650 JH/17 Fakereject: previously logged as a normal message arrival "<="; now
651 distinguished as "(=".
653 JH/18 Bug 1867: make the fail_defer_domains option on a dnslookup router work
654 for missing MX records. Previously it only worked for missing A records.
656 JH/19 Bug 1850: support Radius libraries that return REJECT_RC.
658 JH/20 Bug 1872: Ensure that acl_smtp_notquit is run when the connection drops
659 after the data-go-ahead and data-ack. Patch from Jason Betts.
661 JH/21 Bug 1846: Send DMARC forensic reports for reject and quarantine results,
662 even for a "none" policy. Patch from Tony Meyer.
664 JH/22 Fix continued use of a connection for further deliveries. If a port was
665 specified by a router, it must also match for the delivery to be
668 JH/23 Bug 1874: fix continued use of a connection for further deliveries.
669 When one of the recipients of a message was unsuitable for the connection
670 (has no matching addresses), we lost track of needing to mark it
671 deferred. As a result mail would be lost.
673 JH/24 Bug 1832: Log EHLO response on getting conn-close response for HELO.
675 JH/25 Decoding ACL controls is now done using a binary search; the source code
676 takes up less space and should be simpler to maintain. Merge the ACL
677 condition decode tables also, with similar effect.
679 JH/26 Fix problem with one_time used on a redirect router which returned the
680 parent address unchanged. A retry would see the parent address marked as
681 delivered, so not attempt the (identical) child. As a result mail would
684 JH/27 Fix a possible security hole, wherein a process operating with the Exim
685 UID can gain a root shell. Credit to http://www.halfdog.net/ for
686 discovery and writeup. Ubuntu bug 1580454; no bug raised against Exim
689 JH/28 Enable {spool,log} filesystem space and inode checks as default.
690 Main config options check_{log,spool}_{inodes,space} are now
691 100 inodes, 10MB unless set otherwise in the configuration.
693 JH/29 Fix the connection_reject log selector to apply to the connect ACL.
694 Previously it only applied to the main-section connection policy
697 JH/30 Bug 1897: fix callouts connection fallback from TLS to cleartext.
699 PP/01 Changed default Diffie-Hellman parameters to be Exim-specific, created
700 by me. Added RFC7919 DH primes as an alternative.
702 PP/02 Unbreak build via pkg-config with new hash support when crypto headers
703 are not in the system include path.
705 JH/31 Fix longstanding bug with aborted TLS server connection handling. Under
706 GnuTLS, when a session startup failed (eg because the client disconnected)
707 Exim did stdio operations after fclose. This was exposed by a recent
708 change which nulled out the file handle after the fclose.
710 JH/32 Bug 1909: Fix OCSP proof verification for cases where the proof is
711 signed directly by the cert-signing cert, rather than an intermediate
712 OCSP-signing cert. This is the model used by LetsEncrypt.
714 JH/33 Bug 1914: Ensure socket is nonblocking before draining after SMTP QUIT.
716 HS/01 Fix leak in verify callout under GnuTLS, about 3MB per recipient on
717 an incoming connection.
719 HS/02 Bug 1802: Do not half-close the connection after sending a request
722 HS/03 Use "auto" as the default EC curve parameter. For OpenSSL < 1.0.2
723 fallback to "prime256v1".
725 JH/34 SECURITY: Use proper copy of DATA command in error message.
726 Could leak key material. Remotely exploitable. CVE-2016-9963.
732 JH/01 Bug 1664: Disable OCSP for GnuTLS library versions at/before 3.3.16
733 and 3.4.4 - once the server is enabled to respond to an OCSP request
734 it does even when not requested, resulting in a stapling non-aware
735 client dropping the TLS connection.
737 TF/01 Code cleanup: Overhaul the debug_selector and log_selector machinery to
738 support variable-length bit vectors. No functional change.
740 TF/02 Improve the consistency of logging incoming and outgoing interfaces.
741 The I= interface field on outgoing lines is now after the H= remote
742 host field, same as incoming lines. There is a separate
743 outgoing_interface log selector which allows you to disable the
746 JH/02 Bug 728: Close logfiles after a daemon-process "exceptional" log write.
747 If not running log_selector +smtp_connection the mainlog would be held
748 open indefinitely after a "too many connections" event, including to a
749 deleted file after a log rotate. Leave the per net connection logging
750 leaving it open for efficiency as that will be quickly detected by the
751 check on the next write.
753 HS/01 Bug 1671: Fix post transport crash.
754 Processing the wait-<transport> messages could crash the delivery
755 process if the message IDs didn't exist for some reason. When
756 using 'split_spool_directory=yes' the construction of the spool
757 file name failed already, exposing the same netto behaviour.
759 JH/03 Bug 425: Capture substrings in $regex1, $regex2 etc from regex &
760 mime_regex ACL conditions.
762 JH/04 Bug 1686: When compiled with EXPERIMENTAL_DSN_INFO: Add extra information
763 to DSN fail messages (bounces): remote IP, remote greeting, remote response
764 to HELO, local diagnostic string.
766 JH/05 Downgrade message for a TLS-certificate-based authentication fail from
767 log line to debug. Even when configured with a tls authenticator many
768 client connections are expected to not authenticate in this way, so
769 an authenticate fail is not an error.
771 HS/02 Add the Exim version string to the process info. This way exiwhat
772 gives some more detail about the running daemon.
774 JH/06 Bug 1395: time-limit caching of DNS lookups, to the TTL value. This may
775 matter for fast-change records such as DNSBLs.
777 JH/07 Bug 1678: Always record an interface option value, if set, as part of a
778 retry record, even if constant. There may be multiple transports with
779 different interface settings and the retry behaviour needs to be kept
782 JH/08 Bug 1586: exiqgrep now refuses to run if there are unexpected arguments.
784 JH/09 Bug 1700: ignore space & tab embedded in base64 during decode.
786 JH/10 Bug 840: fix log_defer_output option of pipe transport
788 JH/11 Bug 830: use same host for all RCPTS of a message, even under
789 hosts_randomize. This matters a lot when combined with mua_wrapper.
791 JH/12 Bug 1706: percent and underbar characters are no longer escaped by the
792 ${quote_pgsql:<string>} operator.
794 JH/13 Bug 1708: avoid misaligned access in cached lookup.
796 JH/14 Change header file name for freeradius-client. Relevant if compiling
797 with Radius support; from the Gentoo tree and checked under Fedora.
799 JH/15 Bug 1712: Introduce $prdr_requested flag variable
801 JH/16 Bug 1714: Permit an empty string as expansion result for transport
802 option transport_filter, meaning no filtering.
804 JH/17 Bug 1713: Fix non-PDKIM_DEBUG build. Patch from Jasen Betts.
806 JH/18 Bug 1709: When built with TLS support, the tls_advertise_hosts option now
807 defaults to "*" (all hosts). The variable is now available when not built
808 with TLS, default unset, mainly to enable keeping the testsuite sane.
809 If a server certificate is not supplied (via tls_certificate) an error is
810 logged, and clients will find TLS connections fail on startup. Presumably
811 they will retry in-clear.
812 Packagers of Exim are strongly encouraged to create a server certificate
813 at installation time.
815 HS/03 Add -bP config_file as a synonym for -bP configure_file, for consistency
816 with the $config_file variable.
818 JH/19 Two additional event types: msg:rcpt:defer and msg:rcpt:host:defer. Both
819 in transport context, after the attempt, and per-recipient. The latter type
820 is per host attempted. The event data is the error message, and the errno
821 information encodes the lookup type (A vs. MX) used for the (first) host,
822 and the trailing two digits of the smtp 4xx response.
824 GF/01 Bug 1715: Fix for race condition in exicyclog, where exim could attempt
825 to write to mainlog (or rejectlog, paniclog) in the window between file
826 creation and permissions/ownership being changed. Particularly affects
827 installations where exicyclog is run as root, rather than exim user;
828 result is that the running daemon panics and dies.
830 JH/20 Bug 1701: For MySQL lookups, support MySQL config file option group names.
832 JH/21 Bug 1720: Add support for priority groups and weighted-random proxy
833 selection for the EXPERIMENTAL_SOCKS feature, via new per-proxy options
834 "pri" and "weight". Note that the previous implicit priority given by the
835 list order is no longer honoured.
837 JH/22 Bugs 963, 1721: Fix some corner cases in message body canonicalization
840 JH/23 Move SOCKS5 support from Experimental to mainline, enabled for a build
841 by defining SUPPORT_SOCKS.
843 JH/26 Move PROXY support from Experimental to mainline, enabled for a build
844 by defining SUPPORT_PROXY. Note that the proxy_required_hosts option
845 is renamed to hosts_proxy, and the proxy_{host,target}_{address,port}.
846 variables are renamed to proxy_{local,external}_{address,port}.
848 JH/27 Move Internationalisation support from Experimental to mainline, enabled
849 for a build by defining SUPPORT_I18N
851 JH/28 Bug 1745: Fix redis lookups to handle (quoted) spaces embedded in parts
852 of the query string, and make ${quote_redis:} do that quoting.
854 JH/29 Move Events support from Experimental to mainline, enabled by default
855 and removable for a build by defining DISABLE_EVENT.
857 JH/30 Updated DANE implementation code to current from Viktor Dukhovni.
859 JH/31 Fix bug with hosts_connection_nolog and named-lists which were wrongly
860 cached by the daemon.
862 JH/32 Move Redis support from Experimental to mainline, enabled for a build
863 by defining LOOKUP_REDIS. The libhiredis library is required.
865 JH/33 Bug 1748: Permit ACL dnslists= condition in non-smtp ACLs if explicit
866 keys are given for lookup.
868 JH/34 Bug 1192: replace the embedded copy of PolarSSL RSA routines in the DKIM
869 support, by using OpenSSL or GnuTLS library ones. This means DKIM is
870 only supported when built with TLS support. The PolarSSL SHA routines
871 are still used when the TLS library is too old for convenient support.
873 JH/35 Require SINGLE_DH_USE by default in OpenSSL (main config option
874 openssl_options), for security. OpenSSL forces this from version 1.1.0
875 server-side so match that on older versions.
877 JH/36 Bug 1778: longstanding bug in memory use by the ${run } expansion: A fresh
878 allocation for $value could be released as the expansion processing
879 concluded, but leaving the global pointer active for it.
881 JH/37 Bug 1769: Permit a VRFY ACL to override the default 252 response,
882 and to use the domains and local_parts ACL conditions.
884 JH/38 Fix cutthrough bug with body lines having a single dot. The dot was
885 incorrectly not doubled on cutthrough transmission, hence seen as a
886 body-termination at the receiving system - resulting in truncated mails.
887 Commonly the sender saw a TCP-level error, and retransmitted the message
888 via the normal store-and-forward channel. This could result in duplicates
889 received - but deduplicating mailstores were liable to retain only the
890 initial truncated version.
892 JH/39 Bug 1781: Fix use of DKIM private-keys having trailing '=' in the base-64.
894 JH/40 Fix crash in queryprogram router when compiled with EXPERIMENTAL_SRS.
896 JH/41 Bug 1792: Fix selection of headers to sign for DKIM: bottom-up. While
897 we're in there, support oversigning also; bug 1309.
899 JH/42 Bug 1796: Fix error logged on a malware scanner connection failure.
901 HS/04 Add support for keep_environment and add_environment options.
903 JH/43 Tidy coding issues detected by gcc --fsanitize=undefined. Some remain;
904 either intentional arithmetic overflow during PRNG, or testing config-
907 JH/44 Bug 1800: The combination of a -bhc commandline option and cutthrough
908 delivery resulted in actual delivery. Cancel cutthrough before DATA
911 JH/45 Fix cutthrough, when connection not opened by verify and target hard-
912 rejects a recipient: pass the reject to the originator.
914 JH/46 Multiple issues raised by Coverity. Some were obvious or plausible bugs.
915 Many were false-positives and ignorable, but it's worth fixing the
918 JH/47 Fix build on HP-UX and older Solaris, which need (un)setenv now also
919 for the new environment-manipulation done at startup. Move the routines
920 from being local to tls.c to being global via the os.c file.
922 JH/48 Bug 1807: Fix ${extract } for the numeric/3-string case. While preparsing
923 an extract embedded as result-arg for a map, the first arg for extract
924 is unavailable so we cannot tell if this is a numbered or keyed
925 extraction. Accept either.
931 JH/01 Bug 1545: The smtp transport option "retry_include_ip_address" is now
934 JH/02 The smtp transport option "multi_domain" is now expanded.
936 JH/03 The smtp transport now requests PRDR by default, if the server offers
939 JH/04 Certificate name checking on server certificates, when exim is a client,
940 is now done by default. The transport option tls_verify_cert_hostnames
941 can be used to disable this per-host. The build option
942 EXPERIMENTAL_CERTNAMES is withdrawn.
944 JH/05 The value of the tls_verify_certificates smtp transport and main options
945 default to the word "system" to access the system default CA bundle.
946 For GnuTLS, only version 3.0.20 or later.
948 JH/06 Verification of the server certificate for a TLS connection is now tried
949 (but not required) by default. The verification status is now logged by
950 default, for both outbound TLS and client-certificate supplying inbound
953 JH/07 Changed the default rfc1413 lookup settings to disable calls. Few
956 JH/08 The EXPERIMENTAL_DSN compile option is no longer needed; all Delivery
957 Status Notification (bounce) messages are now MIME format per RFC 3464.
958 Support for RFC 3461 DSN options NOTIFY,ENVID,RET,ORCPT can be advertised
959 under the control of the dsn_advertise_hosts option, and routers may
960 have a dsn_lasthop option.
962 JH/09 A timeout of 2 minutes is now applied to all malware scanner types by
963 default, modifiable by a malware= option. The list separator for
964 the options can now be changed in the usual way. Bug 68.
966 JH/10 The smtp_receive_timeout main option is now expanded before use.
968 JH/11 The incoming_interface log option now also enables logging of the
969 local interface on delivery outgoing connections.
971 JH/12 The cutthrough-routing facility now supports multi-recipient mails,
972 if the interface and destination host and port all match.
974 JH/13 Bug 344: The verify = reverse_host_lookup ACL condition now accepts a
977 JH/14 Bug 1573: The spam= ACL condition now additionally supports Rspamd.
978 Patch from Andrew Lewis.
980 JH/15 Bug 670: The spamd_address main option (for the spam= ACL condition)
981 now supports optional time-restrictions, weighting, and priority
982 modifiers per server. Patch originally by <rommer@active.by>.
984 JH/16 The spamd_address main option now supports a mixed list of local
985 and remote servers. Remote servers can be IPv6 addresses, and
986 specify a port-range.
988 JH/17 Bug 68: The spamd_address main option now supports an optional
989 timeout value per server.
991 JH/18 Bug 1581: Router and transport options headers_add/remove can
992 now have the list separator specified.
994 JH/19 Bug 392: spamd_address, and clamd av_scanner, now support retry
997 JH/20 Bug 1571: Ensure that $tls_in_peerdn is set, when verification fails
1000 JH/21 Support for the A6 type of dns record is withdrawn.
1002 JH/22 Bug 608: The result of a QUIT or not-QUIT toplevel ACL now matters
1003 rather than the verbs used.
1005 JH/23 Bug 1572: Increase limit on SMTP confirmation message copy size
1006 from 255 to 1024 chars.
1008 JH/24 Verification callouts now attempt to use TLS by default.
1010 HS/01 DNSSEC options (dnssec_require_domains, dnssec_request_domains)
1011 are generic router options now. The defaults didn't change.
1013 JH/25 Bug 466: Add RFC2322 support for MIME attachment filenames.
1014 Original patch from Alexander Shikoff, worked over by JH.
1016 HS/02 Bug 1575: exigrep falls back to autodetection of compressed
1017 files if ZCAT_COMMAND is not executable.
1019 JH/26 Bug 1539: Add timeout/retry options on dnsdb lookups.
1021 JH/27 Bug 286: Support SOA lookup in dnsdb lookups.
1023 JH/28 Bug 1588: Do not use the A lookup following an AAAA for setting the FQDN.
1024 Normally benign, it bites when the pair was led to by a CNAME;
1025 modern usage is to not canonicalize the domain to a CNAME target
1026 (and we were inconsistent anyway for A-only vs AAAA+A).
1028 JH/29 Bug 1632: Removed the word "rejected" from line logged for ACL discards.
1030 JH/30 Check the forward DNS lookup for DNSSEC, in addition to the reverse,
1031 when evaluating $sender_host_dnssec.
1033 JH/31 Check the HELO verification lookup for DNSSEC, adding new
1034 $sender_helo_dnssec variable.
1036 JH/32 Bug 1397: Enable ECDHE on OpenSSL, just the NIST P-256 curve.
1038 JH/33 Bug 1346: Note MAIL cmd seen in -bS batch, to avoid smtp_no_mail log.
1040 JH/34 Bug 1648: Fix a memory leak seen with "mailq" and large queues.
1042 JH/35 Bug 1642: Fix support of $spam_ variables at delivery time. Was
1043 documented as working, but never had. Support all but $spam_report.
1045 JH/36 Bug 1659: Guard checking of input smtp commands again pseudo-command
1046 added for tls authenticator.
1048 HS/03 Add perl_taintmode main config option
1054 TL/01 When running the test suite, the README says that variables such as
1055 no_msglog_check are global and can be placed anywhere in a specific
1056 test's script, however it was observed that placement needed to be near
1057 the beginning for it to behave that way. Changed the runtest perl
1058 script to read through the entire script once to detect and set these
1059 variables, reset to the beginning of the script, and then run through
1060 the script parsing/test process like normal.
1062 TL/02 The BSD's have an arc4random API. One of the functions to induce
1063 adding randomness was arc4random_stir(), but it has been removed in
1064 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
1065 function when detected.
1067 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
1068 cause callback expansion.
1070 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
1071 syntax errors in an expansion can be treated as a string instead of
1072 logging or causing an error, due to the internal use of bool_lax
1073 instead of bool when processing it.
1075 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
1076 server certificates when making smtp deliveries.
1078 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
1080 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
1082 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
1084 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
1085 Merged patch from Sebastian Wiedenroth.
1087 JH/05 Fix results-pipe from transport process. Several recipients, combined
1088 with certificate use, exposed issues where response data items split
1089 over buffer boundaries were not parsed properly. This eventually
1090 resulted in duplicates being sent. This issue only became common enough
1091 to notice due to the introduction of connection certificate information,
1092 the item size being so much larger. Found and fixed by Wolfgang Breyha.
1094 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
1095 size buffer was used, resulting in syntax errors when an expansion
1098 JH/07 Add support for directories of certificates when compiled with a GnuTLS
1099 version 3.3.6 or later.
1101 JH/08 Rename the TPDA experimental facility to Event Actions. The #ifdef
1102 is EXPERIMENTAL_EVENT, the main-configuration and transport options
1103 both become "event_action", the variables become $event_name, $event_data
1104 and $event_defer_errno. There is a new variable $verify_mode, usable in
1105 routers, transports and related events. The tls:cert event is now also
1106 raised for inbound connections, if the main configuration event_action
1109 TL/06 In test suite, disable OCSP for old versions of openssl which contained
1110 early OCSP support, but no stapling (appears to be less than 1.0.0).
1112 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
1113 server certificate names available under the smtp transport option
1114 "tls_verify_cert_hostname" now do not permit multi-component wildcard
1117 JH/10 Time-related extraction expansions from certificates now use the main
1118 option "timezone" setting for output formatting, and are consistent
1119 between OpenSSL and GnuTLS compilations. Bug 1541.
1121 JH/11 Fix a crash in mime ACL when meeting a zero-length, quoted or RFC2047-
1122 encoded parameter in the incoming message. Bug 1558.
1124 JH/12 Bug 1527: Autogrow buffer used in reading spool files. Since they now
1125 include certificate info, eximon was claiming there were spoolfile
1128 JH/13 Bug 1521: Fix ldap lookup for single-attr request, multiple-attr return.
1130 JH/14 Log delivery-related information more consistently, using the sequence
1131 "H=<name> [<ip>]" wherever possible.
1133 TL/07 Bug 1547: Omit RFCs from release. Draft and RFCs have licenses which
1134 are problematic for Debian distribution, omit them from the release
1137 JH/15 Updates and fixes to the EXPERIMENTAL_DSN feature.
1139 JH/16 Fix string representation of time values on 64bit time_t architectures.
1142 JH/17 Fix a null-indirection in certextract expansions when a nondefault
1143 output list separator was used.
1148 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
1149 checkers that were complaining about end of non-void function with no
1152 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
1153 This was a regression introduced in 4.83 by another bugfix.
1155 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
1157 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
1158 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
1164 TF/01 Correctly close the server side of TLS when forking for delivery.
1166 When a message was received over SMTP with TLS, Exim failed to clear up
1167 the incoming connection properly after forking off the child process to
1168 deliver the message. In some situations the subsequent outgoing
1169 delivery connection happened to have the same fd number as the incoming
1170 connection previously had. Exim would try to use TLS and fail, logging
1171 a "Bad file descriptor" error.
1173 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
1174 utilities have not been installed.
1176 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
1177 temporary space as the ACL may create new global variables.
1179 TL/01 LDAP support uses per connection or global context settings, depending
1180 upon the detected version of the libraries at build time.
1182 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
1183 to extract and use the src ip:port in logging and expansions as if it
1184 were a direct connection from the outside internet. PPv2 support was
1185 updated based on HAProxy spec change in May 2014.
1187 JH/02 Add ${listextract {number}{list}{success}{fail}}.
1189 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
1190 Properly escape header and check for NULL return.
1192 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
1195 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
1197 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
1198 characters in header names, implemented as a verify condition.
1199 Contributed by Michael Fischer v. Mollard.
1201 TL/05 Rename SPF condition results err_perm and err_temp to standardized
1202 results permerror and temperror. Previous values are deprecated but
1203 still accepted. In a future release, err_perm and err_temp will be
1204 completely removed, which will be a backward incompatibility if the
1205 ACL tests for either of these two old results. Patch contributed by
1206 user bes-internal on the mailing list.
1208 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
1210 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
1211 selectors, in both main and reject logs.
1213 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
1216 JH/07 Add malware type "sock" for talking to simple daemon.
1218 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
1220 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
1221 routers/transports under cutthrough routing.
1223 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
1224 numbers. Touch up "bool" conditional to keep the same definition.
1226 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
1228 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
1230 JH/12 Expand items in router/transport headers_add or headers_remove lists
1231 individually rather than the list as a whole. Bug 1452.
1233 Required for reasonable handling of multiple headers_ options when
1234 they may be empty; requires that headers_remove items with embedded
1235 colons must have them doubled (or the list-separator changed).
1237 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
1238 view the policy declared in the DMARC record. Currently, $dmarc_status
1239 is a combined value of both the record presence and the result of the
1242 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
1244 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
1245 dnslookup router and the smtp transport (applying to the forward
1248 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
1249 of ldap servers used for a specific lookup. Patch provided by Heiko
1252 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
1253 New variable $lookup_dnssec_authenticated for observability.
1255 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
1256 Patch submitted by Lars Timman.
1258 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
1260 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
1261 Requires trusted mode and valid format message id, aborts otherwise.
1262 Patch contributed by Heiko Schlichting.
1264 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
1265 certextract with support for various fields. Bug 1358.
1267 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
1268 is requested by default, modifiable by smtp transport option
1271 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
1272 operate on certificate variables to give certificate fingerprints
1273 Also new ${sha256:cert_variable}.
1275 JH/23 The PRDR feature is moved from being Experimental into the mainline.
1277 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
1278 Christian Aistleitner.
1280 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
1282 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
1283 file. Patch from Wolfgang Breyha.
1285 JH/25 Expand the coverage of the delivery $host and $host_address to
1286 client authenticators run in verify callout. Bug 1476.
1288 JH/26 Port service names are now accepted for tls_on_connect_ports, to
1289 align with daemon_smtp_ports. Bug 72.
1291 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
1292 support and error reporting did not work properly.
1294 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
1295 and is readable. Patch from Andrew Colin Kissa.
1297 TL/14 Enhance documentation of ${run expansion and how it parses the
1298 commandline after expansion, particularly in the case when an
1299 unquoted variable expansion results in an empty value.
1301 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
1303 PP/02 Fix internal collision of T_APL on systems which support RFC3123
1304 by renaming away from it. Addresses GH issue 15, reported by
1307 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
1309 TL/15 SECURITY: prevent double expansion in math comparison functions
1310 (can expand unsanitized data). Not remotely exploitable.
1317 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
1319 PP/02 Make -n do something, by making it not do something.
1320 When combined with -bP, the name of an option is not output.
1322 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
1325 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
1326 $sender_host_name and config options to manage this, and basic check
1329 PP/05 DSCP support for outbound connections and control modifier for inbound.
1331 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
1332 (Only plugin which currently uses this is kerberos4, which nobody should
1333 be using, but we should make it available and other future plugins might
1334 conceivably use it, even though it would break NAT; stuff *should* be
1335 using channel bindings instead).
1337 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
1338 name; added for Sendmail compatibility; requires admin caller.
1339 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
1340 just ignore it); requires trusted caller.
1341 Also parse but ignore: -Ac -Am -X<logfile>
1344 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
1346 TL/02 Add +smtp_confirmation as a default logging option.
1348 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
1349 Patch by Magnus Holmgren from 2007-02-20.
1351 TL/04 Bugzilla 1281 - Spec typo.
1352 Bugzilla 1283 - Spec typo.
1353 Bugzilla 1290 - Spec grammar fixes.
1355 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
1357 TL/06 Add Experimental DMARC support using libopendmarc libraries.
1359 TL/07 Fix an out of order global option causing a segfault. Reported to dev
1360 mailing list by by Dmitry Isaikin.
1362 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
1364 JH/02 Support "G" suffix to numbers in ${if comparisons.
1366 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
1368 NM/01 Bugzilla 1197 - Spec typo
1369 Bugzilla 1196 - Spec examples corrections
1371 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
1373 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
1374 gnutls_enable_pkcs11, but renamed to more accurately indicate its
1377 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
1378 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
1380 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
1381 "acl {{name}{arg}...}", and optional args on acl condition
1384 JH/05 Permit multiple router/transport headers_add/remove lines.
1386 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
1388 JH/07 Avoid using a waiting database for a single-message-only transport.
1389 Performance patch from Paul Fisher. Bugzilla 1262.
1391 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
1394 JH/09 Add $headers_added variable, with content from use of ACL modifier
1395 add_header (but not yet added to the message). Bugzilla 199.
1397 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
1398 Pulled from Bugzilla 817 by Wolfgang Breyha.
1400 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
1402 (nb: this is the same fix as in Exim 4.80.1)
1404 JH/11 Add A= logging on delivery lines, and a client_set_id option on
1407 JH/12 Add optional authenticated_sender logging to A= and a log_selector
1410 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
1412 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
1413 advertise SMTP AUTH mechanism to us, instead of a generic
1414 protocol violation error. Also, make Exim more robust to bad
1415 data from the Dovecot auth socket.
1417 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
1419 When a queue runner is handling a message, Exim first routes the
1420 recipient addresses, during which it prunes them based on the retry
1421 hints database. After that it attempts to deliver the message to
1422 any remaining recipients. It then updates the hints database using
1425 So if a recipient address works intermittently, it can get repeatedly
1426 deferred at routing time. The retry hints record remains fresh so the
1427 address never reaches the final cutoff time.
1429 This is a fairly common occurrence when a user is bumping up against
1430 their storage quota. Exim had some logic in its local delivery code
1431 to deal with this. However it did not apply to per-recipient defers
1432 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
1434 This change adds a proper retry rule check during routing so that the
1435 final cutoff time is checked against the message's age. We only do
1436 this check if there is an address retry record and there is not a
1437 domain retry record; this implies that previous attempts to handle
1438 the address had the retry_use_local_parts option turned on. We use
1439 this as an approximation for the destination being like a local
1440 delivery, as in LMTP.
1442 I suspect this new check makes the old local delivery cutoff check
1443 redundant, but I have not verified this so I left the code in place.
1445 TF/02 Correct gecos expansion when From: is a prefix of the username.
1447 Test 0254 submits a message to Exim with the header
1451 When I ran the test suite under the user fanf2, Exim expanded
1452 the header to contain my full name, whereas it should have added
1453 a Resent-Sender: header. It erroneously treats any prefix of the
1454 username as equal to the username.
1456 This change corrects that bug.
1458 GF/01 DCC debug and logging tidyup
1459 Error conditions log to paniclog rather than rejectlog.
1460 Debug lines prefixed by "DCC: " to remove any ambiguity.
1462 TF/03 Avoid unnecessary rebuilds of lookup-related code.
1464 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
1465 Bug spotted by Jeremy Harris; was flawed since initial commit.
1466 Would have resulted in OCSP responses post-SNI triggering an Exim
1467 NULL dereference and crash.
1469 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
1471 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
1472 Bug detection, analysis and fix by Samuel Thibault.
1473 Bugzilla 1331, Debian bug #698092.
1475 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
1477 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
1478 Server implementation by Todd Lyons, client by JH.
1479 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
1480 config variable "prdr_enable" controls whether the server
1481 advertises the facility. If the client requests PRDR a new
1482 acl_data_smtp_prdr ACL is called once for each recipient, after
1483 the body content is received and before the acl_smtp_data ACL.
1484 The client is controlled by both of: a hosts_try_prdr option
1485 on the smtp transport, and the server advertisement.
1486 Default client logging of deliveries and rejections involving
1487 PRDR are flagged with the string "PRDR".
1489 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
1490 fclose(). Diagnosis by Todd Lyons.
1492 PP/17 Update configure.default to handle IPv6 localhost better.
1493 Patch by Alain Williams (plus minor tweaks).
1496 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
1497 This is now consistent with GnuTLS, and is now documented: the
1498 previous undocumented portable approach to treating the option as
1499 unset was to force an expansion failure. That still works, and
1500 an empty string is now equivalent.
1502 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
1503 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
1504 not performing validation itself.
1506 PP/20 Added force_command boolean option to pipe transport.
1507 Patch from Nick Koston, of cPanel Inc.
1509 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
1512 TF/04 Added udpsend ACL modifier and hexquote expansion operator
1514 PP/21 Fix eximon continuous updating with timestamped log-files.
1515 Broken in a format-string cleanup in 4.80, missed when I repaired the
1516 other false fix of the same issue.
1517 Report and fix from Heiko Schlichting.
1520 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
1521 Report from Prashanth Katuri.
1523 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
1524 It's SecureTransport, so affects any MacOS clients which use the
1525 system-integrated TLS libraries, including email clients.
1527 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
1528 using a MIME ACL for non-SMTP local injection.
1529 Report and assistance in diagnosis by Warren Baker.
1531 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
1533 JH/16 Fix comparisons for 64b. Bugzilla 1385.
1535 TL/09 Add expansion variable $authenticated_fail_id to keep track of
1536 last id that failed so it may be referenced in subsequent ACL's.
1538 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
1541 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
1542 ldap library initialization, allowing self-signed CA's to be
1543 used. Also properly sets require_cert option later in code by
1544 using NULL (global ldap config) instead of ldap handle (per
1545 session). Bug diagnosis and testing by alxgomz.
1547 TL/12 Enhanced documentation in the ratelimit.pl script provided in
1548 the src/util/ subdirectory.
1550 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
1551 renamed to Transport Post Delivery Action by Jeremy Harris, as
1554 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
1555 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
1556 redis_servers = needs to be configured which will be used by the redis
1557 lookup. Patch from Warren Baker, of The Packet Hub.
1559 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
1561 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
1562 hostname or reverse DNS when processing a host list. Used suggestions
1563 from multiple comments on this bug.
1565 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
1567 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
1568 Missed a few lines, added it to make the runtest require no keyboard
1571 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
1572 contains upper case chars. Make router use caseful_local_part.
1574 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
1575 support when GnuTLS has been built with p11-kit.
1581 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
1583 This, or similar/improved, will also be change PP/11 of 4.82.
1589 PP/01 Handle short writes when writing local log-files.
1590 In practice, only affects FreeBSD (8 onwards).
1591 Bugzilla 1053, with thanks to Dmitry Isaikin.
1593 NM/01 Bugzilla 949 - Documentation tweak
1595 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
1598 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
1600 PP/02 Implemented gsasl authenticator.
1602 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
1604 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
1605 `pkg-config foo` for cflags/libs.
1607 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
1608 with rest of GSASL and with heimdal_gssapi.
1610 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
1611 `pkg-config foo` for cflags/libs for the TLS implementation.
1613 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
1614 properties get this fed in as external SSF. A number of robustness
1615 and debugging improvements to the cyrus_sasl authenticator.
1617 PP/08 cyrus_sasl server now expands the server_realm option.
1619 PP/09 Bugzilla 1214 - Log authentication information in reject log.
1620 Patch by Jeremy Harris.
1622 PP/10 Added dbmjz lookup type.
1624 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
1626 PP/12 MAIL args handles TAB as well as SP, for better interop with
1627 non-compliant senders.
1628 Analysis and variant patch by Todd Lyons.
1630 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
1631 Bug report from Lars Müller <lars@samba.org> (via SUSE),
1632 Patch from Dirk Mueller <dmueller@suse.com>
1634 PP/13 tls_peerdn now print-escaped for spool files.
1635 Observed some $tls_peerdn in wild which contained \n, which resulted
1636 in spool file corruption.
1638 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
1639 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
1640 or write after TLS renegotiation, which otherwise led to messages
1643 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
1644 as a tracking header (ie: a signed header comes before the signature).
1645 Patch from Wolfgang Breyha.
1647 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
1648 comma-sep list; embedded commas doubled.
1650 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
1652 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
1654 Report and patch from Dmitry Banschikov.
1656 PP/16 Removed "dont_insert_empty_fragments" from "openssl_options".
1657 Removed SSL_clear() after SSL_new() which led to protocol negotiation
1658 failures. We appear to now support TLS1.1+ with Exim.
1660 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
1661 lets Exim select keys and certificates based upon TLS SNI from client.
1662 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
1663 before an outbound SMTP session. New log_selector, +tls_sni.
1665 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
1666 NULL dereference. Report and patch from Alun Jones.
1668 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
1669 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
1670 Not seeing resolver debug output on NetBSD, but suspect this is a
1671 resolver implementation change.
1673 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
1674 Left warnings. Added "eximon gdb" invocation mode.
1676 PP/21 Defaulting "accept_8bitmime" to true, not false.
1678 PP/22 Added -bw for inetd wait mode support.
1680 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
1681 locate the relevant includes and libraries. Made this the default.
1683 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
1684 Bugzilla 1246, report and most of solution from Tomasz Kusy.
1686 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
1687 This may cause build issues on older platforms.
1689 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
1690 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
1691 gnutls_require_mac & gnutls_require_protocols (no longer supported).
1692 Added SNI support via GnuTLS too.
1693 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
1695 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
1697 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
1698 Applied second patch from Janne, implementing suggestion to default
1699 multiple-strings-in-record handling to match SPF spec.
1701 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
1703 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
1704 read-only, out of scope).
1705 Patch from Wolfgang Breyha, report from Stuart Northfield.
1707 PP/29 Fix three issues highlighted by clang analyser static analysis.
1708 Only crash-plausible issue would require the Cambridge-specific
1709 iplookup router and a misconfiguration.
1710 Report from Marcin Mirosław.
1712 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
1714 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
1715 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
1716 As part of this, removing so much warning spew let me fix some minor
1717 real issues in debug logging.
1719 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
1720 assignment on my part. Fixed.
1722 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
1723 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
1724 Janne Snabb (who went above and beyond: thank you).
1726 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
1727 string otherwise requires a connection and a bunch more work and it's
1728 relatively easy to get wrong. Should also expose TLS library linkage
1731 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
1732 64-bit ${eval} (JH/03).
1734 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
1735 GNU libc to support some of the 64-bit stuff, should not lead to
1736 conflicts. Defined before os.h is pulled in, so if a given platform
1737 needs to override this, it can.
1739 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
1740 protection layer was required, which is not implemented.
1741 Bugzilla 1254, patch from Wolfgang Breyha.
1743 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
1744 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
1745 tls_dhparam take prime identifiers. Also unbreak combination of
1746 OpenSSL+DH_params+TLSSNI.
1748 PP/39 Disable SSLv2 by default in OpenSSL support.
1754 PP/01 Solaris build fix for Oracle's LDAP libraries.
1755 Bugzilla 1109, patch from Stephen Usher.
1757 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
1759 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
1762 TF/02 Fix a couple more cases where we did not log the error message
1763 when unlink() failed. See also change 4.74-TF/03.
1765 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
1766 lock up or crash if it happened to be inside a call to libc when it
1767 got a SIGUSR1 from exiwhat.
1769 The SIGUSR1 handler appends the current process status to the process
1770 log which is later printed by exiwhat. It used to use the general
1771 purpose logging code to do this, but several functions it calls are
1772 not safe for signals.
1774 The new output code in the SIGUSR1 handler is specific to the process
1775 log, and simple enough that it's easy to inspect for signal safety.
1776 Removing some special cases also simplifies the general logging code.
1777 Removing the spurious timestamps from the process log simplifies
1780 TF/04 Improved ratelimit ACL condition.
1782 The /noupdate option has been deprecated in favour of /readonly which
1783 has clearer semantics. The /leaky, /strict, and /readonly update modes
1784 are mutually exclusive. The update mode is no longer included in the
1785 database key; it just determines when the database is updated. (This
1786 means that when you upgrade Exim will forget old rate measurements.)
1788 Exim now checks that the per_* options are used with an update mode that
1789 makes sense for the current ACL. For example, when Exim is processing a
1790 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
1791 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
1792 must specify per_mail/readonly. If you omit the update mode it defaults to
1793 /leaky where that makes sense (as before) or /readonly where required.
1795 The /noupdate option is now undocumented but still supported for
1796 backwards compatibility. It is equivalent to /readonly except that in
1797 ACLs where /readonly is required you may specify /leaky/noupdate or
1798 /strict/noupdate which are treated the same as /readonly.
1800 A useful new feature is the /count= option. This is a generalization
1801 of the per_byte option, so that you can measure the throughput of other
1802 aggregate values. For example, the per_byte option is now equivalent
1803 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
1805 The per_rcpt option has been generalized using the /count= mechanism
1806 (though it's more complicated than the per_byte equivalence). When it is
1807 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
1808 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
1809 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
1810 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
1811 Note that using per_rcpt with a non-readonly update mode in more than
1812 one ACL will cause the recipients to be double-counted. (The per_mail
1813 and per_byte options don't have this problem.)
1815 The handling of very low rates has changed slightly. If the computed rate
1816 is less than the event's count (usually one) then this event is the first
1817 after a long gap. In this case the rate is set to the same as this event's
1818 count, so that the first message of a spam run is counted properly.
1820 The major new feature is a mechanism for counting the rate of unique
1821 events. The new per_addr option counts the number of different
1822 recipients that someone has sent messages to in the last time period. It
1823 behaves like per_rcpt if all the recipient addresses are different, but
1824 duplicate recipient addresses do not increase the measured rate. Like
1825 the /count= option this is a general mechanism, so the per_addr option
1826 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
1827 example, measure the rate that a client uses different sender addresses
1828 with the options per_mail/unique=$sender_address. There are further
1829 details in the main documentation.
1831 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
1833 TF/06 Removed a few PCRE remnants.
1835 TF/07 Automatically extract Exim's version number from tags in the git
1836 repository when doing development or release builds.
1838 PP/02 Raise smtp_cmd_buffer_size to 16kB.
1839 Bugzilla 879. Patch from Paul Fisher.
1841 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
1842 Heavily based on revision 40f9a89a from Simon Arlott's tree.
1845 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
1847 PP/05 Variable $av_failed, true if the AV scanner deferred.
1848 Bugzilla 1078. Patch from John Horne.
1850 PP/06 Stop make process more reliably on build failure.
1851 Bugzilla 1087. Patch from Heiko Schlittermann.
1853 PP/07 Make maildir_use_size_file an _expandable_ boolean.
1854 Bugzilla 1089. Patch from Heiko Schlittermann.
1856 PP/08 Handle ${run} returning more data than OS pipe buffer size.
1857 Bugzilla 1131. Patch from Holger Weiß.
1859 PP/09 Handle IPv6 addresses with SPF.
1860 Bugzilla 860. Patch from Wolfgang Breyha.
1862 PP/10 GnuTLS: support TLS 1.2 & 1.1.
1864 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
1867 PP/11 match_* no longer expand right-hand-side by default.
1868 New compile-time build option, EXPAND_LISTMATCH_RHS.
1869 New expansion conditions, "inlist", "inlisti".
1871 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
1873 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
1875 PP/14 fix log_write() format string regression from TF/03.
1876 Bugzilla 1152. Patch from Dmitry Isaikin.
1882 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
1884 PP/02 Harmonised TLS library version reporting; only show if debugging.
1885 Layout now matches that introduced for other libraries in 4.74 PP/03.
1887 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
1889 PP/04 New "dns_use_edns0" global option.
1891 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
1894 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
1895 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
1897 TK/01 Updated PolarSSL code to 0.14.2.
1898 Bugzilla 1097. Patch from Andreas Metzler.
1900 PP/07 Catch divide-by-zero in ${eval:...}.
1901 Fixes bugzilla 1102.
1903 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
1906 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
1907 format-string attack -- SECURITY: remote arbitrary code execution.
1909 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
1910 time unintentionally subject to list matching rules, letting the header
1911 cause arbitrary Exim lookups (of items which can occur in lists, *not*
1912 arbitrary string expansion). This allowed for information disclosure.
1914 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
1915 INT_MIN/-1 -- value coerced to INT_MAX.
1921 NM/01 Workaround for PCRE version dependency in version reporting
1924 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
1925 This fixes portability to compilers other than gcc, notably
1926 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
1928 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
1929 makefiles for portability to HP-UX and POSIX correctness.
1931 PP/01 Permit LOOKUP_foo enabling on the make command-line.
1932 Also via indented variable definition in the Makefile.
1933 (Debugging by Oliver Heesakkers).
1935 PP/02 Restore caching of spamd results with expanded spamd_address.
1936 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
1938 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
1939 Improves build reliability. Fix from: Frank Elsner
1941 NM/02 Fix wide character breakage in the rfc2047 coding
1942 Fixes bug 1064. Patch from Andrey N. Oktyabrski
1944 NM/03 Allow underscore in dnslist lookups
1945 Fixes bug 1026. Patch from Graeme Fowler
1947 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
1948 Code patches from Adam Ciarcinski of NetBSD.
1950 NM/04 Fixed exiqgrep to cope with mailq missing size issue
1953 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
1954 is logged, to avoid truncation. Patch from John Horne.
1956 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
1957 Patch from Jakob Hirsch.
1959 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
1960 SQL string expansion failure details.
1961 Patch from Andrey Oktyabrski.
1963 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
1964 Patch from Simon Arlott.
1966 PP/09 New lookups functionality failed to compile on old gcc which rejects
1967 extern declarations in function scope.
1968 Patch from Oliver Fleischmann
1970 PP/10 Use sig_atomic_t for flags set from signal handlers.
1971 Check getgroups() return and improve debugging.
1972 Fixed developed for diagnosis in bug 927 (which turned out to be
1975 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
1976 Patch from Mark Zealey.
1978 PP/12 Bugzilla 1056: Improved spamd server selection.
1979 Patch from Mark Zealey.
1981 PP/13 Bugzilla 1086: Deal with maildir quota file races.
1982 Based on patch from Heiko Schlittermann.
1984 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
1985 Patch from Uwe Doering, sign-off by Michael Haardt.
1987 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
1988 variable declaration deep within a block. Bug and patch from
1991 PP/15 lookups-Makefile IRIX compatibility coercion.
1993 PP/16 Make DISABLE_DKIM build knob functional.
1995 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
1996 Patch by Simon Arlott
1998 TF/03 Fix valgrind.h portability to C89 compilers that do not support
1999 variable argument macros. Our copy now differs from upstream.
2005 TF/01 Failure to get a lock on a hints database can have serious
2006 consequences so log it to the panic log.
2008 TF/02 Log LMTP confirmation messages in the same way as SMTP,
2009 controlled using the smtp_confirmation log selector.
2011 TF/03 Include the error message when we fail to unlink a spool file.
2013 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
2014 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
2015 for maintaining out-of-tree patches for some time.
2017 PP/01 Bugzilla 139: Documentation and portability issues.
2018 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
2019 Handle per-OS dynamic-module compilation flags.
2021 PP/02 Let /dev/null have normal permissions.
2022 The 4.73 fixes were a little too stringent and complained about the
2023 permissions on /dev/null. Exempt it from some checks.
2024 Reported by Andreas M. Kirchwitz.
2026 PP/03 Report version information for many libraries, including
2027 Exim version information for dynamically loaded libraries. Created
2028 version.h, now support a version extension string for distributors
2029 who patch heavily. Dynamic module ABI change.
2031 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
2032 privilege escalation vulnerability whereby the Exim run-time user
2033 can cause root to append content of the attacker's choosing to
2036 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
2039 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
2040 If dropping privileges for untrusted macros, we disabled normal logging
2041 on the basis that it would fail; for the Exim run-time user, this is not
2042 the case, and it resulted in successful deliveries going unlogged.
2043 Fixed. Reported by Andreas Metzler.
2049 PP/01 Date: & Message-Id: revert to normally being appended to a message,
2050 only prepend for the Resent-* case. Fixes regression introduced in
2051 Exim 4.70 by NM/22 for Bugzilla 607.
2053 PP/02 Include check_rfc2047_length in configure.default because we're seeing
2054 increasing numbers of administrators be bitten by this.
2056 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
2058 PP/03 Bugzilla 994: added openssl_options main configuration option.
2060 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
2062 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
2064 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
2066 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
2067 without a peer certificate, leading to a segfault because of an
2068 assumption that peers always have certificates. Be a little more
2069 paranoid. Problem reported by Martin Tscholak.
2071 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
2072 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
2073 NB: ClamAV planning to remove STREAM in "middle of 2010".
2074 CL also introduces -bmalware, various -d+acl logging additions and
2075 more caution in buffer sizes.
2077 PP/09 Implemented reverse_ip expansion operator.
2079 PP/10 Bugzilla 937: provide a "debug" ACL control.
2081 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
2083 PP/12 Bugzilla 973: Implement --version.
2085 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
2087 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
2089 PP/15 Bugzilla 816: support multiple condition rules on Routers.
2091 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
2092 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
2093 ignore trailing whitespace.
2095 JJ/02 prevent non-panic DKIM error from being sent to paniclog
2097 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
2100 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
2101 Notification from Dr Andrew Aitchison.
2103 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
2104 ExtendedDetectionInfo response format.
2105 Notification from John Horne.
2107 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
2110 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
2111 XSL and documented dependency on system catalogs, with examples of how
2114 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
2117 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
2118 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
2119 configuration file which is writeable by the Exim user or group.
2121 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
2122 of configuration files to cover files specified with the -C option if
2123 they are going to be used with root privileges, not just the default
2126 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
2127 option (effectively making it always true).
2129 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
2130 files to be used while preserving root privileges.
2132 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
2133 that rogue child processes cannot use them.
2135 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
2136 run-time user, instead of root.
2138 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
2139 Exim run-time user without dropping privileges.
2141 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
2142 result string, instead of calling string_vformat() twice with the same
2145 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
2146 for other users. Others should always drop root privileges if they use
2147 -C on the command line, even for a whitelisted configure file.
2149 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
2151 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
2157 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
2158 $data_path, and $header_path variables; fixed documentation bugs and
2161 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
2162 exipick to access non-standard spools, including the "frozen" queue
2165 NM/01 Bugzilla 965: Support mysql stored procedures.
2166 Patch from Alain Williams
2168 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
2170 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
2171 Patch from Andreas Metzler
2173 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
2174 Patch from Kirill Miazine
2176 NM/05 Bugzilla 671: Added umask to procmail example.
2178 JJ/03 installed exipick 20100323.0, fixing doc bug
2180 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
2181 directory. Notification and patch from Dan Rosenberg.
2183 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
2185 TK/02 Improve log output when DKIM signing operation fails.
2187 MH/01 Treat the transport option dkim_domain as a colon separated
2188 list, not as a single string, and sign the message with each element,
2189 omitting multiple occurences of the same signer.
2191 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
2192 Bugzilla 985, 986. Patch by Simon Arlott
2194 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
2195 Patch by Simon Arlott
2197 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
2198 MBX locking. Notification from Dan Rosenberg.
2204 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
2206 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
2208 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
2210 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
2212 NM/04 Bugzilla 915: Flag broken perl installation during build.
2218 TK/01 Added patch by Johannes Berg that expands the main option
2219 "spamd_address" if it starts with a dollar sign.
2221 TK/02 Write list of recipients to X-Envelope-Sender header when building
2222 the mbox-format spool file for content scanning (suggested by Jakob
2225 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
2226 (http://www.dcc-servers.net/) support via dccifd. Activated by
2227 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
2229 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
2230 by Mark Daniel Reidel <mr@df.eu>.
2232 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
2233 When building exim an external PCRE library is now needed -
2234 PCRE is a system library on the majority of modern systems.
2235 See entry on PCRE_LIBS in EDITME file.
2237 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
2238 conversation. Added nologin parameter to request.
2239 Patch contributed by Kirill Miazine.
2241 TF/01 Do not log submission mode rewrites if they do not change the address.
2243 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
2245 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
2246 log files in place. Contributed by Roberto Lima.
2248 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
2250 TF/03 Bugzilla 615: When checking the local_parts router precondition
2251 after a local_part_suffix or local_part_prefix option, Exim now
2252 does not use the address's named list lookup cache, since this
2253 contains cached lookups for the whole local part.
2255 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
2256 Robert Millan. Documentation is in experimental-spec.txt.
2258 TF/04 Bugzilla 668: Fix parallel build (make -j).
2260 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
2262 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
2263 Patch provided by Jan Srzednicki.
2265 TF/05 Leading white space used to be stripped from $spam_report which
2266 wrecked the formatting. Now it is preserved.
2268 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
2269 that they are available at delivery time.
2271 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
2273 TF/08 TLS error reporting now respects the incoming_interface and
2274 incoming_port log selectors.
2276 TF/09 Produce a more useful error message if an SMTP transport's hosts
2277 setting expands to an empty string.
2279 NM/06 Bugzilla 744: EXPN did not work under TLS.
2280 Patch provided by Phil Pennock.
2282 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
2283 Patch provided by Richard Godbee.
2285 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
2286 acl_smtp_notquit, added index entry.
2288 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
2289 Patch provided by Eugene Bujak.
2291 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
2292 accept(). Patch provided by Maxim Dounin.
2294 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
2295 Patch provided by Phil Pennock.
2297 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
2299 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
2300 Patch provided by Brad "anomie" Jorsch.
2302 NM/14 Bugzilla 622: Added timeout setting to transport filter.
2303 Patch provided by Dean Brooks.
2305 TK/05 Add native DKIM support (does not depend on external libraries).
2307 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
2308 Patch provided by Graeme Fowler.
2310 NM/16 Bugzilla 851: Documentation example syntax fix.
2312 NM/17 Changed NOTICE file to remove references to embedded PCRE.
2314 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
2317 NM/19 Bugzilla 745: TLS version reporting.
2318 Patch provided by Phil Pennock.
2320 NM/20 Bugzilla 167: bool: condition support.
2321 Patch provided by Phil Pennock.
2323 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
2324 clients. Patch provided by Phil Pennock.
2326 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
2327 Patch provided by Brad "anomie" Jorsch.
2329 NM/23 Bugzilla 687: Fix misparses in eximstats.
2330 Patch provided by Heiko Schlittermann.
2332 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
2333 Patch provided by Heiko Schlittermann.
2335 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
2336 plus update to original patch.
2338 NM/26 Bugzilla 799: Documentation correction for ratelimit.
2340 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
2341 Patch provided by David Brownlee.
2343 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
2345 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
2347 NM/30 Bugzilla 888: TLS documentation bugfixes.
2349 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
2351 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
2352 Unlike the original bugzilla I have changed all shell scripts in src tree.
2354 NM/33 Bugzilla 898: Transport filter timeout fix.
2355 Patch by Todd Rinaldo.
2357 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
2358 Patch by Serge Demonchaux.
2360 NM/35 Bugzilla 39: Base64 decode bug fixes.
2361 Patch by Jakob Hirsch.
2363 NM/36 Bugzilla 909: Correct connect() call in dcc code.
2365 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
2367 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
2369 NM/39 Bugzilla 911: Fixed MakeLinks build script.
2375 TK/01 Add preliminary DKIM support. Currently requires a forked version of
2376 ALT-N's libdkim that I have put here:
2377 http://duncanthrax.net/exim-experimental/
2379 Note to Michael Haardt: I had to rename some vars in sieve.c. They
2380 were called 'true' and it seems that C99 defines that as a reserved
2381 keyword to be used with 'bool' variable types. That means you could
2382 not include C99-style headers which use bools without triggering
2383 build errors in sieve.c.
2385 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
2386 as mailq or other aliases. Changed the --help handling significantly
2387 to do whats expected. exim_usage() emits usage/help information.
2389 SC/01 Added the -bylocaldomain option to eximstats.
2391 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
2393 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
2395 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
2401 PH/01 Another patch from the Sieve maintainer.
2403 PH/02 When an IPv6 address is converted to a string for single-key lookup
2404 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
2405 dots are used instead of colons so that keys in lsearch files need not
2406 contain colons. This was done some time before quoting was made available
2407 in lsearch files. However, iplsearch files do require colons in IPv6 keys
2408 (notated using the quote facility) so as to distinguish them from IPv4
2409 keys. This meant that lookups for IP addresses in host lists did not work
2410 for iplsearch lookups.
2412 This has been fixed by arranging for IPv6 addresses to be expressed with
2413 colons if the lookup type is iplsearch. This is not incompatible, because
2414 previously such lookups could never work.
2416 The situation is now rather anomalous, since one *can* have colons in
2417 ordinary lsearch keys. However, making the change in all cases is
2418 incompatible and would probably break a number of configurations.
2420 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
2423 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
2424 conversion specification without a maximum field width, thereby enabling
2425 a rogue spamd server to cause a buffer overflow. While nobody in their
2426 right mind would setup Exim to query an untrusted spamd server, an
2427 attacker that gains access to a server running spamd could potentially
2428 exploit this vulnerability to run arbitrary code as the Exim user.
2430 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
2431 $primary_hostname instead of what libspf2 thinks the hosts name is.
2433 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
2434 a directory entry by the name of the lookup key. Previously, if a
2435 symlink pointed to a non-existing file or a file in a directory that
2436 Exim lacked permissions to read, a lookup for a key matching that
2437 symlink would fail. Now it is enough that a matching directory entry
2438 exists, symlink or not. (Bugzilla 503.)
2440 PH/03 The body_linecount and body_zerocount variables are now exported in the
2443 PH/04 Added the $dnslist_matched variable.
2445 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
2446 This means they are set thereafter only if the connection becomes
2449 PH/06 Added the client_condition to authenticators so that some can be skipped
2450 by clients under certain conditions.
2452 PH/07 The error message for a badly-placed control=no_multiline_responses left
2453 "_responses" off the end of the name.
2455 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
2457 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
2458 (without spaces) instead of just copying the configuration text.
2460 PH/10 Added the /noupdate option to the ratelimit ACL condition.
2462 PH/11 Added $max_received_linelength.
2464 PH/12 Added +ignore_defer and +include_defer to host lists.
2466 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
2467 way in which PCRE > 7.0 is built.
2469 PH/14 Implemented queue_only_load_latch.
2471 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
2472 MAIL command. The effect was to mangle the value on 64-bit systems.
2474 PH/16 Another patch from the Sieve maintainer.
2476 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
2478 PH/18 If a system quota error occurred while trying to create the file for
2479 a maildir delivery, the message "Mailbox is full" was not appended to the
2480 bounce if the delivery eventually timed out. Change 4.67/27 below applied
2481 only to a quota excession during the actual writing of the file.
2483 PH/19 It seems that peer DN values may contain newlines (and other non-printing
2484 characters?) which causes problems in log lines. The DN values are now
2485 passed through string_printing() before being added to log lines.
2487 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
2488 and InterBase are left for another time.)
2490 PH/21 Added message_body_newlines option.
2492 PH/22 Guard against possible overflow in moan_check_errorcopy().
2494 PH/23 POSIX allows open() to be a macro; guard against that.
2496 PH/24 If the recipient of an error message contained an @ in the local part
2497 (suitably quoted, of course), incorrect values were put in $domain and
2498 $local_part during the evaluation of errors_copy.
2504 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
2505 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
2508 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
2509 issue a MAIL command.
2511 PH/02 In an ACL statement such as
2513 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
2515 if a client was not listed at all, or was listed with a value other than
2516 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
2517 the condition was not true (as it should be), so access was not denied.
2518 The bug was that the ! inversion was incorrectly passed on to the second
2519 item. This has been fixed.
2521 PH/03 Added additional dnslists conditions == and =& which are different from
2522 = and & when the dns lookup returns more than one IP address.
2524 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
2525 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
2527 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
2528 FSYNC, which compiles an option called disable_fsync that allows for
2529 bypassing fsync(). The documentation is heavily laced with warnings.
2531 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
2533 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
2534 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
2535 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
2536 including adding "make clean"; (3) Added -fPIC when compiling the test
2537 dynamically loaded module, to get rid of a warning.
2539 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
2540 message fails, move_frozen_messages = true and ignore_bounce_errors_after
2541 = 0s. The bug is otherwise harmless.
2543 PH/07 There was a bug in the dovecot authenticator such that the value of
2544 $auth1 could be overwritten, and so not correctly preserved, after a
2545 successful authentication. This usually meant that the value preserved by
2546 the server_setid option was incorrect.
2548 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
2550 PH/09 Installed PCRE release 7.0.
2552 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
2553 run for batched SMTP input. It is now run at the start of every message
2554 in the batch. While fixing this I discovered that the process information
2555 (output by running exiwhat) was not always getting set for -bs and -bS
2556 input. This is fixed, and it now also says "batched" for BSMTP.
2558 PH/11 Added control=no_pipelining.
2560 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
2561 patch, slightly modified), and move the expansion of helo_data till after
2562 the connection is made in the smtp transport (so it can use these
2565 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
2567 PH/14 Added log_selector = +pid.
2569 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
2571 PH/16 Add ${if forany and ${if forall.
2573 PH/17 Added dsn_from option to vary the From: line in DSNs.
2575 PH/18 Flush SMTP output before performing a callout, unless control =
2576 no_callout_flush is set.
2578 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
2579 was true (the default) a successful delivery failed to delete the retry
2580 item, thus causing premature timeout of the address. The bug is now
2583 PH/20 Added hosts_avoid_pipelining to the smtp transport.
2585 PH/21 Long custom messages for fakedefer and fakereject are now split up
2586 into multiline responses in the same way that messages for "deny" and
2587 other ACL rejections are.
2589 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
2590 with slight modification.
2592 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
2593 draft, changing the syntax and factoring some duplicate code".
2595 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
2596 for deliveries of the second and subsequent messages over the same SMTP
2599 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
2600 ${reduce, with only minor "tidies".
2602 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
2604 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
2605 expansion side effects.
2607 PH/27 When a message times out after an over-quota error from an Exim-imposed
2608 quota, the bounce message says "mailbox is full". This message was not
2609 being given when it was a system quota that was exceeded. It now should
2612 MH/03 Made $recipients available in local_scan(). local_scan() already has
2613 better access to the recipient list through recipients_list[], but
2614 $recipients can be useful in postmaster-provided expansion strings.
2616 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
2617 in the case of a MAIL command with additional options following the
2618 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
2619 were accidentally chopped off.
2621 PH/29 SMTP synchronization checks are implemented when a command is read -
2622 there is a check that no more input is waiting when there shouldn't be
2623 any. However, for some commands, a delay in an ACL can mean that it is
2624 some time before the response is written. In this time, more input might
2625 arrive, invalidly. So now there are extra checks after an ACL has run for
2626 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
2627 pipelining has not been advertised.
2629 PH/30 MH's patch to allow iscntrl() characters to be list separators.
2631 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
2632 returned in the SMTP response when smtp_return_error_details was false.
2633 This has been fixed.
2635 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
2636 instead of the C I/O that was originally supplied, because problems were
2637 reported on Solaris.
2639 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
2640 Exim which did not show up earlier: it was assuming that a call to
2641 SSL_CTX_set_info_callback() might give an error value. In fact, there is
2642 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
2643 was a macro that became an assignment, so it seemed to work. This has
2644 changed to a proper function call with a void return, hence the compile
2645 error. Exim's code has been fixed.
2647 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
2650 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
2652 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
2654 PH/37 If a message is not accepted after it has had an id assigned (e.g.
2655 because it turns out to be too big or there is a timeout) there is no
2656 "Completed" line in the log. When some messages of this type were
2657 selected by exigrep, they were listed as "not completed". Others were
2658 picked up by some special patterns. I have improved the selection
2659 criteria to be more general.
2661 PH/38 The host_find_failed option in the manualroute router can now be set
2662 to "ignore", to completely ignore a host whose IP address cannot be
2663 found. If all hosts are ignored, the behaviour is controlled by the new
2664 host_all_ignored option.
2666 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
2667 homing or because of multiple MX records with /mx) generated more than
2668 one IP address, and the following item turned out to be the local host,
2669 all the secondary addresses of the first item were incorrectly removed
2670 from the list, along with the local host and any following hosts (which
2671 is what is supposed to happen).
2673 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
2674 whoever called Exim into the -H file. In the case of the daemon it was
2675 behaving confusingly. When first started, it used values for whoever
2676 started the daemon, but after a SIGHUP it used the Exim user (because it
2677 calls itself on a restart). I have changed the code so that it now always
2680 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
2681 message are rejected with the same error (e.g. no authentication or bad
2682 sender address), and a DATA command is nevertheless sent (as can happen
2683 with PIPELINING or a stupid MUA), the error message that was given to the
2684 RCPT commands is included in the rejection of the DATA command. This is
2685 intended to be helpful for MUAs that show only the final error to their
2688 PH/42 Another patch from the Sieve maintainer.
2690 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
2691 Eximstats - Fixed some broken HTML links and added missing column headers
2693 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
2694 columns for Rejects, Temp Rejects, Ham, and Spam rows.
2696 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
2698 PH/43 Yet another patch from the Sieve maintainer.
2700 PH/44 I found a way to check for a TCP/IP connection going away before sending
2701 the response to the final '.' that terminates a message, but only in the
2702 case where the client has not sent further data following the '.'
2703 (unfortunately, this is allowed). However, in many cases there won't be
2704 any further data because there won't be any more messages to send. A call
2705 to select() can be used: if it shows that the input is "ready", there is
2706 either input waiting, or the socket has been closed. An attempt to read
2707 the next input character can distinguish the two cases. Previously, Exim
2708 would have sent an OK response which the client would never have see.
2709 This could lead to message repetition. This fix should cure that, at
2710 least in a lot of common cases.
2712 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
2713 advertised in response to EHLO.
2719 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
2720 fixed by 4.65/MH/01 (is this a record?) are fixed:
2722 (i) An empty string was always treated as zero by the numeric comparison
2723 operators. This behaviour has been restored.
2725 (ii) It is documented that the numeric comparison operators always treat
2726 their arguments as decimal numbers. This was broken in that numbers
2727 starting with 0 were being interpreted as octal.
2729 While fixing these problems I realized that there was another issue that
2730 hadn't been noticed. Values of message_size_limit (both the global option
2731 and the transport option) were treated as octal if they started with 0.
2732 The documentation was vague. These values are now always treated as
2733 decimal, and I will make that clear in the documentation.
2739 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
2740 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
2743 MH/01 Don't check that the operands of numeric comparison operators are
2744 integers when their expansion is in "skipping" mode (fixes bug
2745 introduced by 4.64-PH/07).
2747 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
2748 child addresses, Exim now panics and dies. Previously, because the count
2749 is held in a short int, deliveries were likely to be lost. As such a
2750 large number of recipients for a single message is ridiculous
2751 (performance will be very, very poor), I have chosen to impose a limit
2752 rather than extend the field.
2758 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
2759 leftover -K file (the existence of which was triggered by #402).
2760 While we were at it, introduced process PID as part of the -K
2761 filename. This should rule out race conditions when creating
2764 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
2765 processing considerably. Previous code took too long for large mails,
2766 triggering a timeout which in turn triggers #401.
2768 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
2769 in the DK code in transports.c. sendfile() is not really portable,
2770 hence the _LINUX specificness.
2772 TF/01 In the add_headers option to the mail command in an Exim filter,
2773 there was a bug that Exim would claim a syntax error in any
2774 header after the first one which had an odd number of characters
2777 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
2778 callout verification, Exim cached a "reject" for the entire domain. This
2779 is correct for most verifications, but it is not correct for a recipient
2780 verification with use_sender or use_postmaster set, because in that case
2781 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
2782 case of MAIL FROM:<> rejection from other early rejections (e.g.
2783 rejection of HELO). When verifying a recipient using a non-null MAIL
2784 address, the cache is ignored if it shows MAIL FROM:<> rejection.
2785 Whatever the result of the callout, the value of the domain cache is
2786 left unchanged (for any other kind of callout, getting as far as trying
2787 RCPT means that the domain itself is ok).
2789 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
2792 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
2793 manifest itself as EPIPE rather than ECONNECT. When tidying away a
2794 session, the daemon ignores ECONNECT errors and logs others; it now
2795 ignores EPIPE as well.
2797 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
2798 (quoted-printable decoding).
2800 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
2801 later the small subsequent patch to fix an introduced bug.
2803 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
2805 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
2807 PH/08 An error is now given if message_size_limit is specified negative.
2809 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
2810 to be given (somewhat) arbitrary names.
2812 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
2815 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
2816 miscellaneous code fixes
2818 PH/10 Added the log_reject_target ACL modifier to specify where to log
2821 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
2822 hostname. This is wrong, because it relates to the incoming message (and
2823 probably the interface on which it is arriving) and not to the outgoing
2824 callout (which could be using a different interface). This has been
2825 changed to use the value of the helo_data option from the smtp transport
2826 instead - this is what is used when a message is actually being sent. If
2827 there is no remote transport (possible with a router that sets up host
2828 addresses), $smtp_active_hostname is used.
2830 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
2831 tweaks were necessary in order to get it to work (see also 21 below):
2832 (a) The code assumed that strncpy() returns a negative number on buffer
2833 overflow, which isn't the case. Replaced with Exim's string_format()
2835 (b) There were several signed/unsigned issues. I just did the minimum
2836 hacking in of casts. There is scope for a larger refactoring.
2837 (c) The code used strcasecmp() which is not a standard C function.
2838 Replaced with Exim's strcmpic() function.
2839 (d) The code set only $1; it now sets $auth1 as well.
2840 (e) A simple test gave the error "authentication client didn't specify
2841 service in request". It would seem that Dovecot has changed its
2842 interface. Fortunately there's a specification; I followed it and
2843 changed what the client sends and it appears to be working now.
2845 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
2848 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
2849 address A is aliased to B and C, where B exists and C does not. Without
2850 -v the output is "A verified" because verification stops after a
2851 successful redirection if more than one address is generated. However,
2852 with -v the child addresses are also verified. Exim was outputting "A
2853 failed to verify" and then showing the successful verification for C,
2854 with its parentage. It now outputs "B failed to verify", showing B's
2855 parentage before showing the successful verification of C.
2857 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
2858 look up a TXT record in a specific list after matching in a combined
2861 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
2862 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
2863 they consult the DNS. I had assumed they would set it the way they
2864 wanted; and indeed my experiments on Linux seem to show that in some
2865 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
2866 To be on the safe side, however, I have now made the interface to
2867 host_find_byname() similar to host_find_bydns(), with an argument
2868 containing the DNS resolver options. The host_find_byname() function now
2869 sets these options at its start, just as host_find_bydns() does. The smtp
2870 transport options dns_qualify_single and dns_search_parents are passed to
2871 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
2872 of host_find_byname() use the default settings of RES_DEFNAMES
2873 (qualify_single) but not RES_DNSRCH (search_parents).
2875 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
2876 spool_read_header() do less string testing, by means of a preliminary
2877 switch on the second character of optional "-foo" lines. (This is
2878 overdue, caused by the large number of possibilities that now exist.
2879 Originally there were few.) While I was there, I also converted the
2880 str(n)cmp tests so they don't re-test the leading "-" and the first
2881 character, in the hope this might squeeze out yet more improvement.
2883 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
2884 flag allowing group syntax was set by the header_syntax check but not
2885 turned off, possible causing trouble later; (2) The flag was not being
2886 set at all for the header_verify test, causing "group"-style headers to
2887 be rejected. I have now set it in this case, and also caused header_
2888 verify to ignore an empty address taken from a group. While doing this, I
2889 came across some other cases where the code for allowing group syntax
2890 while scanning a header line wasn't quite right (mostly, not resetting
2891 the flag correctly in the right place). These bugs could have caused
2892 trouble for malformed header lines. I hope it is now all correct.
2894 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
2895 with the "reply" argument non-NULL. The code, however (which originally
2896 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
2897 but it didn't always do it. This confused somebody who was copying the
2898 code for some other use. I have removed all the tests.
2900 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
2901 feature that was used to support insecure browsers during the U.S. crypto
2902 embargo. It requires special client support, and Exim is probably the
2903 only MTA that supported it -- and would never use it because real RSA is
2904 always available. This code has been removed, because it had the bad
2905 effect of slowing Exim down by computing (never used) parameters for the
2906 RSA_EXPORT functionality.
2908 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
2909 authenticator to fail if there's a tab character in the incoming data
2910 (there should never be unless someone is messing about, as it's supposed
2911 to be base64-encoded). Also added, on Timo's advice, the "secured" option
2912 if the connection is using TLS or if the remote IP is the same as the
2913 local IP, and the "valid-client-cert option" if a client certificate has
2916 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
2917 authenticators. This can be used for authorization after authentication
2918 succeeds. (In the case of plaintext, it servers for both authentication
2921 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
2922 if any retry times were supplied.
2924 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
2925 connection, where there is no client IP address to check. In this
2926 situation, the verify now always succeeds.
2928 PH/25 Applied John Jetmore's -Mset patch.
2930 PH/26 Added -bem to be like -Mset, but loading a message from a file.
2932 PH/27 In a string expansion for a processed (not raw) header when multiple
2933 headers of the same name were present, leading whitespace was being
2934 removed from all of them, but trailing whitespace was being removed only
2935 from the last one. Now trailing whitespace is removed from each header
2936 before concatenation. Completely empty headers in a concatenation (as
2937 before) are ignored.
2939 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
2940 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
2942 PH/29 [Removed. This was a change that I later backed out, and forgot to
2943 correct the ChangeLog entry (that I had efficiently created) before
2944 committing the later change.]
2946 PH/30 Exim was sometimes attempting to deliver messages that had suffered
2947 address errors (4xx response to RCPT) over the same connection as other
2948 messages routed to the same hosts. Such deliveries are always "forced",
2949 so retry times are not inspected. This resulted in far too many retries
2950 for the affected addresses. The effect occurred only when there were more
2951 hosts than the hosts_max_try setting in the smtp transport when it had
2952 the 4xx errors. Those hosts that it had tried were not added to the list
2953 of hosts for which the message was waiting, so if all were tried, there
2954 was no problem. Two fixes have been applied:
2956 (i) If there are any address or message errors in an SMTP delivery, none
2957 of the hosts (tried or untried) are now added to the list of hosts
2958 for which the message is waiting, so the message should not be a
2959 candidate for sending over the same connection that was used for a
2960 successful delivery of some other message. This seems entirely
2961 reasonable: after all the message is NOT "waiting for some host".
2962 This is so "obvious" that I'm not sure why it wasn't done
2963 previously. Hope I haven't missed anything, but it can't do any
2964 harm, as the worst effect is to miss an optimization.
2966 (ii) If, despite (i), such a delivery is accidentally attempted, the
2967 routing retry time is respected, so at least it doesn't keep
2968 hammering the server.
2970 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
2971 in ${readsocket because some servers need this prod.
2973 PH/32 Added some extra debug output when updating a wait-xxx database.
2975 PH/33 The hint "could be header name not terminated by colon", which has been
2976 given for certain expansion errors for a long time, was not being given
2977 for the ${if def:h_colon_omitted{... case.
2979 PH/34 The spec says: "With one important exception, whenever a domain list is
2980 being scanned, $domain contains the subject domain." There was at least
2981 one case where this was not true.
2983 PH/35 The error "getsockname() failed: connection reset by peer" was being
2984 written to the panic log as well as the main log, but it isn't really
2985 panic-worthy as it just means the connection died rather early on. I have
2986 removed the panic log writing for the ECONNRESET error when getsockname()
2989 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
2990 runs only) independently of the message's sender address. This meant
2991 that, if the 4xx error was in fact related to the sender, a different
2992 message to the same recipient with a different sender could confuse
2993 things. In particular, this can happen when sending to a greylisting
2994 server, but other circumstances could also provoke similar problems.
2995 I have changed the default so that the retry time for these errors is now
2996 based a combination of the sender and recipient addresses. This change
2997 can be overridden by setting address_retry_include_sender=false in the
3000 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
3001 remote server are returned as part of bounce messages. This was not
3002 happening for LMTP over a pipe (the lmtp transport), but now it is the
3003 same for both kinds of LMTP.
3005 PH/38 Despite being documented as not happening, Exim was rewriting addresses
3006 in header lines that were in fact CNAMEs. This is no longer the case.
3008 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
3009 and queue runs started by the daemon processed all messages. This has
3010 been fixed so that -R and -S can now usefully be given with -q<time>.
3012 PH/40 Import PCRE release 6.7 (fixes some bugs).
3014 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
3016 PH/42 Give an error if -q is specified more than once.
3018 PH/43 Renamed the variables $interface_address and $interface_port as
3019 $received_ip_address and $received_port, to make it clear that these
3020 values apply to message reception, and not to the outgoing interface when
3021 a message is delivered. (The old names remain recognized, of course.)
3023 PH/44 There was no timeout on the connect() call when using a Unix domain
3024 socket in the ${readsocket expansion. There now is.
3026 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
3027 be meaningful with "accept".
3029 SC/01 Eximstats V1.43
3030 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
3032 SC/02 Eximstats V1.44
3033 Use a glob alias rather than an array ref in the generated
3034 parser. This improves both readability and performance.
3036 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
3037 Collect SpamAssassin and rejection statistics.
3038 Don't display local sender or destination tables unless
3039 there is data to show.
3040 Added average volumes into the top table text output.
3042 SC/04 Eximstats V1.46
3043 Collect data on the number of addresses (recipients)
3044 as well as the number of messages.
3046 SC/05 Eximstats V1.47
3047 Added 'Message too big' to the list of mail rejection
3048 reasons (thanks to Marco Gaiarin).
3050 SC/06 Eximstats V1.48
3051 Mainlog lines which have GMT offsets and are too short to
3052 have a flag are now skipped.
3054 SC/07 Eximstats V1.49 (Alain Williams)
3055 Added the -emptyok flag.
3057 SC/08 Eximstats V1.50
3058 Fixes for obtaining the IP address from reject messages.
3060 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
3061 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
3062 whitespace changes from 4.64-PH/27
3064 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
3067 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
3068 are found, allow negative numbers in numeric criteria)
3070 JJ/06 exipick.20061117.2, added new $message_body_missing variable
3072 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
3073 to match changes made in 4.64-PH/43
3075 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
3077 PH/47 Put in an explicit test for a DNS lookup of an address record where the
3078 "domain" is actually an IP address, and force a failure. This locks out
3079 those revolvers/nameservers that support "A-for-A" lookups, in
3080 contravention of the specifications.
3082 PH/48 When a host name was looked up from an IP address, and the subsequent
3083 forward lookup of the name timed out, the host name was left in
3084 $sender_host_name, contrary to the specification.
3086 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
3087 restricted to single-key lookups, Exim was not diagnosing an error if
3088 * or *@ was used with a query-style lookup.
3090 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
3092 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
3093 long ago, but noone interested enough thought of it. Let's just say that
3094 the "1.1" means that there are some new functions that weren't there at
3095 some point in the past.
3097 PH/51 Error processing for expansion failure of helo_data from an smtp
3098 transport during callout processing was broken.
3100 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
3101 tested/used via the -bh/-bhc/-bs options.
3103 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
3104 bug, fixed in subsequent PCRE releases).
3106 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
3107 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
3109 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
3115 SC/01 Use a glob alias rather than an array ref in eximstats generated
3116 parser. This improves both readability and performance.
3118 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
3119 Don't display local sender or destination tables in eximstats unless
3120 there is data to show.
3121 Added average volumes into the eximstats top table text output.
3123 SC/03 Collect data on the number of addresses (recipients) as well
3124 as the number of messages in eximstats.
3126 TF/01 Correct an error in the documentation for the redirect router. Exim
3127 does (usually) call initgroups() when daemonizing.
3129 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
3130 with consistent privilege compared to when running as a daemon.
3132 TF/03 Note in the spec that $authenticated_id is not set for local
3133 submissions from trusted users.
3135 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
3136 Thanks to Dean Brooks <dean@iglou.com> for the patch.
3138 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
3139 by adding some example configuration directives to the default
3140 configuration file. A little bit of work is required to uncomment the
3141 directives and define how usernames and passwords are checked, but
3142 there is now a framework to start from.
3144 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
3145 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
3146 without this. I don't know how relevant this is to other LDAP libraries.
3148 PH/02 Add the verb name to the "unknown ACL verb" error.
3150 PH/03 Magnus Holmgren's patch for filter_prepend_home.
3152 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
3154 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
3155 directory not expanded when it should be if an expanded home directory
3156 was set for the address (which is overridden by the transport).
3158 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
3161 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
3162 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
3163 because it is too late at that time, and has no effect.
3165 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
3166 security issue with \' (bugzilla #107). I could not use the
3167 PQescapeStringConn() function, because it needs a PGconn value as one of
3170 PH/08 When testing addresses using -bt, indicate those final addresses that
3171 are duplicates that would not cause an additional delivery. At least one
3172 person was confused, thinking that -bt output corresponded to deliveries.
3173 (Suppressing duplicates isn't a good idea as you lose the information
3174 about possibly different redirections that led to the duplicates.)
3176 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
3177 systems where poll() doesn't work, in particular OS X.
3179 PH/10 Added more information to debugging output for retry time not reached.
3181 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
3182 operations in malware.c.
3184 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
3187 PH/13 If write_rejectlog was set false when logging was sent to syslog with
3188 syslog_duplication set false, log lines that would normally be written
3189 both the the main log and to the reject log were not written to syslog at
3192 PH/14 In the default configuration, change the use of "message" in ACL warn
3193 statements to "add_header".
3195 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
3196 not followed by a command (e.g. "seen endif").
3198 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
3199 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
3202 PH/17 Added extra conditions to the default value of delay_warning_condition
3206 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
3207 { match{$h_precedence:}{(?i)bulk|list|junk} } \
3208 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
3211 The Auto-Submitted: and various List- headers are standardised, whereas I
3212 don't think Precedence: ever was.
3214 PH/18 Refactored debugging code in route_finduser() to show more information,
3215 in particular, the error code if getpwnam() issues one.
3217 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
3218 This is apparently needed in addition to the PH/07 change above to avoid
3219 any possible encoding problems.
3221 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
3222 but not after initializing Perl.
3224 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
3225 output them only if debugging. By default they are written stderr,
3226 apparently, which is not desirable.
3228 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
3231 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
3234 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
3236 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
3237 authenticated or an ident call has been made. Suppress the default
3238 values for $authenticated_id and $authenticated_sender (but permit -oMai
3239 and -oMas) when testing with -bh.
3241 PH/24 Re-jigged the order of the tests in the default configuration so that the
3242 tests for valid domains and recipients precede the DNS black list and CSA
3243 tests, on the grounds that those ones are more expensive.
3245 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
3246 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
3247 This bug exists in every version of Exim that I still have, right back to
3250 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
3251 However, an attempt to turn on case-sensitivity in a regex key by
3252 including (?-i) didn't work because the subject string was already
3253 lowercased, and the effects were non-intuitive. It turns out that a
3254 one-line patch can be used to allow (?-i) to work as expected.
3260 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
3261 other effects) broke the use of negated acl sub-conditions.
3263 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
3266 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
3267 "Deny" causes Exim to reject the incoming connection with a 554 error.
3268 Unfortunately, if there is a major crisis, such as a disk failure,
3269 tcp-wrappers gives "deny", whereas what one would like would be some
3270 kind of temporary error. A kludge has been added to help with this.
3271 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
3272 554 error is used if errno is still zero or contains ENOENT (which occurs
3273 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
3276 PH/03 Add -lutil to the default FreeBSD LIBS setting.
3278 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
3279 errors. Otherwise a message that provokes a temporary error (when other
3280 messages do not) can cause a whole host to time out.
3282 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
3283 addresses were routed directly to files or pipes from a redirect router.
3284 File deliveries just didn't batch; pipe deliveries might have suffered
3287 PH/06 A failure to get a lock for a hints database would erroneously always say
3288 "Failed to get write lock", even when it was really a read lock.
3290 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
3291 of 0600. This has been changed to use the value of the lockfile_mode
3292 option (which defaults to 0600).
3294 PH/08 Applied small patch from the Sieve maintainer.
3296 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
3297 folder from quota calculations, a direct delivery into this folder messed
3298 up the contents of the maildirsize file. This was because the regex was
3299 used only to exclude .Trash (or whatever) when the size of the mailbox
3300 was calculated. There was no check that a delivery was happening into an
3301 excluded directory. This bug has been fixed by ignoring all quota
3302 processing for deliveries into excluded directories.
3304 PH/10 Added the maildirfolder_create_regex option to appendfile.
3310 PH/01 The code for finding all the local interface addresses on a FreeBSD
3311 system running IPv6 was broken. This may well have applied to all BSD
3312 systems, as well as to others that have similar system calls. The broken
3313 code found IPv4 interfaces correctly, but gave incorrect values for the
3314 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
3315 that it would not match correctly against @[] and not recognize the IPv6
3318 PH/02 The ipliteral router was not recognizing addresses of the form user@
3319 [ipv6:....] because it didn't know about the "ipv6:" prefix.
3321 PH/03 Added disable_ipv6.
3323 PH/04 Changed $reply_address to use the raw form of the headers instead of the
3324 decoded form, because it is most often used to construct To: headers
3325 lines in autoreplies, and the decoded form may well be syntactically
3326 invalid. However, $reply_address has leading white space removed, and all
3327 newlines turned into spaces so that the autoreply transport does not
3330 PH/05 If group was specified without a user on a router, and no group or user
3331 was specified on a transport, the group from the router was ignored.
3333 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
3334 for visible compile-time settings that can be used to change these
3335 numbers, for those that want even more. Backwards compatibility with old
3336 spool files has been maintained. However, going back to a previous Exim
3337 release will lost any variables that are in spool files.
3339 PH/07 Two small changes when running in the test harness: increase delay when
3340 passing a TCP/IP connection to a new process, in case the original
3341 process has to generate a bounce, and remove special handling of
3342 127.0.0.2 (sic), which is no longer necessary.
3344 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
3345 be the same on different OS.
3347 PH/09 Moved a debug statement in filter processing to avoid a race problem when
3350 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
3351 whether --show-vars was specified or not
3353 JJ/02 exipick: Added support for new ACL variable spool format introduced
3356 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
3357 syntactically invalid From: or Reply-to: line, and a filter used this to
3358 generate an autoreply, and therefore failed to obtain an address for the
3359 autoreply, Exim could try to deliver to a non-existent relative file
3360 name, causing unrelated and misleading errors. What now happens is that
3361 it logs this as a hard delivery error, but does not attempt to create a
3364 PH/11 The exinext utility has a -C option for testing purposes, but although
3365 the given file was scanned by exinext itself; it wasn't being passed on
3366 when Exim was called.
3368 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
3369 an end-of-file indication when reading a command response.
3371 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
3372 compiled. In many other places in Exim, IPv6 addresses are always
3373 recognized, so I have changed this. It also means that IPv4 domain
3374 literals of the form [IPV4:n.n.n.n] are now always recognized.
3376 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
3377 used if the router is not running as root, for example, when verifying at
3378 ACL time, or when using -bh. The debugging output from this situation was
3379 non-existent - all you got was a failure to exec. I have made two
3382 (a) Failures to set uid/gid, the current directory, or a process leader
3383 in a subprocess such as that created by queryprogram now generate
3384 suitable debugging output when -d is set.
3386 (b) The queryprogram router detects when it is not running as root,
3387 outputs suitable debugging information if -d is set, and then runs
3388 the subprocess without attempting to change uid/gid.
3390 PH/15 Minor change to Makefile for building test_host (undocumented testing
3393 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
3394 additional section of a DNS packet that returns MX or SRV records.
3395 Instead, it always explicitly searches for A/AAAA records. This avoids
3396 major problems that occur when a DNS server includes only records of one
3397 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
3398 fixed another bug: if SRV records were looked up and the corresponding
3399 address records were *not* found in the additional section, the port
3400 values from the SRV records were lost.
3402 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
3403 using the correct key (the original address) when searching the retry
3404 rules in order to find which one to use for generating the retry hint.
3406 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
3407 adding the default one. Similarly, if it contains a Reply-To: header, the
3408 errors_reply_to option, if set, is not used.
3410 PH/19 When calculating a retry time, Exim used to measure the "time since
3411 failure" by looking at the "first failed" field in the retry record. Now
3412 it does not use this if it is later than than the arrival time of the
3413 message. Instead it uses the arrival time. This makes for better
3414 behaviour in cases where some deliveries succeed, thus re-setting the
3415 "first failed" field. An example is a quota failure for a huge message
3416 when small messages continue to be delivered. Without this change, the
3417 "time since failure" will always be short, possible causing more frequent
3418 delivery attempts for the huge message than are intended.
3419 [Note: This change was subsequently modified - see PH/04 for 4.62.]
3421 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
3422 $1, $2, $3) because the numerical variables can be reset during some
3423 expansion items (e.g. "match"), thereby losing the authentication data.
3425 PH/21 Make -bV show the size of off_t variables so that the test suite can
3426 decide whether to run tests for quotas > 2G.
3428 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
3429 mailbox_size, and mailbox_filecount in the appendfile transport. If a
3430 filecount value is greater than 2G or if a quota value is greater than 2G
3431 on a system where the size of off_t is not greater than 4, a panic error
3434 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
3435 never match. The debug and -bh output now contains an explicit error
3436 message indicating a malformed IPv4 address or mask.
3438 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
3439 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
3440 PH/23 above applies.
3442 PH/25 Do not write to syslog when running in the test harness. The only
3443 occasion when this arises is a failure to open the main or panic logs
3444 (for which there is an explicit test).
3446 PH/26 Added the /no_tell option to "control=freeze".
3448 PH/27 If a host name lookup failed very early in a connection, for example, if
3449 the IP address matched host_lookup and the reverse lookup yielded a name
3450 that did not have a forward lookup, an error message of the form "no IP
3451 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
3452 could be logged. Now it outputs the IP address instead of "NULL".
3454 PH/28 An enabling patch from MH: add new function child_open_exim2() which
3455 allows the sender and the authenticated sender to be set when
3456 submitting a message from within Exim. Since child_open_exim() is
3457 documented for local_scan(), the new function should be too.
3459 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
3460 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
3461 results in an empty string is now treated as unset.
3463 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
3465 PH/31 Added sender_verify_fail logging option.
3467 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
3468 needed by RFC 822 but not by RFC 2822 was commented out. I have now
3469 tidied the source and removed it altogether.
3471 PH/33 When a queue run was abandoned because the load average was too high, a
3472 log line was always written; now it is written only if the queue_run log
3473 selector is set. In addition, the log line for abandonment now contains
3474 information about the queue run such as the pid. This is always present
3475 in "start" and "stop" lines but was omitted from the "abandon" line.
3477 PH/34 Omit spaces between a header name and the colon in the error message that
3478 is given when verify = headers_syntax fails (if there are lots of them,
3479 the message gets confusing).
3481 PH/35 Change the default for dns_check_names_pattern to allow slashes within
3482 names, as there are now some PTR records that contain slashes. This check
3483 is only to protect against broken name servers that fall over on strange
3484 characters, so the fact that it applies to all lookups doesn't matter.
3486 PH/36 Now that the new test suite is complete, we can remove some of the
3487 special code in Exim that was needed for the old test suite. For example,
3488 sorting DNS records because real resolvers return them in an arbitrary
3489 order. The new test suite's fake resolver always returns records in the
3492 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
3493 bounces) except when queue_only is set, to avoid logging races between
3494 the different processes.
3496 PH/38 Panic-die if .include specifies a non-absolute path.
3498 PH/39 A tweak to the "H" retry rule from its user.
3500 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
3501 a label. They prevented compilation on older perls.
3503 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
3504 a warning to be raised on newish perls.
3506 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
3507 on queue. Changes to match documented behaviour of showing count of
3508 messages matching specified criteria.
3510 PH/40 Changed the default ident timeout from 30s to 5s.
3512 PH/41 Added support for the use of login_cap features, on those BSD systems
3513 that have them, for controlling the resources used by pipe deliveries.
3515 PH/42 The content-scanning code uses fopen() to create files in which to put
3516 message data. Previously it was not paying any attention to the mode of
3517 the files. Exim runs with umask(0) because the rest of the code creates
3518 files with open(), and sets the required mode explicitly. Thus, these
3519 files were ending up world-writeable. This was not a big issue, because,
3520 being within the spool directory, they were not world-accessible. I have
3521 created a function called modefopen, which takes an additional mode
3522 argument. It sets umask(777), creates the file, chmods it to the required
3523 mode, then resets the umask. All the relevant calls to fopen() in the
3524 content scanning code have been changed to use this function.
3526 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
3527 to 24 hours. This avoids potential overflow problems when processing G
3528 and H retry rules. I suspect nobody ever tinkers with this value.
3530 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
3532 PH/45 When the plaintext authenticator is running as a client, the server's
3533 challenges are checked to ensure they are valid base64 strings. By
3534 default, the authentication attempt is cancelled if an invalid string is
3535 received. Setting client_ignore_invalid_base64 true ignores these errors.
3536 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
3537 they are received. Thus, the responses can be made to depend on the
3538 challenges. If an invalid string is ignored, an empty string is placed in
3541 PH/46 Messages that are created by the autoreply transport now contains a
3542 References: header, in accordance with RFCs 2822 and 3834.
3544 PH/47 Added authenticated_sender_force to the smtp transport.
3546 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
3548 PH/49 Installed latest patch from the Sieve maintainer.
3550 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
3551 was also set, the appendfile transport was unnecessarily scanning a
3552 directory of message files (e.g. for maildir delivery) to find the count
3553 of files (along with the size), even though it did not need this
3554 information. It now does the scan only if it needs to find either the
3555 size of the count of files.
3557 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
3559 PH/52 Two bugs concerned with error handling when the smtp transport is
3562 (i) Exim was not creating retry information for temporary errors given
3563 for individual recipients after the DATA command when the smtp transport
3564 was used in LMTP mode. This meant that they could be retried too
3565 frequently, and not timed out correctly.
3567 (ii) Exim was setting the flag that allows error details to be returned
3568 for LMTP errors on RCPT commands, but not for LMTP errors for individual
3569 recipients that were returned after the DATA command.
3571 PH/53 This is related to PH/52, but is more general: for any failing address,
3572 when detailed error information was permitted to be returned to the
3573 sender, but the error was temporary, then after the final timeout, only
3574 "retry timeout exceeded" was returned. Now it returns the full error as
3575 well as "retry timeout exceeded".
3577 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
3578 do this, and (what is worse) MTAs that accept it.
3580 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
3581 will now be deprecated.
3583 PH/56 New os.c-cygwin from the Cygwin maintainer.
3585 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
3586 formats (previously only available in exim formats via -bpr, -bpru,
3587 and -bpra. Now also available in native and exiqgrep formats)
3589 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
3590 with very large, slow to parse queues
3592 JJ/08 exipick: added ! as generic prefix to negate any criteria format
3594 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
3596 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
3597 responses to authentication challenges, though it was showing the
3598 challenges; (ii) I've removed the CR characters from the debug output for
3601 PH/58 Allow for the insertion of a newline as well as a space when a string
3602 is turned into more than one encoded-word during RFC 2047 encoding. The
3603 Sieve code now uses this.
3605 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
3606 data_4xx, lost_connection, tls_required.
3608 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
3609 message was being sent as an SMTP response.
3611 PH/61 Add -l and -k options to exicyclog.
3613 PH/62 When verifying, if an address was redirected to one new address, so that
3614 verification continued, and the new address failed or deferred after
3615 having set something in $address_data, the value of $address_data was not
3616 passed back to the ACL. This was different to the case when no
3617 redirection occurred. The value is now passed back in both cases.
3619 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
3620 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
3621 use login_cap.h, so on its own it isn't the distinguishing feature. The
3622 new name refers directly to the setclassresources() function.
3624 PH/65 Added configuration files for NetBSD3.
3626 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
3628 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
3629 is preferred over IPv4.
3631 PH/68 The bounce_return_message and bounce_return_body options were not being
3632 honoured for bounces generated during the reception of non-SMTP messages.
3633 In particular, this applied to messages rejected by the ACL. This bug has
3634 been fixed. However, if bounce_return_message is true and bounce_return_
3635 body is false, the headers that are returned for a non-SMTP message
3636 include only those that have been read before the error was detected.
3637 (In the case of an ACL rejection, they have all been read.)
3639 PH/69 The HTML version of the specification is now built in a directory called
3640 spec_html instead of spec.html, because the latter looks like a path with
3641 a MIME-type, and this confuses some software.
3643 PH/70 Catch two compiler warnings in sieve.c.
3645 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
3646 function verify_get_ident() calls ip_connect() to connect a socket, but
3647 if the "connect()" function timed out, ip_connect() used to close the
3648 socket. However, verify_get_ident() also closes the socket later, and in
3649 between Exim writes to the log, which may get opened at this point. When
3650 the socket was closed in ip_connect(), the log could get the same file
3651 descriptor number as the socket. This naturally causes chaos. The fix is
3652 not to close the socket in ip_connect(); the socket should be closed by
3653 the function that creates it. There was only one place in the code where
3654 this was missing, in the iplookup router, which I don't think anybody now
3655 uses, but I've fixed it anyway.
3657 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
3658 well as to direct DNS lookups. Otherwise the handling of names in host
3659 lists is inconsistent and therefore confusing.
3665 PH/01 Two changes to the default runtime configuration:
3667 (1) Move the checks for relay_from_hosts and authenticated clients from
3668 after to before the (commented out) DNS black list checks.
3670 (2) Add control=submission to the relay_from_hosts and authenticated
3671 clients checks, on the grounds that messages accepted by these
3672 statements are most likely to be submissions.
3674 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
3676 (1) Generate an error if the third argument for the ${prvs expansion is
3679 (2) Treat a missing third argument of ${prvscheck as if it were an empty
3682 (3) Reset the variables that are obtained from the first argument of
3683 ${prvscheck and used in the second argument before leaving the code,
3684 because their memory is reclaimed, so using them afterwards may do
3687 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
3688 one (it's much easier than Tom thought :-).
3690 (5) Because of (4), we can now allow for the use of $prvscheck_result
3691 inside the third argument.
3693 PH/03 For some reason, the default setting of PATH when running a command from
3694 a pipe transport was just "/usr/bin". I have changed it to
3697 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
3698 anything to be listed in the output from -bV.
3700 PH/05 When a filter generated an autoreply, the entire To: header line was
3701 quoted in the delivery log line, like this:
3703 => >A.N.Other <ano@some.domain> <original@ddress> ...
3705 This has been changed so that it extracts the operative address. There
3706 may be more than one such address. If so, they are comma-separated, like
3709 => >ano@some.domain,ona@other.domain <original@ddress> ...
3711 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
3712 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
3713 not being looked up in the rDNS to get a host name, Exim was showing the
3714 IP address twice in Received: lines, even though the IP addresses were
3715 identical. For example:
3717 Received: from [1.2.3.4] (helo=[1.2.3.4])
3719 However, if the real host name was known, it was omitting the HELO data
3720 if it matched the actual IP address. This has been tidied up so that it
3721 doesn't show the same IP address twice.
3723 PH/07 When both +timestamp and +memory debugging was on, the value given by
3724 $tod_xxx expansions could be wrong, because the tod_stamp() function was
3725 called by the debug printing, thereby overwriting the timestamp buffer.
3726 Debugging no longer uses the tod_stamp() function when +timestamp is set.
3728 PH/08 When the original message was included in an autoreply transport, it
3729 always said "this is a copy of the message, including all the headers",
3730 even if body_only or headers_only was set. It now gives an appropriate
3733 PH/09 Applied a patch from the Sieve maintainer which:
3735 o fixes some comments
3736 o adds the (disabled) notify extension core
3737 o adds some debug output for the result of if/elsif tests
3738 o points to the current vacation draft in the documentation
3739 and documents the missing references header update
3743 o fixes a bug in processing the envelope test (when testing
3744 multiple envelope elements, the last element determined the
3747 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
3748 Electronic Mail") by including:
3750 Auto-submitted: auto-generated
3752 in the messages that it generates (bounce messages and others, such as
3753 warnings). In the case of bounce messages for non-SMTP messages, there was
3754 also a typo: it was using "Auto_submitted" (underscore instead of
3755 hyphen). Since every message generated by Exim is necessarily in response
3756 to another message, thes have all been changed to:
3758 Auto-Submitted: auto-replied
3760 in accordance with these statements in the RFC:
3762 The auto-replied keyword:
3764 - SHOULD be used on messages sent in direct response to another
3765 message by an automatic process,
3767 - MUST NOT be used on manually-generated messages,
3769 - MAY be used on Delivery Status Notifications (DSNs) and Message
3770 Disposition Notifications (MDNs),
3772 - MUST NOT be used on messages generated by automatic or periodic
3773 processes, except for messages which are automatic responses to
3776 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
3777 to the default Received: header definition.
3779 PH/12 Added log selector acl_warn_skipped (default on).
3781 PH/13 After a successful wildlsearch lookup, discard the values of numeric
3782 variables because (a) they are in the wrong storage pool and (b) even if
3783 they were copied, it wouldn't work properly because of the caching.
3785 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
3786 checking when decoding. Apparently there are clients that generate
3787 overlong encoded strings. Why am I not surprised?
3789 PH/15 If the first argument of "${if match_address" was not empty, but did not
3790 contain an "@" character, Exim crashed. Now it writes a panic log message
3791 and treats the condition as false.
3793 PH/16 In autoreply, treat an empty string for "once" the same as unset.
3795 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
3796 extension "envelope-auth". The code is finished and in agreement with
3797 other implementations, but there is no documentation so far and in fact,
3798 nobody wrote the draft yet. This extension is currently #undef'ed, thus
3799 not changing the active code.
3801 Print executed "if" and "elsif" statements when debugging is used. This
3802 helps a great deal to understand what a filter does.
3804 Document more things not specified clearly in RFC3028. I had all this
3805 sorted out, when out of a sudden new issues came to my mind. Oops."
3807 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
3810 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
3811 canonical form (as documented). However, after a host name lookup from
3812 the IP address, check_host() was doing a simple string comparison with
3813 addresses acquired from the DNS when checking that the found name did
3814 have the original IP as one of its addresses. Since any found IPv6
3815 addresses are likely to be in abbreviated form, the comparison could
3816 fail. Luckily, there already exists a function for doing the comparison
3817 by converting both addresses to binary, so now that is used instead of
3818 the text comparison.
3820 PH/20 There was another similar case to PH/19, when a complete host name was
3821 given in a host list; looking up its IP address could give an abbreviated
3822 form, whereas the current host's name might or might not be abbreviated.
3823 The same fix has been applied.
3829 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
3830 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
3833 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
3834 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
3836 PH/03 Typo: missing ".o" in src/pcre/Makefile.
3838 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
3839 header line, restrict the check to what is listed in RFCs 2369 and 2929.
3840 Also, for "Auto-Submitted", treat anything other than "no" as
3841 non-personal, in accordance with RFC 3834. (Previously it treated
3842 anything starting "auto-" as non-personal.)
3844 TF/01 The control=submission/name=... option had a problem with syntax
3845 errors if the name included a slash character. The /name= option
3846 now slurps the rest of the string, so it can include any characters
3847 but it must come last in the list of options (after /sender_retain
3850 PH/05 Some modifications to the interface to the fake nameserver for the new
3858 TK/01 Added the "success_on_redirect" address verification option. See
3859 NewStuff for rationale and an example.
3861 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
3863 PH/02 Patch to exigrep to allow it to work on syslog lines.
3865 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
3866 fread() to skip over the body file's header line, because in Cygwin the
3867 header line is locked and is inaccessible.
3869 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
3870 co-exist for some time) to make it clear that it is the Exim ID that is
3871 referenced, not the Message-ID: header line.
3873 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
3874 string_format() function, because snprintf() does not exist on all
3877 PH/06 The use of forbid_filter_existstest now also locks out the use of the
3878 ${stat: expansion item.
3880 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
3881 protocol synchronization error", to keep the pedants happy.
3883 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
3884 well as for IRIX systems, when gcc is being used. See the host.c source
3887 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
3889 PH/10 Named domain lists were not working if used in a queue_smtp_domains
3892 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
3893 transport and to the smtp transport in LMTP mode.
3895 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
3897 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
3898 run a filter in a subprocess. This could lead to confusion in subsequent
3899 lookups in the parent process. There should also be a search_tidyup() at
3900 the end of the subprocess.
3902 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
3903 only if the host matched helo_try_verify_hosts, which caused the
3904 verification to occur when the EHLO/HELO command was issued. The ACL just
3905 tested the remembered result. Now, if a previous verification attempt has
3906 not happened, "verify = helo" does it there and then.
3908 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
3910 TK/03 Fix log output including CR from clamd.
3912 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
3913 exist provoked a memory error which could cause a segfault.
3915 PH/15 Installed PCRE 6.2
3917 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
3919 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
3920 of the problem. Specifically, suggested +O2 rather than +O1 for the
3923 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
3925 PH/20 If a delivery was routed to a non-standard port by means of an SRV
3926 record, the port was not correctly logged when the outgoing_port log
3927 selector was set (it logged the transort's default port).
3929 PH/21 Added support for host-specific ports to manualroute, queryprogram,
3930 fallback_hosts, and "hosts" in the smtp transport.
3932 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
3933 host errors such as "Connection refused".
3935 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
3936 authentication with radiusclient 0.4.9:
3938 - Error returned from rc_read_config was caught wrongly
3939 - Username/password not passed on to radius server due to wrong length.
3941 The presumption is that some radiusclient API changes for 4.51/PH/17
3942 were not taken care of correctly. The code is still untested by me (my
3943 Linux distribution still has 0.3.2 of radiusclient), but it was
3944 contributed by a Radius user.
3946 PH/24 When doing a callout, the value of $domain wasn't set correctly when
3947 expanding the "port" option of the smtp transport.
3949 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
3950 while reading a MIME header. Thanks to Tom Hughes for a patch.
3952 PH/24 Include config.h inside local_scan.h so that configuration settings are
3955 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
3956 that in an ACL for RCPT (for example), you can examine exactly what was
3959 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
3960 commands, but it was not correctly comparing the address with the actual
3961 client host address. Thus, it would show the EHLO address in Received:
3962 header lines when this was not necessary.
3964 PH/27 Added the % operator to ${eval:}.
3966 PH/28 Exim tries to create and chdir to its spool directory when it starts;
3967 it should be ignoring failures (because with -C, for example, it has lost
3968 privilege). It wasn't ignoring creation failures other than "already
3971 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
3974 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
3975 because an input line was too long, either on its own, or by virtue of
3976 too many continuations, the temporary file was not being removed, and the
3977 return code was incorrect.
3979 PH/31 Missing "BOOL" in function definition in filtertest.c.
3981 PH/32 Applied Sieve patches from the maintainer.
3983 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
3985 PH/33 Added "verify = not_blind".
3987 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
3988 Local/Makefile (with some defaults set). These are used in built scripts
3989 such as exicyclog, but they have never been used in the exim_install
3990 script (though there are many overriding facilities there). I have
3991 arranged that the exim_install script now takes note of these two
3994 PH/35 Installed configuration files for Dragonfly.
3996 PH/36 When a locally submitted message by a trusted user did not contain a
3997 From: header, and the sender address was obtained from -f or from an SMTP
3998 MAIL command, and the trusted user did not use -F to supply a sender
3999 name, $originator_name was incorrectly used when constructing a From:
4000 header. Furthermore, $originator_name was used for submission mode
4001 messages from external hosts without From: headers in a similar way,
4002 which is clearly wrong.
4004 PH/37 Added control=suppress_local_fixups.
4006 PH/38 When log_selector = +received_sender was set, and the addition of the
4007 sender made the log line's construction buffer exactly full, or one byte
4008 less than full, an overflow happened when the terminating "\n" was
4011 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
4012 when the result of a list match is failure because a DNS lookup failed.
4014 PH/40 RM_COMMAND is now used in the building process.
4016 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
4017 the "build-* directories that it finds.
4019 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
4020 address in a domain literal was a prefix of an interface address.
4022 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
4023 when verifying a sender address, unless rewrite_headers is false.
4025 PH/44 Wrote a long comment about why errors_to addresses are verified as
4026 recipients, not senders.
4028 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
4029 the ratelimit ACL was added.
4031 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
4033 PH/46 Added extra information about PostgreSQL errors to the error string.
4035 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
4036 avoiding the need to install special zones in a real server. This is
4037 backwards compatible; if it can't find the fake resolver, it drops back.
4038 Thus, both old and new test suites can be run.
4040 TF/02 Added util/ratelimit.pl
4042 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
4043 clock is set back in time.
4045 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
4046 Candler <B.Candler@pobox.com>.
4048 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
4049 OK for addresses that are the result of redirections.
4051 PH/48 A number of further additions for the benefit of the new test suite,
4052 including a fake gethostbyname() that interfaces to the fake DNS resolver
4055 TF/06 The fix for widen_domains has also been applied to qualify_single and
4056 search_parents which are the other dnslookup options that can cause
4059 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
4062 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
4064 TF/07 Exim produced the error message "an SRV record indicated no SMTP
4065 service" if it encountered an MX record with an empty target hostname.
4066 The message is now "an MX or SRV record indicated no SMTP service".
4068 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
4069 if the DNS of the sending site is misconfigured. This is quite a
4070 common situation. This change restores the behaviour of treating a
4071 helo verification defer as a failure.
4073 PH/51 If self=fail was set on a router, the bounce message did not include the
4074 actual error message.
4080 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
4082 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
4083 command itself ended in error, the underlying message about the transport
4084 filter timeout was being overwritten with the pipe command error. Now the
4085 underlying error message should be appended to the second error message.
4087 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
4089 PH/02 Reduce the amount of output that "make" produces by default. Full output
4090 can still be requested.
4092 PH/03 The warning log line about a condition test deferring for a "warn" verb
4093 was being output only once per connection, rather than after each
4094 occurrence (because it was using the same function as for successful
4095 "warn" verbs). This seems wrong, so I have changed it.
4097 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
4098 it should not have, which might have caused a crash in the right
4099 circumstances, but probably never did.
4101 PH/04 Installed a modified version of Tony Finch's patch to make submission
4102 mode fix the return path as well as the Sender: header line, and to
4103 add a /name= option so that you can make the user's friendly name appear
4106 TF/03 Added the control = fakedefer ACL modifier.
4108 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
4109 Mark Lowes for thorough testing.
4111 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
4113 TK/03 Merged latest SRS patch from Miles Wilton.
4115 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
4116 with the definition in sysexits.h (which is #included earlier).
4117 Fortunately, Exim does not actually use EX_OK. The code used to try to
4118 preserve the sysexits.h value, by assuming that macro definitions were
4119 scanned for macro replacements. I have been disabused of this notion,
4120 so now the code just undefines EX_OK before #including unistd.h.
4122 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
4123 in the smtp transport. When a block could not be written in a single
4124 write() function, the timeout was being re-applied to each part-write.
4125 This seems wrong - if the receiver was accepting one byte at a time it
4126 would take for ever. The timeout is now adjusted when this happens. It
4127 doesn't have to be particularly precise.
4129 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
4130 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
4132 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
4133 without a domain if the check to <postmaster@domain> fails.
4135 SC/01 Eximstats: added -xls and the ability to specify output files
4136 (patch written by Frank Heydlauf).
4138 SC/02 Eximstats: use FileHandles for outputting results.
4140 SC/03 Eximstats: allow any combination of xls, txt, and html output.
4142 SC/04 Eximstats: fixed display of large numbers with -nvr option
4144 SC/05 Eximstats: fixed merging of reports with empty tables.
4146 SC/06 Eximstats: added the -include_original_destination flag
4148 SC/07 Eximstats: removed tabs and trailing whitespace.
4150 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
4152 TK/06 MBOX spool code: Add real "From " MBOX separator line
4153 so the .eml file is really in mbox format (even though
4154 most programs do not really care). Patch from Alex Miller.
4156 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
4157 The latter is generated from $received_to and is only set if the
4158 message has one envelope recipient. SA can use these headers,
4159 obviously out-of-the-box. Patch from Alex Miller.
4161 PH/08 The ${def test on a variable was returning false if the variable's
4162 value was "0", contrary to what the specification has always said!
4163 The result should be true unless the variable is empty.
4165 PH/09 The syntax error of a character other than { following "${if
4166 def:variable_name" (after optional whitespace) was not being diagnosed.
4167 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
4168 accidental colon was present, for example, could give incorrect results.
4170 PH/10 Tidied the code in a number of places where the st_size field of a stat()
4171 result is used (not including appendfile, where other changes are about
4174 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
4175 This involved changing a lot of size variables from int to off_t. It
4176 should work with maildirs and everything.
4178 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
4179 spamd dying while we are connected to it.
4181 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
4184 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
4185 function for BDB 4.3.
4187 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
4189 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
4190 the sources for PCRE have been re-arranged and more files are now
4193 PH/15 The code I had for printing potentially long long variables in PH/11
4194 above was not the best (it lost precision). The length of off_t variables
4195 is now inspected at build time, and an appropriate printing format (%ld
4196 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
4197 to be "long long int" or "long int". This is needed for the internal
4198 formatting function string_vformat().
4200 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
4201 the configuration file to be ":syslog", then the script "guesses" where
4202 the logs files are, rather than using the compiled in default. In our
4203 case the guess is not the same as the compiled default, so the script
4204 suddenly stopped working when I started to use syslog. The patch checks
4205 to see if log_file_path is "". If so, it attempts to read it from exim
4206 with no configuration file to get the compiled in version, before it
4207 falls back to the previous guessing code."
4209 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
4210 implementing BATV in an Exim configuration. See NewStuff for the gory
4213 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
4214 Makefile that are specific to HP-UX.
4216 PH/18 If the "use_postmaster" option was set for a recipient callout together
4217 with the "random" option, the postmaster address was used as the MAIL
4218 FROM address for the random test, but not for the subsequent recipient
4219 test. It is now used for both.
4221 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
4222 patch removes a few documentation additions to RFC 3028, because the
4223 latest draft now contains them. It adds the new en;ascii-case comparator
4224 and a new error check for 8bit text in MIME parts. Comparator and
4225 require names are now matched exactly. I enabled the subaddress
4226 extension, but it is not well tested yet (read: it works for me)."
4228 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
4229 rework some of the code of TK/09 above to avoid the hardwired use of
4230 "%lld" and "long long". Replaced the call to snprintf() with a call to
4233 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
4234 records point to non-existent hosts", "retry timeout exceeded", and
4235 "retry time not reached for any host after a long failure period".
4237 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
4238 experimental DomainKeys support:
4240 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
4241 (2) On an error such as an illegally used "control", the wrong name for
4242 the control was given.
4244 These problems did NOT occur unless DomainKeys support was compiled.
4246 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
4248 PH/24 Added ${if match_ip condition.
4250 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
4251 databases so that it will be absolutely obvious if a crash occurs in the
4252 DB library. This is a regular occurrence (often caused by mis-matched
4255 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
4256 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
4257 picked up on a user's system that detects such things. There doesn't seem
4258 to be a gcc warning option for this - only an attribute that has to be
4259 put on the function's prototype. It seems that in Fedora Core 4 they have
4260 set this on a number of new functions. No doubt there will be more in due
4263 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
4264 specify a transport. However, if an address that was verified by such a
4265 router was the subject of a callout, Exim crashed because it tried to
4266 read the rcpt_include_affixes from the non-existent transport. Now it
4267 just assumes that the setting of that option is false. This bug was
4268 introduced by 4.51/PH/31.
4270 PH/28 Changed -d+all to exclude +memory, because that information is very
4271 rarely of interest, but it makes the output a lot bigger. People tend to
4272 do -d+all out of habit.
4274 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
4275 code in os-type was giving problems when libc.so lives in lib64, like on
4278 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
4279 aren't the modern standard, and it seems that some systems' include files
4280 don't always have them. Exim was already checking for some of the newer
4281 ones like T_AAAA, and defining it itself. I've added checks for all the
4282 record types that Exim uses.
4284 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
4285 not automatically generating a new one, as it is supposed to. This
4286 prevented TLS from working. If the file did exist, but contained invalid
4287 data, a new version was generated, as expected. It was only the case of a
4288 non-existent file that was broken.
4290 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
4291 with a change in libdomainkeys > 0.64.
4293 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
4294 from DNS. If the selector record carries the flag, it now has
4295 precedence over the domain-wide flag.
4297 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
4299 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
4300 the use of a port name that isn't defined in /etc/services) occurred, the
4301 message was deferred as in a normal delivery, and thus remained on the
4302 spool, instead of being failed because of the mua_wrapper setting. This
4303 is now fixed, and I tidied up some of the mua_wrapper messages at the
4306 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
4307 the messages in a hash of arrays rather than using individual hashes.
4308 This is a bit cleaner and results in dramatic memory savings, albeit
4309 at a slight CPU cost.
4311 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
4312 as requested by Marc Sherman.
4314 SC/10 Eximstats: added histograms for user specified patterns as requested
4317 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
4319 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
4320 fopen() in the content-scanning modules that did not already have it.
4326 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
4327 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
4329 TK/02 Fix ACL "control" statement not being available in MIME ACL.
4331 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
4333 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
4334 to test Sieve filters that use "vacation".
4336 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
4337 that changes the way the GnuTLS parameters are stored in the cache file.
4338 The new format can be generated externally. For backward compatibility,
4339 if the data in the cache doesn't make sense, Exim assumes it has read an
4340 old-format file, and it generates new data and writes a new file. This
4341 means that you can't go back to an older release without removing the
4344 PH/03 A redirect router that has both "unseen" and "one_time" set does not
4345 work if there are any delivery delays because "one_time" forces the
4346 parent to be marked "delivered", so its unseen clone is never tried
4347 again. For this reason, Exim now forbids the simultaneous setting of
4350 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
4351 redirected to themselves ("homonym" addresses). Read the long ChangeLog
4352 entry if you want to know the details. The fix, however, neglected to
4353 consider the case when local delivery batching is involved. The test for
4354 "previously delivered" was not happening when checking to see if an
4355 address could be batched with a previous (undelivered) one; under
4356 certain circumstances this could lead to multiple deliveries to the same
4359 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
4360 in its include files, and this causes problems building Exim.
4362 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
4363 header_syntax) but Exim was just ignoring anything given after a slash.
4364 In particular, this caused confusion with an attempt to use "verify =
4365 reverse_host_lookup/defer_ok". An error is now given when options are
4366 supplied for verify items that do not have them. (Maybe reverse_host_
4367 lookup should have a defer_ok option, but that's a different point.)
4369 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
4370 defined by RFC 821) to 2048, because there were problems with some AUTH
4371 commands, and RFC 1869 says the size should be increased for extended
4372 SMTP commands that take arguments.
4374 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
4377 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
4378 "unknown" error; now it says that the functionality isn't in the binary.
4380 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
4381 an address' error message when a string expansion fails (syntax or
4382 whatever). Otherwise the password may appear in the log. Following change
4383 PH/42 below, there is no longer a chance of it appearing in a bounce
4386 PH/11 Installed exipick version 20050225.0 from John Jetmore.
4388 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
4389 of its addresses was ever tried. (Bugzilla bug #2.)
4391 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
4392 the result incorrectly in the debug output. (It correctly added a newline
4393 to what was transported.)
4395 TF/01 Added $received_time.
4397 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
4398 commented out examples of how to interface to a virus scanner and to
4399 SpamAssassin. Also added commented examples of av_scanner and
4400 spamd_address settings.
4402 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
4403 and controls are allowed in which ACLs. There were a couple of minor
4404 errors. Some of the entries in the conditions table (which is a table of
4405 where they are NOT allowed) were getting very unwieldy; rewrote them as a
4406 negation of where the condition IS allowed.
4408 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
4410 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
4411 header file does not have a version number, so I've had to invent a new
4412 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
4413 API. The code is untested by me (my Linux distribution still has 0.3.2 of
4414 radiusclient), but it was contributed by a Radius user.
4416 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
4417 files or directories, for OpenSSL.
4419 PH/19 When an Exim process that is running as root has to create an Exim log
4420 file, it does so in a subprocess that runs as exim:exim so as to get the
4421 ownership right at creation (otherwise, other Exim processes might see
4422 the file with the wrong ownership). There was no test for failure of this
4423 fork() call, which would lead to the process getting stuck as it waited
4424 for a non-existent subprocess. Forks do occasionally fail when resources
4425 run out. I reviewed all the other calls to fork(); they all seem to check
4428 PH/20 When checking for unexpected SMTP input at connect time (before writing
4429 the banner), Exim was not dealing correctly with a non-positive return
4430 from the read() function. If the client had disconnected by this time,
4431 the result was a log entry for a synchronization error with an empty
4432 string after "input=" when read() returned zero. If read() returned -1
4433 (an event I could not check), uninitialized data bytes were printed.
4434 There were reports of junk text (parts of files, etc) appearing after
4437 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
4439 PH/22 Added support for macro redefinition, and (re)definition in between
4440 driver and ACL definitions.
4442 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
4443 forgetting to use the resulting value; it was using the unexpanded value.
4445 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
4446 hadn't been configured. The fix is from Juergen Kreileder, who
4447 understands it better than I do:
4449 "Here's what I see happening with three configured cyrus_sasl
4450 authenticators configured (plain, login, cram-md5):
4452 On startup auth_cyrus_sasl_init() gets called for each of these.
4453 This means three calls to sasl_listmech() without a specified mech_list.
4454 => SASL tests which mechs of all available mechs actually work
4455 => three warnings about OTP not working
4456 => the returned list contains: plain, login, cram-md5, digest-md5, ...
4458 With the patch, sasl_listmech() also gets called three times. But now
4459 SASL's mech_list option is set to the server_mech specified in the the
4460 authenticator. Or in other words, the answer from sasl_listmech()
4461 gets limited to just the mech you're testing for (which is different
4463 => the return list contains just 'plain' or 'login', 'cram-md5' or
4464 nothing depending on the value of ob->server_mech.
4466 I've just tested the patch: Authentication still works fine,
4467 unavailable mechs specified in the exim configuration are still
4468 caught, and the auth.log warnings about OTP are gone."
4470 PH/25 When debugging is enabled, the contents of the command line are added
4471 to the debugging output, even when log_selector=+arguments is not
4474 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
4475 answer is "GNU", and only if the return is "GNU/something" is the answer
4478 PH/27 $acl_verify_message is now set immediately after the failure of a
4479 verification in an ACL, and so is available in subsequent modifiers. In
4480 particular, the message can be preserved by coding like this:
4482 warn !verify = sender
4483 set acl_m0 = $acl_verify_message
4485 Previously, $acl_verify_message was set only while expanding "message"
4486 and "log_message" when a very denied access.
4488 PH/28 Modified OS/os.c-Linux with
4490 -#ifndef OS_LOAD_AVERAGE
4491 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
4493 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
4494 nomenclature these days.)
4496 PH/29 Installed patch from the Sieve maintainer that adds the options
4497 sieve_useraddress and sieve_subaddress to the redirect router.
4499 PH/30 In these circumstances:
4500 . Two addresses routed to the same list of hosts;
4501 . First host does not offer TLS;
4502 . First host accepts first address;
4503 . First host gives temporary error to second address;
4504 . Second host offers TLS and a TLS session is established;
4505 . Second host accepts second address.
4506 Exim incorrectly logged both deliveries with the TLS parameters (cipher
4507 and peerdn, if requested) that were in fact used only for the second
4510 PH/31 When doing a callout as part of verifying an address, Exim was not paying
4511 attention to any local part prefix or suffix that was matched by the
4512 router that accepted the address. It now behaves in the same way as it
4513 does for delivery: the affixes are removed from the local part unless
4514 rcpt_include_affixes is set on the transport.
4516 PH/32 Add the sender address, as F=<...>, to the log line when logging a
4517 timeout during the DATA phase of an incoming message.
4519 PH/33 Sieve envelope tests were broken for match types other than :is. I have
4520 applied a patch sanctioned by the Sieve maintainer.
4522 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
4523 the uid or gid is negative. A case of a negative gid caused this to be
4524 noticed. The fix allows for either to be negative.
4526 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
4527 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
4530 PH/36 The change PH/12 above was broken. Fixed it.
4532 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
4533 the grounds that routing the same address twice would always produce the
4534 same answer. This might have been true once, but it is certainly no
4535 longer true now. Routing a child address may depend on the previous
4536 routing that produced that child. Some complicated redirection strategies
4537 went wrong when messages had multiple recipients, and made Exim's
4538 behaviour dependent on the order in which the addresses were given.
4540 I have moved the duplicate checking until after the routing is complete.
4541 Exim scans the addresses that are assigned to local and remote
4542 transports, and removes any duplicates. This means that more work will be
4543 done, as duplicates will always all be routed, but duplicates are
4544 presumably rare, so I don't expect this is of any significance.
4546 For deliveries to pipes, files, and autoreplies, the duplicate checking
4547 still happens during the routing process, since they are not going to be
4550 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
4551 It corrects a timeout issue with spamd. This is Ian's comment: "The
4552 background is that sometimes spamd either never reads data from a
4553 connection it has accepted, or it never writes response data. The exiscan
4554 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
4555 blindly assumes that writes won't block so it may never time out."
4557 PH/39 Allow G after quota size as well as K and M.
4559 PH/40 The value set for $authenticated_id in an authenticator may not contain
4560 binary zeroes or newlines because the value is written to log lines and
4561 to spool files. There was no check on this. Now the value is run through
4562 the string_printing() function so that such characters are converted to
4563 printable escape sequences.
4565 PH/41 $message_linecount is a new variable that contains the total number of
4566 lines in the message. Compare $body_linecount, which is the count for the
4569 PH/42 Exim no longer gives details of delivery errors for specific addresses in
4570 bounce and delay warning messages, except in certain special cases, which
4573 (a) An SMTP error message from a remote host;
4574 (b) A message specified in a :fail: redirection;
4575 (c) A message specified in a "fail" command in a system filter;
4576 (d) A message specified in a FAIL return from the queryprogram router;
4577 (e) A message specified by the cannot_route_message router option.
4579 In these cases only, Exim does include the error details in bounce and
4580 warning messages. There are also a few cases where bland messages such
4581 as "unrouteable address" or "local delivery error" are given.
4583 PH/43 $value is now also set for the "else" part of a ${run expansion.
4585 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
4586 being worked on, but at least Exim now implements the latest version to
4589 PH/45 In a pipe transport, although a timeout while waiting for the pipe
4590 process to complete was treated as a delivery failure, a timeout while
4591 writing the message to the pipe was logged, but erroneously treated as a
4592 successful delivery. Such timeouts include transport filter timeouts. For
4593 consistency with the overall process timeout, these timeouts are now
4594 treated as errors, giving rise to delivery failures by default. However,
4595 there is now a new Boolean option for the pipe transport called
4596 timeout_defer, which, if set TRUE, converts the failures into defers for
4597 both kinds of timeout. A transport filter timeout is now identified in
4600 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
4601 systems where "make" and "gmake" are different, calling "gmake" at top
4602 level broke things. I've arranged for the value of $(MAKE) to be passed
4603 from the Makefile to this script so that it can call the same version of
4607 A note about Exim versions 4.44 and 4.50
4608 ----------------------------------------
4610 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
4611 changes of various kinds. As a consequence, a big documentation update was
4612 needed. This delayed the release for rather longer than seemed good, especially
4613 in the light of a couple of (minor) security issues. Therefore, the changes
4614 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
4615 release. So 4.44 and 4.50 are in effect two different branches that both start
4618 I have left the 4.50 change log unchanged; it contains all the changes since
4619 4.43. The change log for 4.44 is below; many of its items are identical to
4620 those for 4.50. This seems to be the most sensible way to preserve the
4621 historical information.
4627 1. Minor wording change to the doc/README.SIEVE file.
4629 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
4630 computation of the current number of files was incorrect.
4632 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4633 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4636 4. Give more explanation in the error message when the command for a transport
4637 filter fails to execute.
4639 5. There are several places where Exim runs a non-Exim command in a
4640 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4641 was being done only for the command run by the queryprogram router. It is
4642 now done for all such subprocesses. The other cases are: ${run, transport
4643 filters, and the commands run by the lmtp and pipe transports.
4645 6. Added CONFIGURE_GROUP build-time option.
4647 7. Some older OS have a limit of 256 on the maximum number of file
4648 descriptors. Exim was using setrlimit() to set 1000 as a large value
4649 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4650 systems. I've change it so that if it can't get 1000, it tries for 256.
4652 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4653 was an oversight, and furthermore, ever since the addition of extra
4654 controls (e.g. 4.43/32), the checks on when to allow different forms of
4655 "control" were broken. There should now be diagnostics for all cases when a
4656 control that does not make sense is encountered.
4658 9. Added the /retain_sender option to "control=submission".
4660 10. $recipients is now available in the predata ACL (oversight).
4662 11. Tidy the search cache before the fork to do a delivery from a message
4663 received from the command line. Otherwise the child will trigger a lookup
4664 failure and thereby defer the delivery if it tries to use (for example) a
4665 cached ldap connection that the parent has called unbind on.
4667 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4668 of $address_data from the recipient verification was clobbered by the
4669 sender verification.
4671 13. The value of address_data from a sender verification is now available in
4672 $sender_address_data in subsequent conditions in the ACL statement.
4674 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
4676 15. Added a new option "connect=<time>" to callout options, to set a different
4679 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4680 was its contents. (It was OK if the option was not defined at all.)
4682 17. A "Completed" log line is now written for messages that are removed from
4683 the spool by the -Mrm option.
4685 18. New variables $sender_verify_failure and $recipient_verify_failure contain
4686 information about exactly what failed.
4688 19. Added -dd to debug only the daemon process.
4690 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
4691 handles timeouts, both on the server side and network timeouts. Renamed the
4692 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
4694 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
4695 It is now set to "smtps".
4697 22. $host_address is now set to the target address during the checking of
4698 ignore_target_hosts.
4700 23. When checking ignore_target_hosts for an ipliteral router, no host name was
4701 being passed; this would have caused $sender_host_name to have been used if
4702 matching the list had actually called for a host name (not very likely,
4703 since this list is usually IP addresses). A host name is now passed as
4706 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4707 code that specifies a non-restarting handler (typically sigaction() in
4708 modern systems) in an attempt to fix a rare and obscure crash bug.
4710 25. Narrowed the window for a race in the daemon that could cause it to ignore
4711 SIGCHLD signals. This is not a major problem, because they are used only to
4712 wake it up if nothing else does.
4714 26. A malformed maildirsize file could cause Exim to calculate negative values
4715 for the mailbox size or file count. Odd effects could occur as a result.
4716 The maildirsize information is now recalculated if the size or filecount
4719 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4720 support for a long time. Removed HAVE_SYS_VFS_H.
4722 28. Installed the latest version of exipick from John Jetmore.
4724 29. In an address list, if the pattern was not a regular expression, an empty
4725 subject address (from a bounce message) matched only if the pattern was an
4726 empty string. Non-empty patterns were not even tested. This was the wrong
4727 because it is perfectly reasonable to use an empty address as part of a
4728 database query. An empty address is now tested by patterns that are
4729 lookups. However, all the other forms of pattern expect the subject to
4730 contain a local part and a domain, and therefore, for them, an empty
4731 address still always fails if the pattern is not itself empty.
4733 30. Exim went into a mad DNS loop when attempting to do a callout where the
4734 host was specified on an smtp transport, and looking it up yielded more
4735 than one IP address.
4737 31. Re-factored the code for checking spool and log partition space into a
4738 function that finds that data and another that does the check. The former
4739 is then used to implement four new variables: $spool_space, $log_space,
4740 $spool_inodes, and $log_inodes.
4742 32. The RFC2047 encoding function was originally intended for short strings
4743 such as real names; it was not keeping to the 75-character limit for
4744 encoded words that the RFC imposes. It now respects the limit, and
4745 generates multiple encoded words if necessary. To be on the safe side, I
4746 have increased the buffer size for the ${rfc2047: expansion operator from
4749 33. It is now permitted to omit both strings after an "if" condition; if the
4750 condition is true, the result is "true". As before, when the second string
4751 is omitted, a false condition yields an empty string. This makes it less
4752 cumbersome to write custom ACL and router conditions.
4754 34. Failure to deliver a bounce message always caused it to be frozen, even if
4755 there was an errors_to setting on the router. The errors_to setting is now
4758 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4759 canonical form (fully expanded) before being placed in
4760 $sender_host_address.
4762 36. The table in the code that translates DNS record types into text (T_A to
4763 "A" for instance) was missing entries for NS and CNAME. It is just possible
4764 that this could have caused confusion if both these types were looked up
4765 for the same domain, because the text type is used as part of Exim's
4766 per-process caching. But the chance of anyone hitting this buglet seems
4769 37. The dnsdb lookup has been extended in a number of ways.
4771 (1) There is a new type, "zns", which walks up the domain tree until it
4772 finds some nameserver records. It should be used with care.
4774 (2) There is a new type, "mxh", which is like "mx" except that it returns
4775 just the host names, not the priorities.
4777 (3) It is now possible to give a list of domains (or IP addresses) to be
4778 looked up. The behaviour when one of the lookups defers can be
4779 controlled by a keyword.
4781 (4) It is now possible to specify the separator character for use when
4782 multiple records are returned.
4784 38. The dnslists ACL condition has been extended: it is now possible to supply
4785 a list of IP addresses and/or domains to be looked up in a particular DNS
4788 39. Added log_selector=+queue_time_overall.
4790 40. When running the queue in the test harness, wait just a tad after forking a
4791 delivery process, to get repeatability of debugging output.
4793 41. Include certificate and key file names in error message when GnuTLS fails
4794 to set them up, because the GnuTLS error message doesn't include the name
4795 of the failing file when there is a problem reading it.
4797 42. Allow both -bf and -bF in the same test run.
4799 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
4801 44. The "Exiscan patch" is now merged into the mainline Exim source.
4803 45. Sometimes the final signoff response after QUIT could fail to get
4804 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4805 < 0 before doing a fflush(). This bug looks as though it goes back to the
4806 introduction of TLS in release 3.20, but "sometimes" must have been rare
4807 because the tests only now provoked it.
4809 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
4810 (this can affect the format of dates).
4812 47. exim_tidydb, when checking for the continued existence of a message for
4813 which it has found a message-specific retry record, was not finding
4814 messages that were in split spool directories. Consequently, it was
4815 deleting retry records that should have stayed in existence.
4817 48. Steve fixed some bugs in eximstats.
4819 49. The SPA authentication driver was not abandoning authentication and moving
4820 on to the next authenticator when an expansion was forced to fail,
4821 contradicting the general specification for all authenticators. Instead it
4822 was generating a temporary error. It now behaves as specified.
4824 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
4825 (the order specifies the preference for clients). The order is now AES256,
4826 AES128, 3DES, ARCFOUR128.
4828 51. Small patch to Sieve code - explicitly set From: when generating an
4831 52. Exim crashed if a remote delivery caused a very long error message to be
4832 recorded - for instance if somebody sent an entire SpamAssassin report back
4833 as a large number of 550 error lines. This bug was coincidentally fixed by
4834 increasing the size of one of Exim's internal buffers (big_buffer) that
4835 happened as part of the Exiscan merge. However, to be on the safe side, I
4836 have made the code more robust (and fixed the comments that describe what
4839 53. Now that there can be additional text after "Completed" in log lines (if
4840 the queue_time_overall log selector is set), a one-byte patch to exigrep
4841 was needed to allow it to recognize "Completed" as not the last thing in
4844 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
4845 patch that reportedly fixes this has been added. I am not expert enough to
4846 create a test for it. This is what the patch creator wrote:
4848 "I found a little strange behaviour of ldap code when working with
4849 Windows 2003 AD Domain, where users was placed in more than one
4850 Organization Units. When I tried to give exim partial DN, the exit code
4851 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
4852 But simultaneously result of request was absolutely normal ldap result,
4853 so I produce this patch..."
4855 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
4856 REFERENCE, so I have modified the code to exclude the patch when that macro
4859 55. Some experimental protocols are using DNS PTR records for new purposes. The
4860 keys for these records are domain names, not reversed IP addresses. The
4861 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4862 leaves it alone. Component reversal etc. now happens only for IP addresses.
4865 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
4867 57. Double the size of the debug message buffer (to 2048) so that more of very
4868 long debug lines gets shown.
4870 58. The exicyclog utility now does better if the number of log files to keep
4871 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
4873 59. Two changes related to the smtp_active_hostname option:
4875 (1) $smtp_active_hostname is now available as a variable.
4876 (2) The default for smtp_banner uses $smtp_active_hostname instead
4877 of $primary_hostname.
4879 60. The host_aton() function is supposed to be passed a string that is known
4880 to be a valid IP address. However, in the case of IPv6 addresses, it was
4881 not checking this. This is a hostage to fortune. Exim now panics and dies
4882 if the condition is not met. A case was found where this could be provoked
4883 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4884 components; fortuitously, this particular loophole had already been fixed
4885 by change 4.50/55 above.
4887 If there are any other similar loopholes, the new check in host_aton()
4888 itself should stop them being exploited. The report I received stated that
4889 data on the command line could provoke the exploit when Exim was running as
4890 exim, but did not say which command line option was involved. All I could
4891 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4892 running as the user.
4895 61. There was a buffer overflow vulnerability in the SPA authentication code
4896 (which came originally from the Samba project). I have added a test to the
4897 spa_base64_to_bits() function which I hope fixes it.
4900 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
4901 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
4903 63. The daemon start-up calls getloadavg() while still root for those OS that
4904 need the first call to be done as root, but it missed one case: when
4905 deliver_queue_load_max is set with deliver_drop_privilege. This is
4906 necessary for the benefit of the queue runner, because there is no re-exec
4907 when deliver_drop_privilege is set.
4909 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
4910 This has been fixed.
4912 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4913 was in use, was not putting the data itself into the right store pool;
4914 consequently, it could be overwritten for a subsequent message in the same
4915 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4918 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
4920 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
4921 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
4922 the function were treating the return as a boolean value, which happened to
4923 work because 0=false and not-0=true, but is not correct code.
4925 68. The host_aton() function was not handling scoped IPv6 addresses (those
4926 with, for example, "%eth0" on the end) correctly.
4928 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
4929 negated items (that is, ~something) in unsigned ints. Some compilers
4930 apparently mutter when there is no cast.
4932 70. If an address verification called from an ACL failed, and did not produce a
4933 user-specific message (i.e. there was only a "system" message), nothing was
4934 put in $acl_verify_message. In this situation, it now puts the system
4937 71. Change 4.23/11 added synchronization checking at the start of an SMTP
4938 session; change 4.31/43 added the unwanted input to the log line - except
4939 that it did not do this in the start of session case. It now does.
4941 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
4942 This is wrong and can cause the other end to generate a synchronization
4943 error if it is another Exim or anything else that does the synchronization
4944 check. A QUIT command is no longer sent after a timeout.
4946 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
4947 during host lookups.
4949 74. The defer_ok option of callout verification was not working if it was used
4950 when verifying addresses in header lines, that is, for this case:
4952 verify = header_sender/callout=defer_ok
4954 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
4955 those file descriptors could be used for SMTP connections. If anything
4956 wrote to stderr (the example that came up was "warn" in embedded Perl), it
4957 could be sent to the SMTP client, causing chaos. The daemon now opens
4958 stdin, stdout, and stderr to /dev/null when it puts itself into the
4961 76. Arrange for output from Perl's "warn" command to be written to Exim's main
4962 log by default. The user can override this with suitable Perl magic.
4964 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
4965 the log message when discard triggers, was not working for the DATA ACL or
4966 for the non-SMTP ACL.
4968 78. Error message wording change in sieve.c.
4970 79. If smtp_accept_max_per_host was set, the number of connections could be
4971 restricted to fewer than expected, because the daemon was trying to set up
4972 a new connection before checking whether the processes handling previous
4973 connections had finished. The check for completed processes is now done
4974 earlier. On busy systems, this bug wouldn't be noticed because something
4975 else would have woken the daemon, and it would have reaped the completed
4978 80. If a message was submitted locally by a user whose login name contained one
4979 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
4980 It caused a spool format error. I have fixed the spool reading code. A
4981 related problem was that the "from" clause in the Received: line became
4982 illegal because of the space(s). It is now covered by ${quote_local_part.
4984 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
4987 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
4989 83. Patch from Sieve maintainer to fix unterminated string problem in
4990 "vacation" handling.
4992 84. Some minor changes to the Linux configuration files to help with other
4993 OS variants using glibc.
4995 85. One more patch for Sieve to update vacation handling to latest spec.
4998 ----------------------------------------------------
4999 See the note above about the 4.44 and 4.50 releases.
5000 ----------------------------------------------------
5006 1. Change 4.43/35 introduced a bug that caused file counts to be
5007 incorrectly computed when quota_filecount was set in an appendfile
5010 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
5011 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
5014 3. Give more explanation in the error message when the command for a transport
5015 filter fails to execute.
5017 4. There are several places where Exim runs a non-Exim command in a
5018 subprocess. The SIGUSR1 signal should be disabled for these processes. This
5019 was being done only for the command run by the queryprogram router. It is
5020 now done for all such subprocesses. The other cases are: ${run, transport
5021 filters, and the commands run by the lmtp and pipe transports.
5023 5. Some older OS have a limit of 256 on the maximum number of file
5024 descriptors. Exim was using setrlimit() to set 1000 as a large value
5025 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
5026 systems. I've change it so that if it can't get 1000, it tries for 256.
5028 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
5029 was an oversight, and furthermore, ever since the addition of extra
5030 controls (e.g. 4.43/32), the checks on when to allow different forms of
5031 "control" were broken. There should now be diagnostics for all cases when a
5032 control that does not make sense is encountered.
5034 7. $recipients is now available in the predata ACL (oversight).
5036 8. Tidy the search cache before the fork to do a delivery from a message
5037 received from the command line. Otherwise the child will trigger a lookup
5038 failure and thereby defer the delivery if it tries to use (for example) a
5039 cached ldap connection that the parent has called unbind on.
5041 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
5042 of $address_data from the recipient verification was clobbered by the
5043 sender verification.
5045 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
5046 was its contents. (It was OK if the option was not defined at all.)
5048 11. A "Completed" log line is now written for messages that are removed from
5049 the spool by the -Mrm option.
5051 12. $host_address is now set to the target address during the checking of
5052 ignore_target_hosts.
5054 13. When checking ignore_target_hosts for an ipliteral router, no host name was
5055 being passed; this would have caused $sender_host_name to have been used if
5056 matching the list had actually called for a host name (not very likely,
5057 since this list is usually IP addresses). A host name is now passed as
5060 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
5061 code that specifies a non-restarting handler (typically sigaction() in
5062 modern systems) in an attempt to fix a rare and obscure crash bug.
5064 15. Narrowed the window for a race in the daemon that could cause it to ignore
5065 SIGCHLD signals. This is not a major problem, because they are used only to
5066 wake it up if nothing else does.
5068 16. A malformed maildirsize file could cause Exim to calculate negative values
5069 for the mailbox size or file count. Odd effects could occur as a result.
5070 The maildirsize information is now recalculated if the size or filecount
5073 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
5074 support for a long time. Removed HAVE_SYS_VFS_H.
5076 18. Updated exipick to current release from John Jetmore.
5078 19. Allow an empty sender to be matched against a lookup in an address list.
5079 Previously the only cases considered were a regular expression, or an
5082 20. Exim went into a mad DNS lookup loop when doing a callout where the
5083 host was specified on the transport, if the DNS lookup yielded more than
5086 21. The RFC2047 encoding function was originally intended for short strings
5087 such as real names; it was not keeping to the 75-character limit for
5088 encoded words that the RFC imposes. It now respects the limit, and
5089 generates multiple encoded words if necessary. To be on the safe side, I
5090 have increased the buffer size for the ${rfc2047: expansion operator from
5093 22. Failure to deliver a bounce message always caused it to be frozen, even if
5094 there was an errors_to setting on the router. The errors_to setting is now
5097 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
5098 canonical form (fully expanded) before being placed in
5099 $sender_host_address.
5101 24. Updated eximstats to version 1.33
5103 25. Include certificate and key file names in error message when GnuTLS fails
5104 to set them up, because the GnuTLS error message doesn't include the name
5105 of the failing file when there is a problem reading it.
5107 26. Expand error message when OpenSSL has problems setting up cert/key files.
5110 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
5111 (this can affect the format of dates).
5113 28. exim_tidydb, when checking for the continued existence of a message for
5114 which it has found a message-specific retry record, was not finding
5115 messages that were in split spool directories. Consequently, it was
5116 deleting retry records that should have stayed in existence.
5118 29. eximstats updated to version 1.35
5119 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
5120 1.35 - bugfix such that pie charts by volume are generated correctly
5122 30. The SPA authentication driver was not abandoning authentication and moving
5123 on to the next authenticator when an expansion was forced to fail,
5124 contradicting the general specification for all authenticators. Instead it
5125 was generating a temporary error. It now behaves as specified.
5127 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
5128 (the order specifies the preference for clients). The order is now AES256,
5129 AES128, 3DES, ARCFOUR128.
5131 31. Small patch to Sieve code - explicitly set From: when generating an
5134 32. Exim crashed if a remote delivery caused a very long error message to be
5135 recorded - for instance if somebody sent an entire SpamAssassin report back
5136 as a large number of 550 error lines. This bug was coincidentally fixed by
5137 increasing the size of one of Exim's internal buffers (big_buffer) that
5138 happened as part of the Exiscan merge. However, to be on the safe side, I
5139 have made the code more robust (and fixed the comments that describe what
5142 33. Some experimental protocols are using DNS PTR records for new purposes. The
5143 keys for these records are domain names, not reversed IP addresses. The
5144 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
5145 leaves it alone. Component reversal etc. now happens only for IP addresses.
5148 34. The host_aton() function is supposed to be passed a string that is known
5149 to be a valid IP address. However, in the case of IPv6 addresses, it was
5150 not checking this. This is a hostage to fortune. Exim now panics and dies
5151 if the condition is not met. A case was found where this could be provoked
5152 from a dnsdb PTR lookup with an IPv6 address that had more than 8
5153 components; fortuitously, this particular loophole had already been fixed
5154 by change 4.50/55 or 4.44/33 above.
5156 If there are any other similar loopholes, the new check in host_aton()
5157 itself should stop them being exploited. The report I received stated that
5158 data on the command line could provoke the exploit when Exim was running as
5159 exim, but did not say which command line option was involved. All I could
5160 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
5161 running as the user.
5164 35. There was a buffer overflow vulnerability in the SPA authentication code
5165 (which came originally from the Samba project). I have added a test to the
5166 spa_base64_to_bits() function which I hope fixes it.
5169 36. The daemon start-up calls getloadavg() while still root for those OS that
5170 need the first call to be done as root, but it missed one case: when
5171 deliver_queue_load_max is set with deliver_drop_privilege. This is
5172 necessary for the benefit of the queue runner, because there is no re-exec
5173 when deliver_drop_privilege is set.
5175 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
5176 was in use, was not putting the data itself into the right store pool;
5177 consequently, it could be overwritten for a subsequent message in the same
5178 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
5181 38. Sometimes the final signoff response after QUIT could fail to get
5182 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
5183 < 0 before doing a fflush(). This bug looks as though it goes back to the
5184 introduction of TLS in release 3.20, but "sometimes" must have been rare
5185 because the tests only now provoked it.
5191 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
5192 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
5193 Now it returns an integer. A number of calls were still expecting a T/F
5194 return. Fortuitously, in all cases, the tests worked in OK situations,
5195 which is the norm. However, things would have gone wrong on any write
5196 failures on the smtp file descriptor. This function is used when sending
5197 messages over SMTP and also when doing verify callouts.
5199 2. When Exim is called to do synchronous delivery of a locally submitted
5200 message (the -odf or -odi options), it no longer closes stderr before doing
5203 3. Implemented the mua_wrapper option.
5205 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
5207 5. Implemented the functions header_remove(), header_testname(),
5208 header_add_at_position(), and receive_remove_recipient(), and exported them
5211 6. If an ACL "warn" statement specified the addition of headers, Exim already
5212 inserted X-ACL-Warn: at the start if there was no header name. However, it
5213 was not making this test for the second and subsequent header lines if
5214 there were newlines in the string. This meant that an invalid header could
5215 be inserted if Exim was badly configured.
5217 7. Allow an ACL "warn" statement to add header lines at the start or after all
5218 the Received: headers, as well as at the end.
5220 8. Added the rcpt_4xx retry error code.
5222 9. Added postmaster_mailfrom=xxx to callout verification option.
5224 10. Added mailfrom=xxxx to the callout verification option, for verify=
5227 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
5228 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
5230 12. Inserted some casts to stop certain compilers warning when using pointer
5231 differences as field lengths or precisions in printf-type calls (mostly
5232 affecting debugging statements).
5234 13. Added optional readline() support for -be (dynamically loaded).
5236 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
5237 same clock tick as a message's arrival, so that its received time was the
5238 same as the "first fail" time on the retry record, and that message
5239 remained on the queue past the ultimate address timeout, every queue runner
5240 would try a delivery (because it was past the ultimate address timeout) but
5241 after another failure, the ultimate address timeout, which should have then
5242 bounced the address, did not kick in. This was a "< instead of <=" error;
5243 in most cases the first failure would have been in the next clock tick
5244 after the received time, and all would be well.
5246 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
5247 being recognized when the domain list was tested by the match_domain
5248 condition in an expansion string.
5250 16. Added the ${str2b64: operator.
5252 17. Exim was always calling setrlimit() to set a large limit for the number of
5253 processes, without checking whether the existing limit was already
5254 adequate. (It did check for the limit on file descriptors.) Furthermore,
5255 errors from getrlimit() and setrlimit() were being ignored. Now they are
5256 logged to the main and panic logs, but Exim does carry on, to try to do its
5257 job under whatever limits there are.
5259 18. Imported PCRE 5.0.
5261 19. Trivial typo in log message " temporarily refused connection" (the leading
5264 20. If the log selector return_path_on_delivery was set and an address was
5265 redirected to /dev/null, the delivery process crashed because it assumed
5266 that a return path would always be set for a "successful" delivery. In this
5267 case, the whole delivery is bypassed as an optimization, and therefore no
5270 21. Internal re-arrangement: the function for sending a challenge and reading
5271 a response while authentication was assuming a zero-terminated challenge
5272 string. It's now changed to take a pointer and a length, to allow for
5273 binary data in such strings.
5275 22. Added the cyrus_sasl authenticator (code supplied by MBM).
5277 23. Exim was not respecting finduser_retries when seeking the login of the
5278 uid under which it was called; it was always trying 10 times. (The default
5279 setting of finduser_retries is zero.) Also, it was sleeping after the final
5280 failure, which is pointless.
5282 24. Implemented tls_on_connect_ports.
5284 25. Implemented acl_smtp_predata.
5286 26. If the domain in control=submission is set empty, Exim assumes that the
5287 authenticated id is a complete email address when it generates From: or
5288 Sender: header lines.
5290 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
5291 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
5292 chown and chgrp in /bin and hostname in /usr/bin.
5294 28. Exim was keeping the "process log" file open after each use, just as it
5295 does for the main log. This opens the possibility of it remaining open for
5296 long periods when the USR1 signal hits a daemon. Occasional processlog
5297 errors were reported, that could have been caused by this. Anyway, it seems
5298 much more sensible not to leave this file open at all, so that is what now
5301 29. The long-running daemon process does not normally write to the log once it
5302 has entered its main loop, and it closes the log before doing so. This is
5303 so that log files can straightforwardly be renamed and moved. However,
5304 there are a couple of unusual error situations where the daemon does write
5305 log entries, and I had neglected to close the log afterwards.
5307 30. The text of an SMTP error response that was received during a remote
5308 delivery was being truncated at 512 bytes. This is too short for some of
5309 the long messages that one sometimes sees. I've increased the limit to
5312 31. It is now possible to make retry rules that apply only when a message has a
5313 specific sender, in particular, an empty sender.
5315 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
5316 it possible to be selective about when SMTP synchronization is enforced.
5318 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
5320 32. Implemented hosts_connection_nolog.
5322 33. Added an ACL for QUIT.
5324 34. Setting "delay_warning=" to disable warnings was not working; it gave a
5327 35. Added mailbox_size and mailbox_filecount to appendfile.
5329 36. Added control = no_multiline_responses to ACLs.
5331 37. There was a bug in the logic of the code that waits for the clock to tick
5332 in the case where the clock went backwards by a substantial amount such
5333 that the microsecond fraction of "now" was more than the microsecond
5334 fraction of "then" (but the whole seconds number was less).
5336 38. Added support for the libradius Radius client library this is found on
5337 FreeBSD (previously only the radiusclient library was supported).
5343 1. When certain lookups returned multiple values in the form name=value, the
5344 quoting of the values was not always being done properly. Specifically:
5345 (a) If the value started with a double quote, but contained no whitespace,
5347 (b) If the value contained whitespace other than a space character (i.e.
5348 tabs or newlines or carriage returns) it was not quoted.
5349 This fix has been applied to the mysql and pgsql lookups by writing a
5350 separate quoting function and calling it from the lookup code. The fix
5351 should probably also be applied to nisplus, ibase and oracle lookups, but
5352 since I cannot test any of those, I have not disturbed their existing code.
5354 2. A hit in the callout cache for a specific address caused a log line with no
5355 reason for rejecting RCPT. Now it says "Previous (cached) callout
5356 verification failure".
5358 3. There was an off-by-one bug in the queryprogram router. An over-long
5359 return line was truncated at 256 instead of 255 characters, thereby
5360 overflowing its buffer with the terminating zero. As well as fixing this, I
5361 have increased the buffer size to 1024 (and made a note to document this).
5363 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
5364 when Exim is waiting for an SMTP response from a remote server, Exim
5365 restarts its select() call on the socket, thereby resetting its timeout.
5366 This is not a problem when such interrupts are rare. Somebody set up a cron
5367 job to run exiwhat every 2 minutes, which is less than the normal select()
5368 timeout (5 or 10 minutes). This meant that the select() timeout never
5369 kicked in because it was always reset. I have fixed this by comparing the
5370 time when an interrupt arrives with the time at the start of the first call
5371 to select(). If more time than the timeout has elapsed, the interrupt is
5372 treated as a timeout.
5374 5. Some internal re-factoring in preparation for the addition of Sieve
5375 extensions (by MH). In particular, the "personal" test is moved to a
5376 separate function, and given an option for scanning Cc: and Bcc: (which is
5377 not set for Exim filters).
5379 6. When Exim created an email address using the login of the caller as the
5380 local part (e.g. when creating a From: or Sender: header line), it was not
5381 quoting the local part when it contained special characters such as @.
5383 7. Installed new OpenBSD configuration files.
5385 8. Reworded some messages for syntax errors in "and" and "or" conditions to
5386 try to make them clearer.
5388 9. Callout options, other than the timeout value, were being ignored when
5389 verifying sender addresses in header lines. For example, when using
5391 verify = header_sender/callout=no_cache
5393 the cache was (incorrectly) being used.
5395 10. Added a missing instance of ${EXE} to the exim_install script; this affects
5396 only the Cygwin environment.
5398 11. When return_path_on_delivery was set as a log selector, if different remote
5399 addresses in the same message used different return paths and parallel
5400 remote delivery occurred, the wrong values would sometimes be logged.
5401 (Whenever a remote delivery process finished, the return path value from
5402 the most recently started remote delivery process was logged.)
5404 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
5405 lines when AUTH and/or TLS are in use. This is the "received protocol"
5406 field. Exim used to use "asmtp" for authenticated SMTP, without any
5407 indication (in the protocol name) for TLS use. Now it follows the RFC and
5408 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
5409 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
5410 names appear in log lines as well as in Received: header lines.
5412 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
5413 extensions, and comparison tests, and to fix some bugs.
5415 14. Changes to the "personal" filter test:
5417 (1) The test was buggy in that it was just doing the equivalent of
5418 "contains" tests on header lines. For example, if a user's address was
5419 anne@some.where, the "personal" test would incorrectly be true for
5421 To: susanne@some.where
5423 This test is now done by extracting each address from the header in turn,
5424 and checking the entire address. Other tests that are part of "personal"
5425 are now done using regular expressions (for example, to check local parts
5426 of addresses in From: header lines).
5428 (2) The list of non-personal local parts in From: addresses has been
5429 extended to include "listserv", "majordomo", "*-request", and "owner-*",
5430 taken from the Sieve specification recommendations.
5432 (3) If the message contains any header line starting with "List-" it is
5433 treated as non-personal.
5435 (4) The test for "circular" in the Subject: header line has been removed
5436 because it now seems ill-conceived.
5438 15. Minor typos in src/EDITME comments corrected.
5440 16. Installed latest exipick from John Jetmore.
5442 17. If headers_add on a router specified a text string that was too long for
5443 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
5444 of string_sprintf() is now avoided.
5446 18. $message_body_size was not set (it was always zero) when running the DATA
5447 ACL and the local_scan() function.
5449 19. For the "mail" command in an Exim filter, no default was being set for
5450 the once_repeat time, causing a random time value to be used if "once" was
5451 specified. (If the value happened to be <= 0, no repeat happened.) The
5452 default is now 0s, meaning "never repeat". The "vacation" command was OK
5453 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
5454 (I found it when inspecting the code).
5456 20. There is now an overall timeout for performing a callout verification. It
5457 defaults to 4 times the callout timeout, which applies to individual SMTP
5458 commands during the callout. The overall timeout applies when there is more
5459 than one host that can be tried. The timeout is checked before trying the
5460 next host. This prevents very long delays if there are a large number of
5461 hosts and all are timing out (e.g. when the network connections are timing
5462 out). The value of the overall timeout can be changed by specifying an
5463 additional sub-option for "callout", called "maxwait". For example:
5465 verify = sender/callout=5s,maxwait=20s
5467 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
5468 to the end before writing, but this should make it even safer).
5470 22. Exim was forgetting that it had advertised PIPELINING for the second and
5471 subsequent messages on an SMTP connection. It was also not resetting its
5472 memory on STARTTLS and an internal HELO.
5474 23. When Exim logs an SMTP synchronization error within a session, it now
5475 records whether PIPELINING has been advertised or not.
5477 24. Added 3 instances of "(long int)" casts to time_t variables that were being
5478 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
5479 rather than long int.
5481 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
5483 26. Added the never_mail option to autoreply.
5489 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
5490 crash if the getsockname() call failed; this can happen if a connection is
5491 closed very soon after it is established. The problem was simply in the
5492 order in which certain operations were done, causing Exim to try to write
5493 to the SMTP stream before it had set up the file descriptor. The bug has
5494 been fixed by making things happen in the correct order.
5500 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
5501 before the connection was closed, thus losing the rejection response.
5503 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
5504 some early Solaris releases, but causes trouble in current releases where
5505 socklen_t is defined.
5507 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
5510 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
5513 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
5514 that releasing the top of it at the end releases what was used for sub-
5515 expansions (unless the block got too big). However, discard this block if
5516 the first thing is a variable or header, so that we can use its block when
5517 it is dynamic (useful for very large $message_headers, for example).
5519 6. Lookups now cache *every* query, not just the most recent. A new, separate
5520 store pool is used for this. It can be recovered when all lookup caches are
5521 flushed. Lookups now release memory at the end of their result strings.
5522 This has involved some general refactoring of the lookup sources.
5524 7. Some code has been added to the store_xxx() functions to reduce the amount
5525 of flapping under certain conditions.
5527 8. log_incoming_interface used to affect only the <= reception log lines. Now
5528 it causes the local interface and port to be added to several more SMTP log
5529 lines, for example "SMTP connection from", and rejection lines.
5531 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
5533 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
5535 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
5536 could be overwritten at the end of the current message (or the start of a
5537 new message if it was set in a HELO ACL). The value is now preserved for
5538 the duration of the SMTP connection.
5540 12. If a transport had a headers_rewrite setting, and a matching header line
5541 contained an unqualified address, that address was qualified, even if it
5542 did not match any rewriting rules. The underlying bug was that the values
5543 of the flags that permit the existence of unqualified sender and recipient
5544 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
5545 non-local messages, and by -bnq for local messages) were not being
5546 preserved with the message after it was received.
5548 13. When Exim was logging an SMTP synchronization error, it could sometimes log
5549 "next input=" as part of the text comprising the host identity instead of
5550 the correct text. The code was using the same buffer for two different
5551 strings. However, depending on which order the printing function evaluated
5552 its arguments, the bug did not always show up. Under Linux, for example, my
5553 test suite worked just fine.
5555 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
5556 doesn't work with some older versions of Perl. It has been changed to "my",
5557 which in any case is probably the better facility to use.
5559 15. A really picky compiler found some instances of statements for creating
5560 error messages that either had too many or two few arguments for the format
5563 16. The size of the buffer for calls to the DNS resolver has been increased
5564 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
5565 for addresses that have a lot of PTR records. This alleviates a problem; it
5566 does not fully solve it.
5568 17. A dnsdb lookup for PTR records that receives more data than will fit in the
5569 buffer now truncates the list and logs the incident, which is the same
5570 action as happens when Exim is looking up a host name and its aliases.
5571 Previously in this situation something unpredictable would happen;
5572 sometimes it was "internal error: store_reset failed".
5574 18. If a server dropped the connection unexpectedly when an Exim client was
5575 using GnuTLS and trying to read a response, the client delivery process
5576 crashed while trying to generate an error log message.
5578 19. If a "warn" verb in an ACL added multiple headers to a message in a single
5579 string, for example:
5581 warn message = H1: something\nH2: something
5583 the text was added as a single header line from Exim's point of view
5584 though it ended up OK in the delivered message. However, searching for the
5585 second and subsequent header lines using $h_h2: did not work. This has been
5586 fixed. Similarly, if a system filter added multiple headers in this way,
5587 the routers could not see them.
5589 20. Expanded the error message when iplsearch is called with an invalid key to
5590 suggest using net-iplsearch in a host list.
5592 21. When running tests using -bh, any delays imposed by "delay" modifiers in
5593 ACLs are no longer actually imposed (and a message to that effect is
5596 22. If a "gecos" field in a passwd entry contained escaped characters, in
5597 particular, if it contained a \" sequence, Exim got it wrong when building
5598 a From: or a Sender: header from that name. A second bug also caused
5599 incorrect handling when an unquoted " was present following a character
5600 that needed quoting.
5602 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
5603 was not being matched caselessly.
5605 24. Arranged for all hyphens in the exim.8 source to be escaped with
5608 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
5609 properly. Recipient callout cache records were still being keyed to include
5610 the sender, even when use_sender was set false. This led to far more
5611 callouts that were necessary. The sender is no longer included in the key
5612 when use_sender is false.
5614 26. Added "control = submission" modifier to ACLs.
5616 27. Added the ${base62d: operator to decode base 62 numbers.
5618 28. dnsdb lookups can now access SRV records.
5620 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
5621 the configuration file.
5623 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
5624 (-v) mode. This makes the output for a verbose queue run more intelligible.
5626 31. Added a use_postmaster feature to recipient callouts.
5628 32. Added the $body_zerocount variable, containing the number of binary zero
5629 bytes in the message body.
5631 33. The time of last modification of the "new" subdirectory is now used as the
5632 "mailbox time last read" when there is a quota error for a maildir
5635 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
5637 35. Added +ignore_unknown as a special item in host lists.
5639 36. Code for decoding IPv6 addresses in host lists is now included, even if
5640 IPv6 support is not being compiled. This fixes a bug in which an IPv6
5641 address was recognized as an IP address, but was then not correctly decoded
5642 into binary, causing unexpected and incorrect effects when compared with
5649 1. Very minor rewording of debugging text in manualroute to say "list of
5650 hosts" instead of "hostlist".
5652 2. If verify=header_syntax was set, and a header line with an unqualified
5653 address (no domain) and a large number of spaces between the end of the
5654 name and the colon was received, the reception process suffered a buffer
5655 overflow, and (when I tested it) crashed. This was caused by some obsolete
5656 code that should have been removed. The fix is to remove it!
5658 3. When running in the test harness, delay a bit after writing a bounce
5659 message to get a bit more predictability in the log output.
5661 4. Added a call to search_tidyup() just before forking a reception process. In
5662 theory, someone could use a lookup in the expansion of smtp_accept_max_
5663 per_host which, without the tidyup, could leave open a database connection.
5665 5. Added the variables $recipient_data and $sender_data which get set from a
5666 lookup success in an ACL "recipients" or "senders" condition, or a router
5667 "senders" option, similar to $domain_data and $local_part_data.
5669 6. Moved the writing of debug_print from before to after the "senders" test
5672 7. Change 4.31/66 (moving the time when the Received: is generated) caused
5673 problems for message scanning, either using a data ACL, or using
5674 local_scan() because the Received: header was not generated till after they
5675 were called (in order to set the time as the time of reception completion).
5676 I have revised the way this works. The header is now generated after the
5677 body is received, but before the ACL or local_scan() are called. After they
5678 are run, the timestamp in the header is updated.
5684 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
5685 before starting a queue runner without re-exec. This happened only when
5686 deliver_drop_privilege was set or when the Exim user was set to root. The
5687 effect of the bug was that timeouts during subsequent deliveries caused
5688 crashes instead of being properly handled. The handler is now left at its
5689 default (and expected) setting.
5691 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
5692 message, again when deliver_drop_privilege is set or Exim is run as root.
5693 The bug described in (1) was not present in this case, but the tidying up
5694 of the other signals was missing. I have made the two cases consistent.
5696 3. The ignore_target_hosts setting on a manualroute router was being ignored
5697 for hosts that were looked up using the /MX notation.
5699 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
5702 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
5703 operated on the sender address. After changing the $sender_address to <>
5704 for the sender address verify, Exim was re-instated it as the original
5705 (before rewriting) address, but remembering that it had rewritten it, so it
5706 wasn't rewriting it again. This bug also had the effect of breaking the
5707 sender address verification caching when the sender address was rewritten.
5709 6. The ignore_target_hosts option was being ignored by the ipliteral router.
5710 This has been changed so that if the ip literal address matches
5711 ignore_target_hosts, the router declines.
5713 7. Added expansion conditions match_domain, match_address, and match_local_
5714 part (NOT match_host).
5716 8. The placeholder for the Received: header didn't have a length field set.
5718 9. Added code to Exim itself and to exim_lock to test for a specific race
5719 condition that could lead to file corruption when using MBX delivery. The
5720 issue is with the lockfile that is created in /tmp. If this file is removed
5721 after a process has opened it but before that process has acquired a lock,
5722 there is the potential for a second process to recreate the file and also
5723 acquire a lock. This could lead to two Exim processes writing to the file
5724 at the same time. The added code performs the same test as UW imapd; it
5725 checks after acquiring the lock that its file descriptor still refers to
5726 the same named file.
5728 10. The buffer for building added header lines was of fixed size, 8192 bytes.
5729 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
5732 11. Added the smtp_active_hostname option. If used, this will typically be made
5733 to depend on the incoming interface address. Because $interface_address is
5734 not set up until the daemon has forked a reception process, error responses
5735 that can happen earlier (such as "too many connections") no longer contain
5738 12. If an expansion in a condition on a "warn" statement fails because a lookup
5739 defers, the "warn" statement is abandoned, and the next ACL statement is
5740 processed. Previously this caused the whole ACL to be aborted.
5742 13. Added the iplsearch lookup type.
5744 14. Added ident_timeout as a log selector.
5746 15. Added tls_certificate_verified as a log selector.
5748 16. Added a global option tls_require_ciphers (compare the smtp transport
5749 option of the same name). This controls incoming TLS connections.
5751 17. I finally figured out how to make tls_require_ciphers do a similar thing
5752 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
5753 before starting the TLS session.
5755 18. Tabs are now shown as \t in -bP output.
5757 19. If the log selector return_path_on_delivery was set, Exim crashed when
5758 bouncing a message because it had too many Received: header lines.
5760 20. If two routers both had headers_remove settings, and the first one included
5761 a superfluous trailing colon, the final name in the first list and the
5762 first name in the second list were incorrectly joined into one item (with a
5763 colon in the middle).
5769 1. Added -C and -D options to the exinext utility, mainly to make it easier
5770 to include in the automated testing, but these could be helpful when
5771 multiple configurations are in use.
5773 2. The exinext utility was not formatting the output nicely when there was
5774 an alternate port involved in the retry record key, nor when there was a
5775 message id as well (for retries that were specific to a specific message
5776 and a specific host). It was also confused by IPv6 addresses, because of
5777 the additional colons they contain. I have fixed the IPv4 problem, and
5778 patched it up to do a reasonable job for IPv6.
5780 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
5781 delivery, the log line now contains "pipelined" if PIPELINING was used.
5783 4. An SMTP transport process used to panic and die if the bind() call to set
5784 an explicit outgoing interface failed. This has been changed; it is now
5785 treated in the same way as a connect() failure.
5787 5. A reference to $sender_host_name in the part of a conditional expansion
5788 that was being skipped was still causing a DNS lookup. This no longer
5791 6. The def: expansion condition was not recognizing references to header lines
5792 that used bh_ and bheader_.
5794 7. Added the _cache feature to named lists.
5796 8. The code for checking quota_filecount in the appendfile transport was
5797 allowing one more file than it should have been.
5799 9. For compatibility with Sendmail, the command line option
5807 and sets the incoming protocol and host name (for trusted callers). The
5808 host name and its colon can be omitted when only the protocol is to be set.
5809 Note the Exim already has two private options, -pd and -ps, that refer to
5810 embedded Perl. It is therefore impossible to set a protocol value of "d" or
5811 "s", but I don't think that's a major issue.
5813 10. A number of refactoring changes to the code, none of which should affect
5816 (a) The number of logging options was getting close to filling up the
5817 32-bit word that was used as a bit map. I have split them into two classes:
5818 those that are passed in the argument to log_write(), and those that are
5819 only ever tested independently outside of that function. These are now in
5820 separate 32-bit words, so there is plenty of room for expansion again.
5821 There is no change in the user interface or the logging behaviour.
5823 (b) When building, for example, log lines, the code previously used a
5824 macro that called string_cat() twice, in order to add two strings. This is
5825 not really sufficiently general. Furthermore, there was one instance where
5826 it was actually wrong because one of the argument was used twice, and in
5827 one call a function was used. (As it happened, calling the function twice
5828 did not affect the overall behaviour.) The macro has been replaced by a
5829 function that can join an arbitrary number of extra strings onto a growing
5832 (c) The code for expansion conditions now uses a table and a binary chop
5833 instead of a serial search (which was left over from when there were very
5834 few conditions). Also, it now recognizes conditions like "pam" even when
5835 the relevant support is not compiled in: a suitably worded error message is
5836 given if an attempt is made to use such a condition.
5838 11. Added ${time_interval:xxxxx}.
5840 12. A bug was causing one of the ddress fields not to be passed back correctly
5841 from remote delivery subprocesses. The field in question was not being
5842 subsequently used, so this caused to problems in practice.
5844 13. Added new log selectors queue_time and deliver_time.
5846 14. Might have fixed a bug in maildirsizefile handling that threw up
5847 "unexpected character" debug warnings, and recalculated the data
5848 unnecessarily. In any case, I expanded the warning message to give more
5851 15. Added the message "Restricted characters in address" to the statements in
5852 the default ACL that block characters like @ and % in local parts.
5854 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
5855 Three changes have been made:
5857 (a) There was a serious bug; a negative response to MAIL caused the whole
5858 recipient domain to be cached as invalid, thereby blocking all messages
5859 to all local parts at the same domain, from all senders. This bug has
5860 been fixed. The domain is no longer cached after a negative response to
5861 MAIL if the sender used is not empty.
5863 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
5866 (c) A new callout option, "use_sender" has been added for people who want
5867 the modified behaviour.
5873 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
5876 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
5877 indeed breaks things for older releases.
5879 3. Added additional logging to the case where there is a problem reading data
5880 from a filter that is running in a subprocess using a pipe, in order to
5881 try to track down a specific problem.
5883 4. Testing facility fudge: when running in the test harness and attempting
5884 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
5885 getting "No route to host". Convert this to a timeout.
5887 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
5890 6. Some OS don't have socklen_t but use size_t instead. This affects the
5891 fifth argument of getsockopt() amongst other things. This is now
5892 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
5893 can be set for individual OS. I have set it for SunOS5, OSF1, and
5894 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
5895 some earlier ones do not.
5897 7. Change 4.30/15 was not doing the test caselessly.
5899 8. The standard form for an IPv6 address literal was being rejected by address
5900 parsing in, for example, MAIL and RCPT commands. An example of this kind of
5901 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5902 this, as well as the form without the "IPv6" on the front (but only when
5903 address literals are enabled, of course).
5905 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
5907 10. Exim crashed if a message with an empty sender address specified by -f
5908 encountered a router with an errors_to setting. This could be provoked only
5909 by a command such as
5913 where an empty string was supplied; "<>" did not hit this bug.
5915 11. Installed PCRE release 4.5.
5917 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
5918 remained set. It is now erased.
5920 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
5921 times from message ids (which are base 36 rather than the normal 62).
5923 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
5924 were being counted as actual protocol errors, and logged if the log
5925 selector +smtp_protocol_error was set. One cannot be perfect in this test,
5926 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
5927 and DATA following a set of rejected RCPTs do not count as protocol errors.
5928 In other words, Exim assumes they were pipelined, though this may not
5929 actually be the case. Of course, in all cases the client gets an
5930 appropriate error code.
5932 15. If a lookup fails in an ACL condition, a message about the failure may
5933 be available; it is used if testing the ACL cannot continue, because most
5934 such messages specify what the cause of the deferral is. However, some
5935 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
5936 that caused an old message to be retained and used if a later statement
5937 caused a defer, replacing the real cause of the deferral.
5939 16. If an IP address had so many PTR records that the DNS lookup buffer
5940 was not large enough to hold them, Exim could crash while trying to process
5941 the truncated data. It now detects and logs this case.
5943 17. Further to 4.21/58, another change has been made: if (and only if) the
5944 first line of a message (the first header line) ends with CRLF, a bare LF
5945 in a subsequent header line has a space inserted after it, so as not to
5946 terminate the header.
5948 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
5949 unnecessarily, used atoi() instead of strtol(), and didn't check the
5950 termination when getting file sizes from file names by regex.
5952 19. Completely re-implemented the support for maildirsize files, in the light
5953 of a number of problems with the previous contributed implementation
5954 (4.30/29). In particular:
5956 . If the quota is zero, the maildirsize file is maintained, but no quota is
5959 . If the maildir directory does not exist, it is created before any attempt
5960 to write a maildirsize file.
5962 . The quota value in the file is just a cache; if the quota is changed in
5963 the transport, the new value overrides.
5965 . A regular expression is available for excluding directories from the
5968 20. The autoreply transport checks the characters in options that define the
5969 message's headers; it allows continued headers, but it was checking with
5970 isspace() after an embedded newline instead of explicitly looking for a
5973 21. If all the "regular" hosts to which an address was routed had passed their
5974 expiry times, and had not reached their retry times, the address was
5975 bounced, even if fallback hosts were defined. Now Exim should go on to try
5978 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
5979 equivalent code in the SMTP transport. Some hosts send humungous responses
5980 to HELO/EHLO, more than 1024 it seems.
5982 23. Refactoring: code in filter.c used (void *) for "any old type" but this
5983 gives compiler warnings in some environments. I've now done it "properly",
5986 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
5987 (because of problems with the built-in one) was declared to return uschar *
5988 instead of char *, causing compiler failure.
5990 25. Fixed a file descriptor leak when processing alias/forward files.
5992 26. Fixed a minor format string issue in dbfn.c.
5994 27. Typo in exim.c: ("dmbnz" for "dbmnz").
5996 28. If a filter file refered to $h_xxx or $message_headers, and the headers
5997 contained RFC 2047 "words", Exim's memory could, under certain conditions,
6000 29. When a sender address is verified, it is cached, to save repeating the test
6001 when there is more than one recipient in a message. However, when the
6002 verification involves a callout, it is possible for different callout
6003 options to be set for different recipients. It is too complicated to keep
6004 track of this in the cache, so now Exim always runs a verification when a
6005 callout is required, relying on the callout cache for the optimization.
6006 The overhead is duplication of the address routing, but this should not be
6009 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
6010 to be verified with callout=postmaster, and the main callout worked but the
6011 postmaster check failed, the verification correctly failed. However, if a
6012 subsequent RCPT command asked for sender verification *without* the
6013 postmaster check, incorrect caching caused this verification also to fail,
6016 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
6017 it was not caching the DNS options (qualify_single, search_parents) that
6018 were used when the lookup failed. A subsequent lookup with different
6019 options therefore always gave the same answer, though there were cases
6020 where it should not have. (Example: a "domains = !$mx_any" option on a
6021 dnslookup router: the "domains" option is always processed without any
6022 widening, but the router might have qualify_single set.) Now Exim uses the
6023 cached value only when the same options are set.
6025 32. Added John Jetmore's "exipick" utility to the distribution.
6027 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
6028 than a timeout (e.g. a certificate is required, and is not provided), an
6029 Exim server now closes the connection immediately. Previously it waited for
6030 the client to close - but if the client is SSL, it seems that they each
6031 wait for each other, leading to a delay before one of them times out.
6033 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
6034 maintained 0.8.x compatibility because I don't think many are using it, and
6035 it is clearly obsolete.
6037 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
6040 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
6041 client certificate was expired. A simple patch fixes this, though I don't
6042 understand the full logic of why the verify callback is called multiple
6045 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
6046 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
6047 which causes problems with some clients (such as the Certicom SSL Plus
6048 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
6049 disables the coutermeasure allowing Eudora to connect."
6051 38. Exim was not checking that a write() to a log file succeeded. This could
6052 lead to Bad Things if a log got too big, in particular if it hit a file
6053 size limit. Exim now panics and dies if it cannot write to a log file, just
6054 as it does if it cannot open a log file.
6056 39. Modified OS/Makefile-Linux so that it now contains
6058 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
6060 The two -D definitions ensure that Exim is compiled with large file
6061 support, which makes it possible to handle log files that are bigger than
6064 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
6065 instance) a domain was checked against a named list that involved a lookup,
6066 causing $domain_data to be set, then another domain was checked against the
6067 same list, then the first domain was re-checked, the value of $domain_data
6068 after the final check could be wrong. In particular, if the second check
6069 failed, it could be set empty. This bug probably also applied to
6072 41. The strip_trailing_dot option was not being applied to the address given
6073 with the -f command-line option.
6075 42. The code for reading a message's header from the spool was incrementing
6076 $received_count, but never initializing it. This meant that the value was
6077 incorrect (doubled) while delivering a message in the same process in which
6078 it was received. In the most common configuration of Exim, this never
6079 happens - a fresh exec is done - but it can happen when
6080 deliver_drop_privilege is set.
6082 43. When Exim logs an SMTP synchronization error - client data sent too soon -
6083 it now includes up to 150 characters of the unexpected data in the log
6086 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
6087 and building data strings. The size of both of these buffers was 10 000
6088 bytes - far larger than anybody would *ever* want, thought I. Needless to
6089 say, somebody hit the limit. I have increased the maximum line length to
6090 20 000 and the maximum data length of concatenated lines to 100 000. I have
6091 also fixed two bugs, because there was no checking on these buffers. Tsk,
6092 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
6093 buffer is too small.
6095 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
6096 lsearch lookups. Now it does.
6098 46. When parsing a route_list item in a manualroute router, a fixed-length
6099 buffer was used for the list of hosts. I made this 1024 bytes long,
6100 thinking that nobody would ever have a list of hosts that long. Wrong.
6101 Somebody had a whole pile of complicated expansion conditions, and the
6102 string was silently truncated, leading to an expansion error. It turns out
6103 that it is easier to change to an unlimited length (owing to other changes
6104 that have happened since this code was originally written) than to build
6105 structure for giving a limitation error. The length of the item that
6106 expands into the list of hosts is now unlimited.
6108 47. The lsearch lookup could not handle data where the length of text line was
6109 more than 4095 characters. Such lines were truncated, leading to shortened
6110 data being returned. It should now handle lines of any length.
6112 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
6113 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
6116 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
6117 replacing strings such as BIN_DIRECTORY in a source file: the replacement
6118 no longer happens in comment lines. A list of replacements is now placed
6119 at the head of all of the source files, except those whose only change is
6120 to replace PERL_COMMAND in the very first #! line.
6122 50. Replaced the slow insertion sort in queue.c, for sorting the list of
6123 messages on the queue, with a bottom-up merge sort, using code contributed
6124 by Michael Haardt. This should make operations like -bp somewhat faster on
6125 large queues. It won't affect queue runners, except when queue_run_in_order
6128 51. Installed eximstats 1.31 in the distribution.
6130 52. Added support for SRV lookups to the dnslookup router.
6132 53. If an ACL referred to $message_body or $message_body_end, the value was not
6133 reset for any messages that followed in the same SMTP session.
6135 54. The store-handling optimization for building very long strings was not
6136 differentiating between the different store pools. I don't think this
6137 actually made any difference in practice, but I've tidied it.
6139 55. While running the routers to verify a sender address, $sender_address
6140 was still set to the sender address. This is wrong, because when routing to
6141 send a bounce to the sender, it would be empty. Therefore, I have changed
6142 it so that, while verifying a sender address, $sender_address is set to <>.
6143 (There is no change to what happens when verifying a recipient address.)
6145 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
6146 target A or AAAA records (if not already returned) without resetting the
6147 qualify_single or search_parents options of the DNS resolver. These are
6148 inappropriate in this case because the targets of MX and SRV records must
6149 be FQDNs. A broken DNS record could cause trouble if it happened to have a
6150 target that, when qualified, matched something in the local domain. These
6151 two options are now turned off when doing these lookups.
6153 57. It seems that at least some releases of Reiserfs (which does not have the
6154 concept of a fixed number of inodes) returns zero and not -1 for the
6155 number of available inodes. This interacted badly with check_spool_inodes,
6156 which assumed that -1 was the "no such thing" setting. What I have done is
6157 to check that the total number of inodes is greater than zero before doing
6158 the test of how many are available.
6160 58. When a "warn" ACL statement has a log_message modifier, the message is
6161 remembered, and not repeated. This is to avoid a lot of repetition when a
6162 message has many recipients that cause the same warning to be written.
6163 However, Exim was preserving the list of already written lines for an
6164 entire SMTP session, which doesn't seem right. The memory is now reset if a
6165 new message is started.
6167 59. The "rewrite" debugging flag was not showing the result of rewriting in the
6168 debugging output unless log_rewrite was also set.
6170 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
6171 of (int)(handle) when we know that handle contains (void *)(-1).
6173 61. The Exim daemon panic-logs an error return when it closes the incoming
6174 connection. However "connection reset by peer" seems to be common, and
6175 isn't really an error worthy of noting specially, so that particular error
6178 62. When Exim is trying to find all the local interfaces, it used to panic and
6179 die if the ioctl to get the interface flags failed. However, it seems that
6180 on at least one OS (Solaris 9) it is possible to have an interface that is
6181 included in the list of interfaces, but for which you get a failure error
6182 for this call. This happens when the interface is not "plumbed" into a
6183 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
6184 failure of the "get flags" call assumes that the interface is down.
6186 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
6187 makes life easier for people who are doing arithmetic on fields extracted
6188 from dates, where you often get leading zeros that should not be
6189 interpreted as octal.
6191 64. Added qualify_domain to the redirect router, to override the global
6194 65. If a pathologically long header line contained very many addresses (the
6195 report of this problem mentioned 10 000) and each of them was rewritten,
6196 Exim could use up a very large amount of memory. (It kept on making new
6197 copies of the header line as it rewrote, and never released the old ones.)
6198 At the expense of a bit more processing, the header rewriting function has
6199 been changed so that it no longer eats memory in this way.
6201 66. The generation of the Received: header has been moved from the time that a
6202 message starts to be received, to the time that it finishes. The timestamp
6203 in the Received: header should now be very close to that of the <= log
6204 line. There are two side-effects of this change:
6206 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
6207 logged header lines no longer include the local Received: line, because
6208 it has not yet been created. The same applies to a copy of the message
6209 that is returned to a non-SMTP sender when a message is rejected.
6211 (b) When a filter file is tested using -bf, no additional Received: header
6212 is added to the test message. After some thought, I decided that this
6215 This change does not affect the value of $received_for. It is still set
6216 after address rewriting, but before local_scan() is called.
6218 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
6220 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
6221 gave an unhelpful panic error message, and a defer error. I have managed to
6222 change this behaviour so that it now rejects any supplied certificate,
6223 which seems right, as the list of acceptable certificates is empty.
6225 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
6226 gave an unhelpful defer error. I have not managed to make this reject any
6227 supplied certificates, but the error message it gives is "no certificate
6228 supplied", which is not helpful.
6230 70. exigrep's output now also includes lines that are not associated with any
6231 message, but which match the given pattern. Implemented by a patch from
6232 Martin Sluka, which also tidied up the Perl a bit.
6234 71. Recipient callout verification, like sender verification, was using <> in
6235 the MAIL FROM command. This isn't really the right thing, since the actual
6236 sender may affect whether the remote host accepts the recipient or not. I
6237 have changed it to use the actual sender in the callout; this means that
6238 the cache record is now keyed on a recipient/sender pair, not just the
6239 recipient address. There doesn't seem to be a real danger of callout loops,
6240 since a callout by the remote host to check the sender would use <>.
6241 [SEE ABOVE: changed after hitting problems.]
6243 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
6244 temporary errors. However, in the case of such a code being given after
6245 the end of a data transmission (i.e. after ".") Exim was failing to write
6246 a retry record for the message. (Yes, there was some broken host that was
6247 actually sending 8xx at this point.)
6249 73. An unknown lookup type in a host list could cause Exim to panic-die when
6250 the list was checked. (An example that provoked this was putting <; in the
6251 middle of a list instead of at the start.) If this happened during a DATA
6252 ACL check, a -D file could be left lying around. This kind of configuration
6253 error no longer causes Exim to die; instead it causes a defer error. The
6254 incident is still logged to the main and panic logs.
6256 74. Buglet left over from Exim 3 conversion. The message "too many messages
6257 in one connection" was written to the rejectlog but not the mainlog, except
6258 when address rewriting (yes!) was being logged.
6260 75. Added write_rejectlog option.
6262 76. When a system filter was run not as root (that is, when system_filter_user
6263 was set), the values of the $n variables were not being returned to the
6264 main process; thus, they were not subsequently available in the $sn
6267 77. Added +return_path_on_delivery log selector.
6269 78. A connection timeout was being treated differently from recipients deferred
6270 when testing hosts_max_try with a message that was older than the host's
6271 retry timeout. (The host should not be counted, thus allowing all hosts to
6272 be tried at least once before bouncing.) This may have been the cause of an
6273 occasionally reported bug whereby a message would remain on the queue
6274 longer than the retry timeout, but would be bounced if a delivery was
6275 forced. I say "may" because I never totally pinned down the problem;
6276 setting up timeout/retry tests is difficult. See also the next item.
6278 79. The ultimate address timeout was not being applied to errors that involved
6279 a combination of host plus message (for example, a timeout on a MAIL
6280 command). When an address resolved to a number of possible hosts, and they
6281 were not all tried for each delivery (e.g. because of hosts_max_try), a
6282 message could remain on the queue longer than the retry timeout.
6284 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
6287 81. Fixed an obscure SMTP outgoing bug which required at least the following
6288 conditions: (a) there was another message waiting for the same server;
6289 (b) the server returned 5xx to all RCPT commands in the first message so
6290 that the message was not completed; (c) the server dropped the connection
6291 or gave a negative response to the RSET that Exim sends to abort the
6292 transaction. The observed case was a dropped connection after DATA that had
6293 been sent in pipelining mode. That is, the server had advertised PIPELINING
6294 but was not implementing it correctly. The effect of the bug was incorrect
6295 behaviour, such as trying another host, and this could lead to a crash.
6301 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
6302 and daemon.c were passed as pointers to ints; they should have been
6303 pointers to socklen_t variables (which are typically unsigned ints).
6305 2. Some signed/unsigned type warnings in the os.c file for Linux have been
6308 3. Fixed a really odd bug that affected only the testing scheme; patching a
6309 certain fixed string in the binary changed the value of another string that
6310 happened to be identical to the end of the original first string.
6312 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
6313 name", it returns that address as the IP address. On some operating
6314 systems (e.g. Solaris), it also passes back the IP address string as the
6315 "host name". However, on others (e.g. Linux), it passes back an empty
6316 string. Exim wasn't checking for this, and was changing the host name to an
6317 empty string, assuming it had been canonicalized.
6319 5. Although rare, it is permitted to have more than one PTR record for a given
6320 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
6321 all the names associated with an address, because they do in Solaris.
6322 However, it seems that they do not in Linux for data that comes from the
6323 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
6324 I found this out when I moved to a new Linux workstation and tried to run
6325 the Exim test suite.
6327 To get round this problem I have changed the code so that it now does its
6328 own call to the DNS to look up PTR records when searching for a host name.
6329 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
6330 addresses that are only in /etc/hosts are still found.
6332 This behaviour is, however, controlled by an option called host_lookup_
6333 order, which defaults to "bydns:byaddr". If people want to use the other
6334 order, or indeed, just use one or the other means of lookup, they can
6335 specify it in this variable.
6337 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
6338 some operating systems, this comes back from gethostbyaddr() as an empty
6339 string, and this is what Exim used to test for. However, it seems that in
6340 other systems, "." is yielded. Exim now tests for this case too.
6342 7. The values of check_spool_space and check_log_space are now held internally
6343 as a number of kilobytes instead of an absolute number of bytes. If a
6344 numbers is specified without 'K' or 'M', it is rounded up to the nearest
6345 kilobyte. This means that much larger values can be stored.
6347 8. Exim monitor: an attempt to get the action menu when not actually pointing
6348 at a message produces an empty menu entitled "No message selected". This
6349 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
6350 no entries in it ("Shell widget menu has zero width and/or height"). So I
6351 have added a single, blank menu entry in this case.
6353 9. Added ${quote_local_part.
6355 10. MIME decoding is now applied to the contents of Subject: header lines when
6358 11. Now that a reference to $sender_host_address automatically causes a reverse
6359 lookup to occur if necessary (4.13/18), there is no need to arrange for a
6360 host lookup before query-style lookups in lists that might use this
6361 variable. This has therefore been abolished, and the "net-" prefix is no
6362 longer necessary for query-style lookups.
6364 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
6365 have been a typo for LFLAGS, so it has been changed.
6367 13. The install script calls Exim with "-C /dev/null" in order to find the
6368 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
6369 to be output. However, since Exim outputs its version number before the
6370 error, it didn't break the script. It just looked ugly. I fixed this by
6371 always allowing "-C /dev/null" if the caller is root.
6373 14. Ignore overlarge ACL variable number when reading spool file - insurance
6374 against a later release with more variables having written the file.
6376 15. The standard form for an IPv6 address literal was being rejected by EHLO.
6377 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
6378 this, as well as the form without the "IPv6" on the front.
6380 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
6381 OS/Makefile-Darwin file.
6383 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
6384 only to LDAP libraries that do not have LDAP_OPT_DEREF.
6386 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
6387 variable. However, some OS use ints for inodes. I've added cast to long int
6388 to get rid of the compiler warning.
6390 19. I had forgotten to lock out "/../" in configuration file names when
6391 ALT_CONFIG_PREFIX was set.
6393 20. Routers used for verification do not need to specify transports. However,
6394 if such a router generated a host list, and callout was configured, Exim
6395 crashed, because it could not find a port number from the (non-existent)
6396 transport. It now assumes port 25 in this circumstance.
6398 21. Added the -t option to exigrep.
6400 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
6401 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
6402 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
6403 them is set, LOOKUP_LSEARCH is forced.
6405 23. "exim -bV" now outputs a list of lookups that are included in the binary.
6407 24. Added sender and host information to the "rejected by local_scan()" log
6408 line; previously there was no indication of these.
6410 25. Added .include_if_exists.
6412 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
6413 Linux. It turns out that not all file systems support this. Apparently some
6414 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
6415 do it.) To cope with this, the error EINVAL, which means that sync-ing is
6416 not supported on the file descriptor, is now ignored when Exim is trying to
6417 sync a directory. This applies only to Linux.
6419 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
6421 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
6422 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
6425 29. Added support for maildirsize files from supplied patch (modified a bit).
6427 30. The use of :fail: followed by an empty string could lead Exim to respond to
6428 sender verification failures with (e.g.):
6430 550 Verification failed for <xxx>
6431 550 Sender verify failed
6433 where the first response line was missing the '-' that indicates it is not
6434 the final line of the response.
6436 31. The loop for finding the name of the user that called Exim had a hardwired
6437 limit of 10; it now uses the value of finduser_retries, which is used for
6438 all other user lookups.
6440 32. Added $received_count variable, available in data and not_smtp ACLs, and at
6443 33. Exim was neglecting to zero errno before one call of strtol() when
6444 expanding a string and expecting an integer value. On some systems this
6445 resulted in spurious "integer overflow" errors. Also, it was casting the
6446 result into an int without checking.
6448 34. Testing for a connection timeout using "timeout_connect" in the retry rules
6449 did not work. The code looks as if it has *never* worked, though it appears
6450 to have been documented since at least release 1.62. I have made it work.
6452 35. The "timeout_DNS" error in retry rules, also documented since at least
6453 1.62, also never worked. As it isn't clear exactly what this means, and
6454 clearly it isn't a major issue, I have abolished the feature by treating it
6455 as "timeout", and writing a warning to the main and panic logs.
6457 36. The display of retry rules for -brt wasn't always showing the error code
6460 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
6461 timeout_connect_A, timeout_connect_MX.
6463 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
6464 to the empty sender.
6466 39. The daemon was not analysing the content of -oX till after it had closed
6467 stderr and disconnected from the controlling terminal. This meant that any
6468 syntax errors were only noted on the panic log, and the return code from
6469 the command was 0. By re-arranging the code a little, I've made the
6470 decoding happen first, so such errors now appear on stderr, and the return
6471 code is 1. However, the actual setting up of the sockets still happens in
6472 the disconnected process, so errors there are still only recorded on the
6475 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
6476 connections (as happens on some IP stacks) was logged at start up time as
6477 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
6478 it from "IPv6 and IPv4", which means that two separate sockets are being
6481 41. The debug output for gethostbyname2() or getipnodebyname() failures now
6482 says whether AF_INET or AF_INET6 was passed as an argument.
6484 42. Exiwhat output was messed up when time zones were included in log
6487 43. Exiwhat now gives more information about the daemon's listening ports,
6488 and whether -tls-on-connect was used.
6490 44. The "port" option of the smtp transport is now expanded.
6492 45. A "message" modifier in a "warn" statement in a non-message ACL was being
6493 silently ignored. Now an error message is written to the main and panic
6496 46. There's a new ACL modifier called "logwrite" which writes to a log file
6497 as soon as it is encountered.
6499 47. Added $local_user_uid and $local_user_gid at routing time.
6501 48. Exim crashed when trying to verify a sender address that was being
6504 49. Exim was recognizing only a space character after ".include". It now also
6505 recognizes a tab character.
6507 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
6508 extracting the relevant information from the specification. The man page no
6509 longer contains scrambled data for the -d option, and I've added a section
6510 at the front about calling Exim under different names.
6512 51. Added "extra_headers" argument to the "mail" command in filter files.
6514 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
6517 53. Installed eximstats 1.29.
6519 54. Added transport_filter_timeout as a generic transport option.
6521 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
6522 Cc: header lines. This was required by RFC 822, but it not required by RFC
6525 56. Exim used to add From:, Date:, and Message-Id: header lines to any
6526 incoming messages that did not have them. Now it does so only if the
6527 message originates locally, that is, if there is no associated remote host
6528 address. When Resent- header lines are present, this applies to the Resent-
6529 lines rather than the non-Resent- lines.
6531 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
6532 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
6534 58. Messages for configuration errors now include the name of the main
6535 configuration file - useful now that there may be more than one file in a
6536 list (.included file names were always shown).
6538 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
6539 for those rare installations that do not start the daemon as root or run it
6540 setuid root. I've cut out the call to initgroups() if the daemon is not
6543 60. The Exim user and group can now be bound into the binary as text strings
6544 that are looked up at the start of Exim's processing.
6546 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
6548 62. Added $mailstore_basename variable.
6550 63. Installed patch to sieve.c from Michael Haardt.
6552 64. When Exim failed to open the panic log after failing to open the main log,
6553 the original message it was trying to log was written to stderr and debug
6554 output, but if they were not available (the usual case in production), it
6555 was lost. Now it is written to syslog before the two lines that record the
6556 failures to open the logs.
6558 65. Users' Exim filters run in subprocesses under the user's uid. It is
6559 possible for a "deliver" command or an alias in a "personal" command to
6560 provoke an address rewrite. If logging of address rewriting is configured,
6561 this fails because the process is not running as root or exim. There may be
6562 a better way of dealing with this, but for the moment (because 4.30 needs
6563 to be released), I have disabled address rewrite logging when running a
6564 filter in a non-root, non-exim process.
6570 1. The buildconfig auxiliary program wasn't quoting the value set for
6571 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
6572 not defined. This bug was masked in 4.22 by the effect that was fixed in
6575 2. Some messages that were rejected after a message id was allocated were
6576 shown as "incomplete" by exigrep. It no longer does this for messages that
6577 are rejected by local_scan() or the DATA or non-SMTP ACLs.
6579 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
6580 have allow_domain_literals set, the ID did not get logged in the <= line.
6581 Domain literals are now always recognized in Message-ID: header lines.
6583 4. The first argument for a ${extract expansion item is the key name or field
6584 number. Leading and trailing spaces in this item were not being ignored,
6585 causing some misleading effects.
6587 5. When deliver_drop_privilege was set, single queue runner processes started
6588 manually (i.e. by the command "exim -q") or by the daemon (which uses the
6589 same command in the process it spins off) were not dropping privilege.
6591 6. When the daemon running as "exim" started a queue runner, it always
6592 re-executed Exim in the spun-off process. This is a waste of effort when
6593 deliver_drop_privilege is set. The new process now just calls the
6594 queue-runner function directly.
6600 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
6603 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
6604 ignored. Though the use of -oP was forcing the writing of a pid file, it
6605 was always written to the default place.
6607 3. If the message "no IP address found for host xxxx" is generated during
6608 incoming verification, it is now followed by identification of the incoming
6609 connection (so you can more easily find what provoked it).
6611 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
6613 5. Added some features to "harden" Exim a bit more against certain attacks:
6615 (a) There is now a build-time option called FIXED_NEVER_USERS that can
6616 be put in Local/Makefile. This is like the never_users runtime option,
6617 but it cannot be overridden. The default setting is "root".
6619 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
6620 prefix string with which any file named in a -C command line option
6623 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
6624 is retained for -C and -D only if the caller of Exim is root. Without
6625 it, the exim user may also use -C and -D and retain privilege.
6627 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
6628 command line option is disabled.
6630 6. Macro names set by the -D option must start with an upper case letter, just
6631 like macro names defined in the configuration file.
6633 7. Added "dereference=" facility to LDAP.
6635 8. Two instances of the typo "uknown" in the source files are fixed.
6637 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
6638 the Configure-Makefile script screwed up while processing it.
6640 10. Incorporated PCRE 4.4.
6642 11. The SMTP synchronization check was not operating right at the start of an
6643 SMTP session. For example, it could not catch a HELO sent before the client
6644 waited for the greeting. There is now a check for outstanding input at the
6645 point when the greeting is written. Because of the duplex, asynchronous
6646 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
6647 way, but not yet received, when the check is performed.
6649 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
6650 on TCP/IP sockets, because this apparently causes some broken clients to
6653 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
6654 unchanged) from the Cygwin maintainer.
6656 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
6657 was supported instead of testing for mbx. Effectively a typo.
6659 15. The spa authenticator server code was not checking that the input it
6660 received was valid base64.
6662 16. The debug output line for the "set" modifier in ACLs was not showing the
6663 name of the variable that was being set.
6665 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
6667 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
6668 lookup on its own. Something else was needed to trigger the lookup. For
6669 example, a match in host_lookup or the need for a host name in a host list.
6670 Now, if $sender_host_name is referenced and the host name has not yet been
6671 looked up, a lookup is performed. If the lookup fails, the variable remains
6672 empty, and $host_lookup_failed is set to "1".
6674 19. Added "eqi" as a case-independent comparison operator.
6676 20. The saslauthd authentication condition could segfault if neither service
6677 nor realm was specified.
6679 21. If an overflowing value such as "2048M" was set for message_size_limit, the
6680 error message that was logged was misleading, and incoming SMTP
6681 connections were dropped. The message is now more accurate, and temporary
6682 errors are given to SMTP connections.
6684 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
6685 (except RSET) with a 421 error, until QUIT is received. However, it was
6686 failing to send a response to QUIT.
6688 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
6689 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
6690 ACL is now run after the helo_try_verify_hosts code.
6692 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
6693 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
6694 case-independent, so other case variants are also recognized). Apparently
6695 some systems use these upper case variants.
6697 25. If more than two messages were waiting for the same host, and a transport
6698 filter was specified for the transport, Exim sent two messages over the
6699 same TCP/IP connection, and then failed with "socket operation on non-
6700 socket" when it tried to send the third.
6702 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
6704 27. The extern definition of crypt16() in expand.c was not being excluded when
6705 the OS had its own crypt16() function.
6707 28. Added bounce_return_body as a new option, and bounce_return_size_limit
6708 as a preferred synonym for return_size_limit, both as an option and as an
6711 29. Added LIBS=-liconv to OS/Makefile-OSF1.
6713 30. Changed the default configuration ACL to relax the local part checking rule
6714 for addresses that are not in any local domains. For these addresses,
6715 slashes and pipe symbols are allowed within local parts, but the sequence
6716 /../ is explicitly forbidden.
6718 31. SPA server authentication was not clearing the challenge buffer before
6721 32. log_message in a "warn" ACL statement was writing to the reject log as
6722 well as to the main log, which contradicts the documentation and doesn't
6723 seem right (because no rejection is happening). So I have stopped it.
6725 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
6726 However, I am unable to do any testing of this.
6728 34. Fixed an infelicity in the appendfile transport. When checking directories
6729 for a mailbox, to see if any needed to be created, it was accidentally
6730 using path names with one or more superfluous leading slashes; tracing
6731 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
6733 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
6734 contents are added to the log line that is written for every discarded
6735 recipient. (Previously a log_message setting was ignored.)
6737 36. The ${quote: operator now quotes the string if it is empty.
6739 37. The install script runs exim in order to find its version number. If for
6740 some reason other than non-existence or emptiness, which it checks, it
6741 could not run './exim', it was installing it with an empty version number,
6742 i.e. as "exim-". This error state is now caught, and the installation is
6745 38. An argument was missing from the function that creates an error message
6746 when Exim fails to connect to the socket for saslauthd authentication.
6747 This could cause Exim to crash, or give a corrupted message.
6749 39. Added isip, isip4, and isip6 to ${if conditions.
6751 40. The ACL variables $acl_xx are now saved with the message, and can be
6752 accessed later in routers, transports, and filters.
6754 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
6755 strings in the file are not string-expanded.
6757 42. If a MAIL command specified a SIZE value that was too large to fit into an
6758 int variable, the check against message_size_limit failed. Such values are
6759 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
6760 day this will have to be increased, but I don't think I want to be around
6761 when emails are that large.
6768 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
6769 iconv() is not standard in FreeBSD.
6771 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
6772 IPv6 enabled. The observed symptom was a segmentation fault on return from
6773 the function os_common_find_running_interfaces() in src/os.c.
6775 3. In the check_special_case() function in daemon.c I had used "errno" as an
6776 argument name, which causes warnings on some systems. This was basically a
6777 typo, since it was named "eno" in the comments!
6779 4. The code that waits for the clock to tick (at a resolution of some fraction
6780 of a second) so as to ensure message-id uniqueness was always waiting for
6781 at least one whole tick, when it could have waited for less. [This is
6782 almost certainly not relevant at current processor speeds, where it is
6783 unlikely to ever wait at all. But we try to future-proof.]
6785 5. The function that sleeps for a time interval that includes fractions of a
6786 second contained a race. It did not block SIGALRM between setting the
6787 timer, and suspending (a couple of lines later). If the interval was short
6788 and the sigsuspend() was delayed until after it had expired, the suspension
6789 never ended. On busy systems this could lead to processes getting stuck for
6792 6. Some uncommon configurations may cause a lookup to happen in a queue runner
6793 process, before it forks any delivery processes. The open lookup caching
6794 mechanism meant that the open file or database connection was passed into
6795 the delivery process. The problem was that delivery processes always tidy
6796 up cached lookup data. This could cause a problem for the next delivery
6797 process started by the queue runner, because the external queue runner
6798 process does not know about the closure. So the next delivery process
6799 still has data in the lookup cache. In the case of a file lookup, there was
6800 no problem because closing a file descriptor in a subprocess doesn't affect
6801 the parent. However, if the lookup was caching a connection to a database,
6802 the connection was closed, and the second delivery process was likely to
6803 see errors such as "PGSQL: query failed: server closed the connection
6804 unexpectedly". The problem has been fixed by closing all cached lookups
6805 in a queue runner before running a delivery process.
6807 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
6808 seem to have the "const" qualifier which it has on other OS. I've
6811 8. Change 4.21/2 was too strict. It is only if there are two authenticators
6812 *of the same type* (client or server) with the same public name that an
6813 error should be diagnosed.
6815 9. When Exim looked up a host name for an IP address, but failed to find the
6816 original IP address when looking up the host name (a safety check), it
6817 output the message "<ip address> does not match any IP for NULL", which was
6818 confusing, to say the least. The bug was that the host name should have
6819 appeared instead of "NULL".
6821 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
6822 user that is built into the binary, and the -C or -D options is used, root
6823 privilege is dropped before the configuration file is read. In addition,
6824 logging is switched to stderr instead of the normal log files. If the
6825 configuration then re-defines the Exim user, the unprivileged environment
6826 is probably not what is expected, so Exim logs a panic warning message (but
6829 However, if deliver_drop_privilege is set, the unprivileged state may well
6830 be exactly what is intended, so the warning has been cut out in that case,
6831 and Exim is allowed to try to write to its normal log files.
6837 1. smtp_return_error_details was not giving details for temporary sender
6838 or receiver verification errors.
6840 2. Diagnose a configuration error if two authenticators have the same public
6843 3. Exim used not to create the message log file for a message until the first
6844 delivery attempt. This could be confusing when incoming messages were held
6845 for policy or load reasons. The message log file is now created at the time
6846 the message is received, and an initial "Received" line is written to it.
6848 4. The automatically generated man page for command line options had a minor
6849 bug that caused no ill effects; however, a more serious problem was that
6850 the procedure for building the man page automatically didn't always
6851 operate. Consequently, release 4.20 contains an out-of-date version. This
6852 shouldn't happen again.
6854 5. When building Exim with embedded Perl support, the script that builds the
6855 Makefile was calling 'perl' to find its compile-time parameters, ignoring
6856 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
6858 6. The freeze_tell option was not being used for messages that were frozen on
6859 arrival, either by an ACL or by local_scan().
6861 7. Added the smtp_incomplete_transaction log selector.
6863 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
6864 was accepting AUTH without a new EHLO.
6866 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
6867 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
6870 10. Logging of TCP/IP connections (when configured) now happens in the main
6871 daemon process instead of the child process, so that the TCP/IP connection
6872 count is more accurate (but it can never be perfect).
6874 11. The use of "drop" in a nested ACL was not being handled correctly in the
6875 outer ACL. Now, if condition failure induced by the nested "drop" causes
6876 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
6877 or "require"), the connection is dropped.
6879 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
6880 that yield "discard" can now be used with an "accept" or a "discard" verb,
6881 but an error is generated for any others (because I can't see a useful way
6882 to define what should happen).
6884 13. When an ACL is read dynamically from a file (or anywhere else), the lines
6885 are now processed in the same way as lines in the Exim configuration file.
6886 In particular, continuation lines are supported.
6888 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
6890 15. Added -ti meaning -t -i.
6892 16. Check for letters, digits, hyphens, and dots in the names of dnslist
6893 domains, and warn by logging if others are found.
6895 17. At least on BSD, alignment is not guaranteed for the array of ifreq's
6896 returned from GIFCONF when Exim is trying to find the list of interfaces on
6897 a host. The code in os.c has been modified to copy each ifreq to an aligned
6898 structure in all cases.
6900 Also, in some cases, the returned ifreq's were being copied to a 'struct
6901 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
6902 means the last couple of bytes of an IPv6 address could be chopped if the
6903 ifreq contained only a normal sockaddr (14 bytes storage).
6905 18. Named domain lists were not supported in the hosts_treat_as_local option.
6906 An entry such as +xxxx was not recognized, and was treated as a literal
6909 19. Ensure that header lines added by a DATA ACL are included in the reject log
6910 if the ACL subsequently rejects the message.
6912 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
6913 MD5 (which is deprecated).
6915 21. When testing a filter file using -bf, Exim was writing a message when it
6916 took the sender from a "From " line in the message, but it was not doing so
6917 when it took $return_path from a Return-Path: header line. It now does.
6919 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
6920 with a valid header line field name (a series of printing characters
6921 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
6923 23. Changed "disc" in the source to "disk" to conform to the documentation and
6924 the book and for uniformity.
6926 24. Ignore Sendmail's -Ooption=value command line item.
6928 25. When execve() failed while trying to run a command in a pipe transport,
6929 Exim was returning EX_UNAVAILABLE (69) from the subprocess. However, this
6930 could be confused with a return value of 69 from the command itself. This
6931 has been changed to 127, the value the shell returns if it is asked to run
6932 a non-existent command. The wording for the related log line suggests a
6933 non-existent command as the problem.
6935 26. If received_header_text expands to an empty string, do not add a Received:
6936 header line to the message. (Well, it adds a token one on the spool, but
6937 marks it "old" so that it doesn't get used or transmitted.)
6939 27. Installed eximstats 1.28 (addition of -nt option).
6941 28. There was no check for failure on the call to getsockname() in the daemon
6942 code. This can fail if there is a shortage of resources on the system, with
6943 ENOMEM, for example. A temporary error is now given on failure.
6945 29. Contrary to the C standard, it seems that in some environments, the
6946 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
6947 program. Exim now does this explicitly; it affects the formatting of
6948 timestamps using strftime().
6950 30. If exiqsumm was given junk data, it threw up some uninitialized variable
6951 complaints. I've now initialized all the variables, to avoid this.
6953 32. Header lines added by a system filter were not being "seen" during
6954 transport-time rewrites.
6956 33. The info_callback() function passed to OpenSSL is set up with type void
6957 (*)(SSL *, int, int), as described somewhere. However, when calling the
6958 function (actually a macro) that sets it up, the type void(*)() is
6959 expected. I've put in a cast to prevent warnings from picky compilers.
6961 34. If a DNS black list lookup found a CNAME record, but there were no A
6962 records associated with the domain it pointed at, Exim crashed.
6964 35. If a DNS black list lookup returned more than one A record, Exim ignored
6965 all but the first. It now scans all returned addresses if a particular IP
6966 value is being sought. In this situation, the contents of the
6967 $dnslist_value variable are a list of all the addresses, separated by a
6970 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
6971 to accept a host name and all its aliases if the forward lookup for any of
6972 them yielded the IP address of the incoming connection. Now it accepts only
6973 those names whose forward lookup yields the correct IP address. Any other
6974 names are discarded. This closes a loophole whereby a rogue DNS
6975 administrator could create reverse DNS records to break through a
6976 wildcarded host restriction in an ACL.
6978 37. If a user filter or a system filter that ran in a subprocess used any of
6979 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
6980 the wrong values were passed to the pipe command ($thisaddress had the
6981 value of $0, $0 had the value of $1, etc). This bug was introduced by
6982 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
6984 38. Improved the line breaking for long SMTP error messages from ACLs.
6985 Previously, if there was no break point between 40 and 75 characters, Exim
6986 left the rest of the message alone. Two changes have been made: (a) I've
6987 reduced the minimum length to 35 characters; (b) if it can't find a break
6988 point between 35 and 75 characters, it looks ahead and uses the first one
6989 that it finds. This may give the occasional overlong line, but at least the
6990 remaining text gets split now.
6992 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
6993 file descriptors might be low, and that setting 1000 would always raise it.
6994 It turns out that in some environments, the limit is already over 1000 and
6995 that lowering it causes trouble. So now Exim takes care not to decrease it.
6997 40. When delivering a message, the value of $return_path is set to $sender_
6998 address at the start of routing (routers may change the value). By an
6999 oversight, this default was not being set up when an address was tested by
7000 -bt or -bv, which affected the outcome if any router or filter referred to
7003 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
7004 log only when "message" or "log_message" are set. However, if one of the
7005 conditions was an address verification, or a call to a nested ACL, the
7006 messages generated by the underlying test were being passed through. This
7007 no longer happens. The underlying message is available in $acl_verify_
7008 message for both "message" and "log_message" expansions, so it can be
7009 passed through if needed.
7011 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
7012 new expansion $bh_ to give the encoded byte string without charset
7013 translation. Translation happens only if iconv() is available; HAVE_ICONV
7014 indicates this at build time. HEADERS_CHARSET gives the charset to
7015 translate to; headers_charset can change it in the configuration, and
7016 "headers charset" can change it in an individual filter file.
7018 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
7019 that creates RFC 2047 encoded "words" labels them as that charset instead
7020 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
7021 expansion operator; (ii) when Exim creates a From: line for a local
7022 message; (iii) when a header line is rewritten to include a "phrase" part.
7024 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
7025 buggy, causing it to skip the first lines of messages whose message ID
7026 ended in 'D'. This would not have bitten before Exim release 4.14, because
7027 message IDs were unlikely to end in 'D' before then. The effect was to have
7028 incorrect size information for certain domains.
7030 45. #include "config.h" was missing at the start of the crypt16.c module. This
7031 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
7034 46. If there was a timeout during a "random" callout check, Exim treated it as
7035 a failure of the random address, and carried on sending RSET and the real
7036 address. If the delay was just some slowness somewhere, the response to the
7037 original RCPT would be taken as a response to RSET and so on, causing
7038 mayhem of various kinds.
7040 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
7041 when I implemented it. It didn't allow for the fact that some option values
7042 may legitimately be negative (e.g. size_addition), and it didn't even do
7043 the right test for positive values.
7045 48. Domain names in DNS records are case-independent. Exim always looks them up
7046 in lower case. Some resolvers return domain names in exactly the case they
7047 appear in the zone file, that is, they may contain uppercase letters. Not
7048 all resolvers do this - some return always lower case. Exim was treating a
7049 change of case by a resolver as a change of domain, similar to a widening
7050 of a domain abbreviation. This triggered its re-routing code and so it was
7051 trying to route what was effectively the same domain again. This normally
7052 caused routing to fail (because the router wouldn't handle the domain
7053 twice). Now Exim checks for this case specially, and just changes the
7054 casing of the domain that it ultimately uses when it transmits the message
7057 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
7060 50. If a filter generated a file delivery with a non-absolute name (possible if
7061 no home directory exists for the router), the forbid_file option was not
7064 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
7065 the existing equality matching.
7067 52. Exim was using ints instead of ino_t variables in some places where it was
7068 dealing with inode numbers.
7070 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
7071 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
7072 called TMPDIR, and if it finds it is different, it changes its value.
7074 54. The smtp_printf() function is now made available to local_scan() so
7075 additional output lines can be written before returning. There is also an
7076 smtp_fflush() function to enable the detection of a dropped connection.
7077 The variables smtp_input and smtp_batched_input are exported to
7080 55. Changed the default runtime configuration: the message "Unknown user"
7081 has been removed from the ACL, and instead placed on the localuser router,
7082 using the cannot_route_message feature. This means that any verification
7083 failures that generate their own messages won't get overridden. Similarly,
7084 the "Unrouteable address" message that was in the ACL for unverifiable
7085 relay addresses has also been removed.
7087 56. Added hosts_avoid_esmtp to the smtp transport.
7089 57. The exicyclog script was not checking for the esoteric option
7090 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
7091 will work only if exicyclog is run under the appropriate euid.
7093 58. Following a discussion on the list, the rules by which Exim recognises line
7094 endings on incoming messages have been changed. The -dropcr and drop_cr
7095 options are now no-ops, retained only for backwards compatibility. The
7096 following line terminators are recognized: LF CRLF CR. However, special
7097 processing applies to CR:
7099 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
7100 nor a local message in the state where . is a terminator.
7102 (ii) If a bare CR is encountered in a header line, an extra space is added
7103 after the line terminator so as not to end the header. The reasoning
7104 behind this is that bare CRs in header lines are most likely either
7105 to be mistakes, or people trying to play silly games.
7107 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
7108 was being incorrectly given as 18 bytes larger than it should have been.
7109 This is a VOB (very old bug).
7111 60. This may never have affected anything current, but just in case it has:
7112 When the local host is found other than at the start of a list of hosts,
7113 the local host, those with the same MX, and any that follow, are discarded.
7114 When the list in question was part of a longer list of hosts, the following
7115 hosts (not currently being processed) were also being discarded. This no
7116 longer happens. I'm not sure if this situation could ever has previously
7119 61. Added the "/MX" feature to lists of hosts in the manualroute and query
7122 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
7123 header. This is something that is recommended in a new Internet Draft, and
7124 is something that is documented as being done by Sendmail. There are two
7125 possible values. For messages generated by the autoreply transport, Exim
7128 Auto-Submitted: auto-replied
7130 whereas for all other generated messages (e.g. bounces) it adds
7132 Auto-Submitted: auto-generated
7134 63. The "personal" condition in filters now includes a test for the
7135 Auto-Submitted: header. If it contains the string "auto-" the message it
7136 not considered personal.
7138 64. Added rcpt_include_affixes as a generic transport option.
7140 65. Added queue_only_override (default true).
7142 66. Added the syslog_duplication option.
7144 67. If what should have been the first header line of a message consisted of
7145 a space followed by a colon, Exim was mis-interpreting it as a header line.
7146 It isn't of course - it is syntactically invalid and should therefore be
7147 treated as the start of the message body. The misbehaviour could have
7148 caused a number of strange effects, including loss of data in subsequent
7149 header lines, and spool format errors.
7151 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
7152 client host had authenticated. This control can now be exercised by an ACL
7153 for more flexibility.
7155 69. By default, callouts do not happen when testing with -bh. There is now a
7156 variant, -bhc, which does actually run the callout code, including
7157 consulting and updating the callout cache.
7159 70. Added support for saslauthd authentication, courtesy of Alexander
7162 71. If statvfs() failed on the spool or log directories while checking their
7163 size for availability, Exim confusingly gave the error "space shortage".
7164 Furthermore, in debugging mode it crashed with a floating point exception.
7165 These checks are done if check_{spool,log}_{space,inodes} are set, and when
7166 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
7167 serious problem, Exim now writes to the main and panic logs when this
7168 happens, with details of the failure. It then refuses to accept the
7169 incoming message, giving the message "spool directory problem" or "log
7170 directory problem" with a 421 code for SMTP messages.
7172 72. When Exim is about to re-exec itself, it ensures that the file descriptors
7173 0, 1, and 2 exist, because some OS complain for execs without them (see
7174 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
7175 descriptors. However, the code omitted to check that the open succeeded,
7176 causing mysterious errors if for some reason the permissions on /dev/null
7177 got screwed. Now Exim writes a message to the main and panic logs, and
7178 bombs out if it can't open /dev/null.
7180 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
7181 interact so that it is all more flexible. It is supposed to remain
7182 backwards compatible. Also added extra_local_interfaces.
7184 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
7185 to bomb out with an assertion failure - to the client this appears as a
7186 connection drop. This problem occurs in the part of the code that was taken
7187 from the Samba project. Fortunately, the assertion is in a very simple
7188 function, so I have fixed this by reproducing the function inline in the
7189 one place where it is called, and arranging for authentication to fail
7190 instead of killing the process with assert().
7192 75. The SPA client code was not working when the server requested OEM rather
7193 than Unicode encoding.
7195 76. Added code to make require_files with a specific uid setting more usable in
7196 the case where statting the file as root fails - usually a non-root-mounted
7197 NFS file system. When this happens and the failure is EACCES, Exim now
7198 forks a subprocess and does the per-uid checking as the relevant uid.
7200 77. Added process_log_path.
7202 78. If log_file_path was not explicitly set, a setting of check_log_space or
7203 check_log_inodes was ignored.
7205 79. If a space check for the spool or log partitions fails, the incident is now
7206 logged. Of course, in the latter case the data may get lost...
7208 80. Added the %p formatting code to string_format() so that it can be used to
7209 print addresses in debug_print(). Adjusted all the address printing in the
7210 debugging in store.c to use %p rather than %d.
7212 81. There was a concern that a line of code in smtp_in.c could overflow a
7213 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
7214 initially expressed, the concern was not well-founded, because trailing
7215 spaces are removed early. However, if the trailing spaces were followed by
7216 a NULL, they did not get removed, so the overflow was possible. Two fixes
7219 (a) I re-wrote the offending code in a cleaner fashion.
7220 (b) If an incoming SMTP command contains a NULL character, it is rejected
7223 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
7224 runs initgroups(), so that if the Exim user is in any additional groups,
7225 they will be used during message reception.
7231 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / users / heiko / exim.git / blob