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.
8 Since Exim version 4.90
9 -----------------------
11 JH/01 Replace the store_release() internal interface with store_newblock(),
12 which internalises the check required to safely use the old one, plus
13 the allocate and data copy operations duplicated in both (!) of the
16 JH/02 Disallow '/' characters in queue names specified for the "queue=" ACL
17 modifier. This matches the restriction on the commandline.
19 JH/03 Fix pgsql lookup for multiple result-tuples with a single column.
20 Previously only the last row was returned.
22 JH/04 Bug 2217: Tighten up the parsing of DKIM signature headers. Previously
23 we assumed that tags in the header were well-formed, and parsed the
24 element content after inspecting only the first char of the tag.
25 Assumptions at that stage could crash the receive process on malformed
28 JH/05 Bug 2215: Fix crash associated with dnsdb lookup done from DKIM ACL.
29 While running the DKIM ACL we operate on the Permanent memory pool so that
30 variables created with "set" persist to the DATA ACL. Also (at any time)
31 DNS lookups that fail create cache records using the Permanent pool. But
32 expansions release any allocations made on the current pool - so a dnsdb
33 lookup expansion done in the DKIM ACL releases the memory used for the
34 DNS negative-cache, and bad things result. Solution is to switch to the
35 Main pool for expansions.
36 While we're in that code, add checks on the DNS cache during store_reset,
37 active in the testsuite.
38 Problem spotted, and debugging aided, by Wolfgang Breyha.
40 JH/06 Fix issue with continued-connections when the DNS shifts unreliably.
41 When none of the hosts presented to a transport match an already-open
42 connection, close it and proceed with the list. Previously we would
43 queue the message. Spotted by Lena with Yahoo, probably involving
46 JH/07 Bug 2214: Fix SMTP responses resulting from non-accept result of MIME ACL.
47 Previously a spurious "250 OK id=" response was appended to the proper
50 JH/08 The "support for" informational output now, which built with Content
51 Scanning support, has a line for the malware scanner interfaces compiled
52 in. Interface can be individually included or not at build time.
54 JH/09 The "aveserver", "kavdaemon" and "mksd" interfaces are now not included
55 by the template makefile "src/EDITME". The "STREAM" support for an older
56 ClamAV interface method is removed.
58 JH/10 Bug 2223: Fix mysql lookup returns for the no-data case (when the number of
59 rows affected is given instead).
61 JH/11 The runtime Berkeley DB library version is now additionally output by
62 "exim -d -bV". Previously only the compile-time version was shown.
64 JH/12 Bug 2230: Fix cutthrough routing for nonfirst messages in an initiating
65 SMTP connection. Previously, when one had more receipients than the
66 first, an abortive onward connection was made. Move to full support for
67 multiple onward connections in sequence, handling cutthrough connection
68 for all multi-message initiating connections.
70 JH/13 Bug 2229: Fix cutthrough routing for nonstandard port numbers defined by
71 routers. Previously, a multi-recipient message would fail to match the
72 onward-connection opened for the first recipient, and cause its closure.
74 JH/14 Bug 2174: A timeout on connect for a callout was also erroneously seen as
75 a timeout on read on a GnuTLS initiating connection, resulting in the
76 initiating connection being dropped. This mattered most when the callout
77 was marked defer_ok. Fix to keep the two timeout-detection methods
80 JH/15 Relax results from ACL control request to enable cutthrough, in
81 unsupported situations, from error to silently (except under debug)
82 ignoring. This covers use with PRDR, frozen messages, queue-only and
85 HS/01 Fix Buffer overflow in base64d() (CVE-2018-6789)
87 JH/16 Fix bug in DKIM verify: a buffer overflow could corrupt the malloc
88 metadata, resulting in a crash in free().
90 PP/01 Fix broken Heimdal GSSAPI authenticator integration.
91 Broken in f2ed27cf5, missing an equals sign for specified-initialisers.
92 Broken also in d185889f4, with init system revamp.
94 JH/17 Bug 2113: Fix conversation closedown with the Avast malware scanner.
95 Previously we abruptly closed the connection after reading a malware-
96 found indication; now we go on to read the "scan ok" response line,
99 JH/18 Bug 2239: Enforce non-usability of control=utf8_downconvert in the mail
100 ACL. Previously, a crash would result.
102 JH/19 Speed up macro lookups during configuration file read, by skipping non-
103 macro text after a replacement (previously it was only once per line) and
104 by skipping builtin macros when searching for an uppercase lead character.
106 JH/20 DANE support moved from Experimental to mainline. The Makefile control
107 for the build is renamed.
109 JH/21 Fix memory leak during multi-message connections using STARTTLS. A buffer
110 was allocated for every new TLS startup, meaning one per message. Fix
111 by only allocating once (OpenSSL) or freeing on TLS-close (GnuTLS).
113 JH/22 Bug 2236: When a DKIM verification result is overridden by ACL, DMARC
114 reported the original. Fix to report (as far as possible) the ACL
115 result replacing the original.
121 JH/01 Rework error string handling in TLS interface so that the caller in
122 more cases is responsible for logging. This permits library-sourced
123 string to be attached to addresses during delivery, and collapses
124 pairs of long lines into single ones.
126 PP/01 Allow PKG_CONFIG_PATH to be set in Local/Makefile and use it correctly
127 during configuration. Wildcards are allowed and expanded.
129 JH/02 Rework error string handling in DKIM to pass more info back to callers.
130 This permits better logging.
132 JH/03 Rework the transport continued-connection mechanism: when TLS is active,
133 do not close it down and have the child transport start it up again on
134 the passed-on TCP connection. Instead, proxy the child (and any
135 subsequent ones) for TLS via a unix-domain socket channel. Logging is
136 affected: the continued delivery log lines do not have any DNSSEC, TLS
137 Certificate or OCSP information. TLS cipher information is still logged.
139 JH/04 Shorten the log line for daemon startup by collapsing adjacent sets of
140 identical IP addresses on different listening ports. Will also affect
143 PP/02 Bug 2070: uClibc defines __GLIBC__ without providing glibc headers;
144 add noisy ifdef guards to special-case this sillyness.
145 Patch from Bernd Kuhls.
147 JH/05 Tighten up the checking in isip4 (et al): dotted-quad components larger
148 than 255 are no longer allowed.
150 JH/06 Default openssl_options to include +no_ticket, to reduce load on peers.
151 Disable the session-cache too, which might reduce our load. Since we
152 currrectly use a new context for every connection, both as server and
153 client, there is no benefit for these.
154 GnuTLS appears to not support tickets server-side by default (we don't
155 call gnutls_session_ticket_enable_server()) but client side is enabled
156 by default on recent versions (3.1.3 +) unless the PFS priority string
159 PP/03 Add $SOURCE_DATE_EPOCH support for reproducible builds, per spec at
160 <https://reproducible-builds.org/specs/source-date-epoch/>.
162 JH/07 Fix smtp transport use of limited max_rcpt under mua_wrapper. Previously
163 the check for any unsuccessful recipients did not notice the limit, and
164 erroneously found still-pending ones.
166 JH/08 Pipeline CHUNKING command and data together, on kernels that support
167 MSG_MORE. Only in-clear (not on TLS connections).
169 JH/09 Avoid using a temporary file during transport using dkim. Unless a
170 transport-filter is involved we can buffer the headers in memory for
171 creating the signature, and read the spool data file once for the
172 signature and again for transmission.
174 JH/10 Enable use of sendfile in Linux builds as default. It was disabled in
175 4.77 as the kernel support then wasn't solid, having issues in 64bit
176 mode. Now, it's been long enough. Add support for FreeBSD also.
178 JH/11 Bug 2104: Fix continued use of a transport connection with TLS. In the
179 case where the routing stage had gathered several addresses to send to
180 a host before calling the transport for the first, we previously failed
181 to close down TLS in the old transport process before passing the TCP
182 connection to the new process. The new one sent a STARTTLS command
183 which naturally failed, giving a failed delivery and bloating the retry
184 database. Investigation and fix prototype from Wolfgang Breyha.
186 JH/12 Fix check on SMTP command input synchronisation. Previously there were
187 false-negatives in the check that the sender had not preempted a response
188 or prompt from Exim (running as a server), due to that code's lack of
189 awareness of the SMTP input buffering.
191 PP/04 Add commandline_checks_require_admin option.
192 Exim drops privileges sanely, various checks such as -be aren't a
193 security problem, as long as you trust local users with access to their
194 own account. When invoked by services which pass untrusted data to
195 Exim, this might be an issue. Set this option in main configuration
196 AND make fixes to the calling application, such as using `--` to stop
199 JH/13 Do pipelining under TLS. Previously, although safe, no advantage was
200 taken. Now take care to pack both (client) MAIL,RCPT,DATA, and (server)
201 responses to those, into a single TLS record each way (this usually means
202 a single packet). As a side issue, smtp_enforce_sync now works on TLS
205 PP/05 OpenSSL/1.1: use DH_bits() for more accurate DH param sizes. This
206 affects you only if you're dancing at the edge of the param size limits.
207 If you are, and this message makes sense to you, then: raise the
208 configured limit or use OpenSSL 1.1. Nothing we can do for older
211 JH/14 For the "sock" variant of the malware scanner interface, accept an empty
212 cmdline element to get the documented default one. Previously it was
215 JH/15 Fix a crash in the smtp transport caused when two hosts in succession
216 are unsuable for non-message-specific reasons - eg. connection timeout,
217 banner-time rejection.
219 JH/16 Fix logging of delivery remote port, when specified by router, under
222 PP/06 Repair manualroute's ability to take options in any order, even if one
223 is the name of a transport.
226 HS/01 Cleanup, prevent repeated use of -p/-oMr (CVE-2017-1000369)
228 JH/17 Change the list-building routines interface to use the expanding-string
229 triplet model, for better allocation and copying behaviour.
231 JH/18 Prebuild the data-structure for "builtin" macros, for faster startup.
232 Previously it was constructed the first time a possibly-matching string
233 was met in the configuration file input during startup; now it is done
236 JH/19 Bug 2141: Use the full-complex API for Berkeley DB rather than the legacy-
237 compatible one, to avoid the (poorly documented) possibility of a config
238 file in the working directory redirecting the DB files, possibly correpting
239 some existing file. CVE-2017-10140 assigned for BDB.
241 JH/20 Bug 2147: Do not defer for a verify-with-callout-and-random which is not
242 cache-hot. Previously, although the result was properly cached, the
243 initial verify call returned a defer.
245 JH/21 Bug 2151: Avoid using SIZE on the MAIL for a callout verify, on any but
246 the main verify for receipient in uncached-mode.
248 JH/22 Retire historical build files to an "unsupported" subdir. These are
249 defined as "ones for which we have no current evidence of testing".
251 JH/23 DKIM: enforce the DNS pubkey record "h" permitted-hashes optional field,
252 if present. Previously it was ignored.
254 JH/24 Start using specified-initialisers in C structure init coding. This is
255 a C99 feature (it's 2017, so now considered safe).
257 JH/25 Use one-bit bitfields for flags in the "addr" data structure. Previously
258 if was a fixed-sized field and bitmask ops via macros; it is now more
261 PP/07 GitHub PR 56: Apply MariaDB build fix.
262 Patch provided by Jaroslav Škarvada.
264 PP/08 Bug 2161: Fix regression in sieve quoted-printable handling introduced
265 during Coverity cleanups [4.87 JH/47]
266 Diagnosis and fix provided by Michael Fischer v. Mollard.
268 JH/26 Fix DKIM bug: when the pseudoheader generated for signing was exactly
269 the right size to place the terminating semicolon on its own folded
270 line, the header hash was calculated to an incorrect value thanks to
271 the (relaxed) space the fold became.
273 HS/02 Fix Bug 2130: large writes from the transport subprocess where chunked
274 and confused the parent.
276 JH/27 Fix SOCKS bug: an unitialized pointer was deref'd by the transport process
277 which could crash as a result. This could lead to undeliverable messages.
279 JH/28 Logging: "next input sent too soon" now shows where input was truncated
282 JH/29 Fix queue_run_in_order to ignore the PID portion of the message ID. This
283 matters on fast-turnover and PID-randomising systems, which were getting
284 out-of-order delivery.
286 JH/30 Fix a logging bug on aarch64: an unsafe routine was previously used for
287 a possibly-overlapping copy. The symptom was that "Remote host closed
288 connection in response to HELO" was logged instead of the actual 4xx
291 JH/31 Fix CHUNKING code to properly flush the unwanted chunk after an error.
292 Previously only that bufferd was discarded, resulting in SYMTP command
295 JH/32 DKIM: when a message has multiple signatures matching an identity given
296 in dkim_verify_signers, run the dkim acl once for each. Previously only
297 one run was done. Bug 2189.
299 JH/33 Downgrade an unfound-list name (usually a typo in the config file) from
300 "panic the current process" to "deliberately defer". The panic log is
301 still written with the problem list name; the mail and reject logs now
302 get a temp-reject line for the message that was being handled, saying
303 something like "domains check lookup or other defer". The SMTP 451
304 message is still "Temporary local problem".
306 JH/34 Bug 2199: Fix a use-after-free while reading smtp input for header lines.
307 A crafted sequence of BDAT commands could result in in-use memory beeing
308 freed. CVE-2017-16943.
310 HS/03 Bug 2201: Fix checking for leading-dot on a line during headers reading
311 from SMTP input. Previously it was always done; now only done for DATA
312 and not BDAT commands. CVE-2017-16944.
314 JH/35 Bug 2201: Flush received data in BDAT mode after detecting an error fatal
315 to the message (such as an overlong header line). Previously this was
316 not done and we did not exit BDAT mode. Followon from the previous item
317 though a different problem.
323 JH/01 Bug 1922: Support IDNA2008. This has slightly different conversion rules
324 than -2003 did; needs libidn2 in addition to libidn.
326 JH/02 The path option on a pipe transport is now expanded before use.
328 PP/01 GitHub PR 50: Do not call ldap_start_tls_s on ldapi:// connections.
329 Patch provided by "Björn", documentation fix added too.
331 JH/03 Bug 2003: fix Proxy Protocol v2 handling: the address size field was
332 missing a wire-to-host endian conversion.
334 JH/04 Bug 2004: fix CHUNKING in non-PIPELINEING mode. Chunk data following
335 close after a BDAT command line could be taken as a following command,
336 giving a synch failure. Fix by only checking for synch immediately
337 before acknowledging the chunk.
339 PP/02 GitHub PR 52: many spelling fixes, which include fixing parsing of
340 no_require_dnssec option and creation of _HAVE_TRANSPORT_APPEND_MAILDIR
341 macro. Patches provided by Josh Soref.
343 JH/05 Have the EHLO response advertise VRFY, if there is a vrfy ACL defined.
344 Previously we did not; the RFC seems ambiguous and VRFY is not listed
345 by IANA as a service extension. However, John Klensin suggests that we
348 JH/06 Bug 2017: Fix DKIM verification in -bh test mode. The data feed into
349 the dkim code may be unix-mode line endings rather than smtp wire-format
350 CRLF, so prepend a CR to any bare LF.
352 JH/07 Rationalise the coding for callout smtp conversations and transport ones.
353 As a side-benfit, callouts can now use PIPELINING hence fewer round-trips.
355 JH/08 Bug 2016: Fix DKIM verification vs. CHUNKING. Any BDAT commands after
356 the first were themselves being wrongly included in the feed into dkim
357 processing; with most chunk sizes in use this resulted in an incorrect
358 body hash calculated value.
360 JH/09 Bug 2014: permit inclusion of a DKIM-Signature header in a received
361 DKIM signature block, for verification. Although advised against by
362 standards it is specifically not ruled illegal.
364 JH/10 Bug 2025: Fix reception of (quoted) local-parts with embedded spaces.
366 JH/11 Bug 2029: Fix crash in DKIM verification when a message signature block is
367 missing a body hash (the bh= tag).
369 JH/12 Bug 2018: Re-order Proxy Protocol startup versus TLS-on-connect startup.
370 It seems that HAProxy sends the Proxy Protocol information in clear and
371 only then does a TLS startup, so do the same.
373 JH/13 Bug 2027: Avoid attempting to use TCP Fast Open for non-transport client
374 TCP connections (such as for Spamd) unless the daemon successfully set
375 Fast Open mode on its listening sockets. This fixes breakage seen on
376 too-old kernels or those not configured for Fast Open, at the cost of
377 requiring both directions being enabled for TFO, and TFO never being used
378 by non-daemon-related Exim processes.
380 JH/14 Bug 2000: Reject messages recieved with CHUNKING but with malformed line
381 endings, at least on the first header line. Try to canonify any that get
382 past that check, despite the cost.
384 JH/15 Angle-bracket nesting (an error inserted by broken sendmails) levels are
385 now limited to an arbitrary five deep, while parsing addresses with the
386 strip_excess_angle_brackets option enabled.
388 PP/03 Bug 2018: For Proxy Protocol and TLS-on-connect, do not over-read and
389 instead leave the unprompted TLS handshake in socket buffer for the
390 TLS library to consume.
392 PP/04 Bug 2018: Also handle Proxy Protocol v2 safely.
394 PP/05 FreeBSD compat: handle that Ports no longer create /usr/bin/perl
396 JH/16 Drop variables when they go out of scope. Memory management drops a whole
397 region in one operation, for speed, and this leaves assigned pointers
398 dangling. Add checks run only under the testsuite which checks all
399 variables at a store-reset and panics on a dangling pointer; add code
400 explicitly nulling out all the variables discovered. Fixes one known
401 bug: a transport crash, where a dangling pointer for $sending_ip_address
402 originally assigned in a verify callout, is re-used.
404 PP/06 Drop '.' from @INC in various Perl scripts.
406 PP/07 Switch FreeBSD iconv to always use the base-system libc functions.
408 PP/08 Reduce a number of compilation warnings under clang; building with
409 CC=clang CFLAGS+=-Wno-dangling-else -Wno-logical-op-parentheses
410 should be warning-free.
412 JH/17 Fix inbound CHUNKING when DKIM disabled at runtime.
414 HS/01 Fix portability problems introduced by PP/08 for platforms where
415 realloc(NULL) is not equivalent to malloc() [SunOS et al].
417 HS/02 Bug 1974: Fix missing line terminator on the last received BDAT
418 chunk. This allows us to accept broken chunked messages. We need a more
419 general solution here.
421 PP/09 Wrote util/chunking_fixqueue_finalnewlines.pl to help recover
422 already-broken messages in the queue.
424 JH/18 Bug 2061: Fix ${extract } corrupting an enclosing ${reduce } $value.
426 JH/19 Fix reference counting bug in routing-generated-address tracking.
432 JH/01 Use SIZE on MAIL FROM in a cutthrough connection, if the destination
433 supports it and a size is available (ie. the sending peer gave us one).
435 JH/02 The obsolete acl condition "demime" is removed (finally, after ten
436 years of being deprecated). The replacements are the ACLs
437 acl_smtp_mime and acl_not_smtp_mime.
439 JH/03 Upgrade security requirements imposed for hosts_try_dane: previously
440 a downgraded non-dane trust-anchor for the TLS connection (CA-style)
441 or even an in-clear connection were permitted. Now, if the host lookup
442 was dnssec and dane was requested then the host is only used if the
443 TLSA lookup succeeds and is dnssec. Further hosts (eg. lower priority
444 MXs) will be tried (for hosts_try_dane though not for hosts_require_dane)
445 if one fails this test.
446 This means that a poorly-configured remote DNS will make it incommunicado;
447 but it protects against a DNS-interception attack on it.
449 JH/04 Bug 1810: make continued-use of an open smtp transport connection
450 non-noisy when a race steals the message being considered.
452 JH/05 If main configuration option tls_certificate is unset, generate a
453 self-signed certificate for inbound TLS connections.
455 JH/06 Bug 165: hide more cases of password exposure - this time in expansions
456 in rewrites and routers.
458 JH/07 Retire gnutls_require_mac et.al. These were nonfunctional since 4.80
459 and logged a warning sing 4.83; now they are a configuration file error.
461 JH/08 Bug 1836: Fix crash in VRFY handling when handed an unqualified name
462 (lacking @domain). Apply the same qualification processing as RCPT.
464 JH/09 Bug 1804: Avoid writing msglog files when in -bh or -bhc mode.
466 JH/10 Support ${sha256:} applied to a string (as well as the previous
469 JH/11 Cutthrough: avoid using the callout hints db on a verify callout when
470 a cutthrough deliver is pending, as we always want to make a connection.
471 This also avoids re-routing the message when later placing the cutthrough
472 connection after a verify cache hit.
473 Do not update it with the verify result either.
475 JH/12 Cutthrough: disable when verify option success_on_redirect is used, and
476 when routing results in more than one destination address.
478 JH/13 Cutthrough: expand transport dkim_domain option when testing for dkim
479 signing (which inhibits the cutthrough capability). Previously only
480 the presence of an option was tested; now an expansion evaluating as
481 empty is permissible (obviously it should depend only on data available
482 when the cutthrough connection is made).
484 JH/14 Fix logging of errors under PIPELINING. Previously the log line giving
485 the relevant preceding SMTP command did not note the pipelining mode.
487 JH/15 Fix counting of empty lines in $body_linecount and $message_linecount.
488 Previously they were not counted.
490 JH/16 DANE: treat a TLSA lookup response having all non-TLSA RRs, the same
491 as one having no matching records. Previously we deferred the message
492 that needed the lookup.
494 JH/17 Fakereject: previously logged as a normal message arrival "<="; now
495 distinguished as "(=".
497 JH/18 Bug 1867: make the fail_defer_domains option on a dnslookup router work
498 for missing MX records. Previously it only worked for missing A records.
500 JH/19 Bug 1850: support Radius libraries that return REJECT_RC.
502 JH/20 Bug 1872: Ensure that acl_smtp_notquit is run when the connection drops
503 after the data-go-ahead and data-ack. Patch from Jason Betts.
505 JH/21 Bug 1846: Send DMARC forensic reports for reject and quarantine results,
506 even for a "none" policy. Patch from Tony Meyer.
508 JH/22 Fix continued use of a connection for further deliveries. If a port was
509 specified by a router, it must also match for the delivery to be
512 JH/23 Bug 1874: fix continued use of a connection for further deliveries.
513 When one of the recipients of a message was unsuitable for the connection
514 (has no matching addresses), we lost track of needing to mark it
515 deferred. As a result mail would be lost.
517 JH/24 Bug 1832: Log EHLO response on getting conn-close response for HELO.
519 JH/25 Decoding ACL controls is now done using a binary search; the source code
520 takes up less space and should be simpler to maintain. Merge the ACL
521 condition decode tables also, with similar effect.
523 JH/26 Fix problem with one_time used on a redirect router which returned the
524 parent address unchanged. A retry would see the parent address marked as
525 delivered, so not attempt the (identical) child. As a result mail would
528 JH/27 Fix a possible security hole, wherein a process operating with the Exim
529 UID can gain a root shell. Credit to http://www.halfdog.net/ for
530 discovery and writeup. Ubuntu bug 1580454; no bug raised against Exim
533 JH/28 Enable {spool,log} filesystem space and inode checks as default.
534 Main config options check_{log,spool}_{inodes,space} are now
535 100 inodes, 10MB unless set otherwise in the configuration.
537 JH/29 Fix the connection_reject log selector to apply to the connect ACL.
538 Previously it only applied to the main-section connection policy
541 JH/30 Bug 1897: fix callouts connection fallback from TLS to cleartext.
543 PP/01 Changed default Diffie-Hellman parameters to be Exim-specific, created
544 by me. Added RFC7919 DH primes as an alternative.
546 PP/02 Unbreak build via pkg-config with new hash support when crypto headers
547 are not in the system include path.
549 JH/31 Fix longstanding bug with aborted TLS server connection handling. Under
550 GnuTLS, when a session startup failed (eg because the client disconnected)
551 Exim did stdio operations after fclose. This was exposed by a recent
552 change which nulled out the file handle after the fclose.
554 JH/32 Bug 1909: Fix OCSP proof verification for cases where the proof is
555 signed directly by the cert-signing cert, rather than an intermediate
556 OCSP-signing cert. This is the model used by LetsEncrypt.
558 JH/33 Bug 1914: Ensure socket is nonblocking before draining after SMTP QUIT.
560 HS/01 Fix leak in verify callout under GnuTLS, about 3MB per recipient on
561 an incoming connection.
563 HS/02 Bug 1802: Do not half-close the connection after sending a request
566 HS/03 Use "auto" as the default EC curve parameter. For OpenSSL < 1.0.2
567 fallback to "prime256v1".
569 JH/34 SECURITY: Use proper copy of DATA command in error message.
570 Could leak key material. Remotely exploitable. CVE-2016-9963.
576 JH/01 Bug 1664: Disable OCSP for GnuTLS library versions at/before 3.3.16
577 and 3.4.4 - once the server is enabled to respond to an OCSP request
578 it does even when not requested, resulting in a stapling non-aware
579 client dropping the TLS connection.
581 TF/01 Code cleanup: Overhaul the debug_selector and log_selector machinery to
582 support variable-length bit vectors. No functional change.
584 TF/02 Improve the consistency of logging incoming and outgoing interfaces.
585 The I= interface field on outgoing lines is now after the H= remote
586 host field, same as incoming lines. There is a separate
587 outgoing_interface log selector which allows you to disable the
590 JH/02 Bug 728: Close logfiles after a daemon-process "exceptional" log write.
591 If not running log_selector +smtp_connection the mainlog would be held
592 open indefinitely after a "too many connections" event, including to a
593 deleted file after a log rotate. Leave the per net connection logging
594 leaving it open for efficiency as that will be quickly detected by the
595 check on the next write.
597 HS/01 Bug 1671: Fix post transport crash.
598 Processing the wait-<transport> messages could crash the delivery
599 process if the message IDs didn't exist for some reason. When
600 using 'split_spool_directory=yes' the construction of the spool
601 file name failed already, exposing the same netto behaviour.
603 JH/03 Bug 425: Capture substrings in $regex1, $regex2 etc from regex &
604 mime_regex ACL conditions.
606 JH/04 Bug 1686: When compiled with EXPERIMENTAL_DSN_INFO: Add extra information
607 to DSN fail messages (bounces): remote IP, remote greeting, remote response
608 to HELO, local diagnostic string.
610 JH/05 Downgrade message for a TLS-certificate-based authentication fail from
611 log line to debug. Even when configured with a tls authenticator many
612 client connections are expected to not authenticate in this way, so
613 an authenticate fail is not an error.
615 HS/02 Add the Exim version string to the process info. This way exiwhat
616 gives some more detail about the running daemon.
618 JH/06 Bug 1395: time-limit caching of DNS lookups, to the TTL value. This may
619 matter for fast-change records such as DNSBLs.
621 JH/07 Bug 1678: Always record an interface option value, if set, as part of a
622 retry record, even if constant. There may be multiple transports with
623 different interface settings and the retry behaviour needs to be kept
626 JH/08 Bug 1586: exiqgrep now refuses to run if there are unexpected arguments.
628 JH/09 Bug 1700: ignore space & tab embedded in base64 during decode.
630 JH/10 Bug 840: fix log_defer_output option of pipe transport
632 JH/11 Bug 830: use same host for all RCPTS of a message, even under
633 hosts_randomize. This matters a lot when combined with mua_wrapper.
635 JH/12 Bug 1706: percent and underbar characters are no longer escaped by the
636 ${quote_pgsql:<string>} operator.
638 JH/13 Bug 1708: avoid misaligned access in cached lookup.
640 JH/14 Change header file name for freeradius-client. Relevant if compiling
641 with Radius support; from the Gentoo tree and checked under Fedora.
643 JH/15 Bug 1712: Introduce $prdr_requested flag variable
645 JH/16 Bug 1714: Permit an empty string as expansion result for transport
646 option transport_filter, meaning no filtering.
648 JH/17 Bug 1713: Fix non-PDKIM_DEBUG build. Patch from Jasen Betts.
650 JH/18 Bug 1709: When built with TLS support, the tls_advertise_hosts option now
651 defaults to "*" (all hosts). The variable is now available when not built
652 with TLS, default unset, mainly to enable keeping the testsuite sane.
653 If a server certificate is not supplied (via tls_certificate) an error is
654 logged, and clients will find TLS connections fail on startup. Presumably
655 they will retry in-clear.
656 Packagers of Exim are strongly encouraged to create a server certificate
657 at installation time.
659 HS/03 Add -bP config_file as a synonym for -bP configure_file, for consistency
660 with the $config_file variable.
662 JH/19 Two additional event types: msg:rcpt:defer and msg:rcpt:host:defer. Both
663 in transport context, after the attempt, and per-recipient. The latter type
664 is per host attempted. The event data is the error message, and the errno
665 information encodes the lookup type (A vs. MX) used for the (first) host,
666 and the trailing two digits of the smtp 4xx response.
668 GF/01 Bug 1715: Fix for race condition in exicyclog, where exim could attempt
669 to write to mainlog (or rejectlog, paniclog) in the window between file
670 creation and permissions/ownership being changed. Particularly affects
671 installations where exicyclog is run as root, rather than exim user;
672 result is that the running daemon panics and dies.
674 JH/20 Bug 1701: For MySQL lookups, support MySQL config file option group names.
676 JH/21 Bug 1720: Add support for priority groups and weighted-random proxy
677 selection for the EXPERIMENTAL_SOCKS feature, via new per-proxy options
678 "pri" and "weight". Note that the previous implicit priority given by the
679 list order is no longer honoured.
681 JH/22 Bugs 963, 1721: Fix some corner cases in message body canonicalization
684 JH/23 Move SOCKS5 support from Experimental to mainline, enabled for a build
685 by defining SUPPORT_SOCKS.
687 JH/26 Move PROXY support from Experimental to mainline, enabled for a build
688 by defining SUPPORT_PROXY. Note that the proxy_required_hosts option
689 is renamed to hosts_proxy, and the proxy_{host,target}_{address,port}.
690 variables are renamed to proxy_{local,external}_{address,port}.
692 JH/27 Move Internationalisation support from Experimental to mainline, enabled
693 for a build by defining SUPPORT_I18N
695 JH/28 Bug 1745: Fix redis lookups to handle (quoted) spaces embedded in parts
696 of the query string, and make ${quote_redis:} do that quoting.
698 JH/29 Move Events support from Experimental to mainline, enabled by default
699 and removable for a build by defining DISABLE_EVENT.
701 JH/30 Updated DANE implementation code to current from Viktor Dukhovni.
703 JH/31 Fix bug with hosts_connection_nolog and named-lists which were wrongly
704 cached by the daemon.
706 JH/32 Move Redis support from Experimental to mainline, enabled for a build
707 by defining LOOKUP_REDIS. The libhiredis library is required.
709 JH/33 Bug 1748: Permit ACL dnslists= condition in non-smtp ACLs if explicit
710 keys are given for lookup.
712 JH/34 Bug 1192: replace the embedded copy of PolarSSL RSA routines in the DKIM
713 support, by using OpenSSL or GnuTLS library ones. This means DKIM is
714 only supported when built with TLS support. The PolarSSL SHA routines
715 are still used when the TLS library is too old for convenient support.
717 JH/35 Require SINGLE_DH_USE by default in OpenSSL (main config option
718 openssl_options), for security. OpenSSL forces this from version 1.1.0
719 server-side so match that on older versions.
721 JH/36 Bug 1778: longstanding bug in memory use by the ${run } expansion: A fresh
722 allocation for $value could be released as the expansion processing
723 concluded, but leaving the global pointer active for it.
725 JH/37 Bug 1769: Permit a VRFY ACL to override the default 252 response,
726 and to use the domains and local_parts ACL conditions.
728 JH/38 Fix cutthrough bug with body lines having a single dot. The dot was
729 incorrectly not doubled on cutthrough transmission, hence seen as a
730 body-termination at the receiving system - resulting in truncated mails.
731 Commonly the sender saw a TCP-level error, and retransmitted the message
732 via the normal store-and-forward channel. This could result in duplicates
733 received - but deduplicating mailstores were liable to retain only the
734 initial truncated version.
736 JH/39 Bug 1781: Fix use of DKIM private-keys having trailing '=' in the base-64.
738 JH/40 Fix crash in queryprogram router when compiled with EXPERIMENTAL_SRS.
740 JH/41 Bug 1792: Fix selection of headers to sign for DKIM: bottom-up. While
741 we're in there, support oversigning also; bug 1309.
743 JH/42 Bug 1796: Fix error logged on a malware scanner connection failure.
745 HS/04 Add support for keep_environment and add_environment options.
747 JH/43 Tidy coding issues detected by gcc --fsanitize=undefined. Some remain;
748 either intentional arithmetic overflow during PRNG, or testing config-
751 JH/44 Bug 1800: The combination of a -bhc commandline option and cutthrough
752 delivery resulted in actual delivery. Cancel cutthrough before DATA
755 JH/45 Fix cutthrough, when connection not opened by verify and target hard-
756 rejects a recipient: pass the reject to the originator.
758 JH/46 Multiple issues raised by Coverity. Some were obvious or plausible bugs.
759 Many were false-positives and ignorable, but it's worth fixing the
762 JH/47 Fix build on HP-UX and older Solaris, which need (un)setenv now also
763 for the new environment-manipulation done at startup. Move the routines
764 from being local to tls.c to being global via the os.c file.
766 JH/48 Bug 1807: Fix ${extract } for the numeric/3-string case. While preparsing
767 an extract embedded as result-arg for a map, the first arg for extract
768 is unavailable so we cannot tell if this is a numbered or keyed
769 extraction. Accept either.
775 JH/01 Bug 1545: The smtp transport option "retry_include_ip_address" is now
778 JH/02 The smtp transport option "multi_domain" is now expanded.
780 JH/03 The smtp transport now requests PRDR by default, if the server offers
783 JH/04 Certificate name checking on server certificates, when exim is a client,
784 is now done by default. The transport option tls_verify_cert_hostnames
785 can be used to disable this per-host. The build option
786 EXPERIMENTAL_CERTNAMES is withdrawn.
788 JH/05 The value of the tls_verify_certificates smtp transport and main options
789 default to the word "system" to access the system default CA bundle.
790 For GnuTLS, only version 3.0.20 or later.
792 JH/06 Verification of the server certificate for a TLS connection is now tried
793 (but not required) by default. The verification status is now logged by
794 default, for both outbound TLS and client-certificate supplying inbound
797 JH/07 Changed the default rfc1413 lookup settings to disable calls. Few
800 JH/08 The EXPERIMENTAL_DSN compile option is no longer needed; all Delivery
801 Status Notification (bounce) messages are now MIME format per RFC 3464.
802 Support for RFC 3461 DSN options NOTIFY,ENVID,RET,ORCPT can be advertised
803 under the control of the dsn_advertise_hosts option, and routers may
804 have a dsn_lasthop option.
806 JH/09 A timeout of 2 minutes is now applied to all malware scanner types by
807 default, modifiable by a malware= option. The list separator for
808 the options can now be changed in the usual way. Bug 68.
810 JH/10 The smtp_receive_timeout main option is now expanded before use.
812 JH/11 The incoming_interface log option now also enables logging of the
813 local interface on delivery outgoing connections.
815 JH/12 The cutthrough-routing facility now supports multi-recipient mails,
816 if the interface and destination host and port all match.
818 JH/13 Bug 344: The verify = reverse_host_lookup ACL condition now accepts a
821 JH/14 Bug 1573: The spam= ACL condition now additionally supports Rspamd.
822 Patch from Andrew Lewis.
824 JH/15 Bug 670: The spamd_address main option (for the spam= ACL condition)
825 now supports optional time-restrictions, weighting, and priority
826 modifiers per server. Patch originally by <rommer@active.by>.
828 JH/16 The spamd_address main option now supports a mixed list of local
829 and remote servers. Remote servers can be IPv6 addresses, and
830 specify a port-range.
832 JH/17 Bug 68: The spamd_address main option now supports an optional
833 timeout value per server.
835 JH/18 Bug 1581: Router and transport options headers_add/remove can
836 now have the list separator specified.
838 JH/19 Bug 392: spamd_address, and clamd av_scanner, now support retry
841 JH/20 Bug 1571: Ensure that $tls_in_peerdn is set, when verification fails
844 JH/21 Support for the A6 type of dns record is withdrawn.
846 JH/22 Bug 608: The result of a QUIT or not-QUIT toplevel ACL now matters
847 rather than the verbs used.
849 JH/23 Bug 1572: Increase limit on SMTP confirmation message copy size
850 from 255 to 1024 chars.
852 JH/24 Verification callouts now attempt to use TLS by default.
854 HS/01 DNSSEC options (dnssec_require_domains, dnssec_request_domains)
855 are generic router options now. The defaults didn't change.
857 JH/25 Bug 466: Add RFC2322 support for MIME attachment filenames.
858 Original patch from Alexander Shikoff, worked over by JH.
860 HS/02 Bug 1575: exigrep falls back to autodetection of compressed
861 files if ZCAT_COMMAND is not executable.
863 JH/26 Bug 1539: Add timeout/retry options on dnsdb lookups.
865 JH/27 Bug 286: Support SOA lookup in dnsdb lookups.
867 JH/28 Bug 1588: Do not use the A lookup following an AAAA for setting the FQDN.
868 Normally benign, it bites when the pair was led to by a CNAME;
869 modern usage is to not canonicalize the domain to a CNAME target
870 (and we were inconsistent anyway for A-only vs AAAA+A).
872 JH/29 Bug 1632: Removed the word "rejected" from line logged for ACL discards.
874 JH/30 Check the forward DNS lookup for DNSSEC, in addition to the reverse,
875 when evaluating $sender_host_dnssec.
877 JH/31 Check the HELO verification lookup for DNSSEC, adding new
878 $sender_helo_dnssec variable.
880 JH/32 Bug 1397: Enable ECDHE on OpenSSL, just the NIST P-256 curve.
882 JH/33 Bug 1346: Note MAIL cmd seen in -bS batch, to avoid smtp_no_mail log.
884 JH/34 Bug 1648: Fix a memory leak seen with "mailq" and large queues.
886 JH/35 Bug 1642: Fix support of $spam_ variables at delivery time. Was
887 documented as working, but never had. Support all but $spam_report.
889 JH/36 Bug 1659: Guard checking of input smtp commands again pseudo-command
890 added for tls authenticator.
892 HS/03 Add perl_taintmode main config option
898 TL/01 When running the test suite, the README says that variables such as
899 no_msglog_check are global and can be placed anywhere in a specific
900 test's script, however it was observed that placement needed to be near
901 the beginning for it to behave that way. Changed the runtest perl
902 script to read through the entire script once to detect and set these
903 variables, reset to the beginning of the script, and then run through
904 the script parsing/test process like normal.
906 TL/02 The BSD's have an arc4random API. One of the functions to induce
907 adding randomness was arc4random_stir(), but it has been removed in
908 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
909 function when detected.
911 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
912 cause callback expansion.
914 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
915 syntax errors in an expansion can be treated as a string instead of
916 logging or causing an error, due to the internal use of bool_lax
917 instead of bool when processing it.
919 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
920 server certificates when making smtp deliveries.
922 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
924 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
926 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
928 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
929 Merged patch from Sebastian Wiedenroth.
931 JH/05 Fix results-pipe from transport process. Several recipients, combined
932 with certificate use, exposed issues where response data items split
933 over buffer boundaries were not parsed properly. This eventually
934 resulted in duplicates being sent. This issue only became common enough
935 to notice due to the introduction of connection certificate information,
936 the item size being so much larger. Found and fixed by Wolfgang Breyha.
938 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
939 size buffer was used, resulting in syntax errors when an expansion
942 JH/07 Add support for directories of certificates when compiled with a GnuTLS
943 version 3.3.6 or later.
945 JH/08 Rename the TPDA experimental facility to Event Actions. The #ifdef
946 is EXPERIMENTAL_EVENT, the main-configuration and transport options
947 both become "event_action", the variables become $event_name, $event_data
948 and $event_defer_errno. There is a new variable $verify_mode, usable in
949 routers, transports and related events. The tls:cert event is now also
950 raised for inbound connections, if the main configuration event_action
953 TL/06 In test suite, disable OCSP for old versions of openssl which contained
954 early OCSP support, but no stapling (appears to be less than 1.0.0).
956 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
957 server certificate names available under the smtp transport option
958 "tls_verify_cert_hostname" now do not permit multi-component wildcard
961 JH/10 Time-related extraction expansions from certificates now use the main
962 option "timezone" setting for output formatting, and are consistent
963 between OpenSSL and GnuTLS compilations. Bug 1541.
965 JH/11 Fix a crash in mime ACL when meeting a zero-length, quoted or RFC2047-
966 encoded parameter in the incoming message. Bug 1558.
968 JH/12 Bug 1527: Autogrow buffer used in reading spool files. Since they now
969 include certificate info, eximon was claiming there were spoolfile
972 JH/13 Bug 1521: Fix ldap lookup for single-attr request, multiple-attr return.
974 JH/14 Log delivery-related information more consistently, using the sequence
975 "H=<name> [<ip>]" wherever possible.
977 TL/07 Bug 1547: Omit RFCs from release. Draft and RFCs have licenses which
978 are problematic for Debian distribution, omit them from the release
981 JH/15 Updates and fixes to the EXPERIMENTAL_DSN feature.
983 JH/16 Fix string representation of time values on 64bit time_t architectures.
986 JH/17 Fix a null-indirection in certextract expansions when a nondefault
987 output list separator was used.
992 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
993 checkers that were complaining about end of non-void function with no
996 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
997 This was a regression introduced in 4.83 by another bugfix.
999 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
1001 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
1002 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
1008 TF/01 Correctly close the server side of TLS when forking for delivery.
1010 When a message was received over SMTP with TLS, Exim failed to clear up
1011 the incoming connection properly after forking off the child process to
1012 deliver the message. In some situations the subsequent outgoing
1013 delivery connection happened to have the same fd number as the incoming
1014 connection previously had. Exim would try to use TLS and fail, logging
1015 a "Bad file descriptor" error.
1017 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
1018 utilities have not been installed.
1020 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
1021 temporary space as the ACL may create new global variables.
1023 TL/01 LDAP support uses per connection or global context settings, depending
1024 upon the detected version of the libraries at build time.
1026 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
1027 to extract and use the src ip:port in logging and expansions as if it
1028 were a direct connection from the outside internet. PPv2 support was
1029 updated based on HAProxy spec change in May 2014.
1031 JH/02 Add ${listextract {number}{list}{success}{fail}}.
1033 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
1034 Properly escape header and check for NULL return.
1036 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
1039 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
1041 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
1042 characters in header names, implemented as a verify condition.
1043 Contributed by Michael Fischer v. Mollard.
1045 TL/05 Rename SPF condition results err_perm and err_temp to standardized
1046 results permerror and temperror. Previous values are deprecated but
1047 still accepted. In a future release, err_perm and err_temp will be
1048 completely removed, which will be a backward incompatibility if the
1049 ACL tests for either of these two old results. Patch contributed by
1050 user bes-internal on the mailing list.
1052 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
1054 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
1055 selectors, in both main and reject logs.
1057 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
1060 JH/07 Add malware type "sock" for talking to simple daemon.
1062 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
1064 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
1065 routers/transports under cutthrough routing.
1067 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
1068 numbers. Touch up "bool" conditional to keep the same definition.
1070 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
1072 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
1074 JH/12 Expand items in router/transport headers_add or headers_remove lists
1075 individually rather than the list as a whole. Bug 1452.
1077 Required for reasonable handling of multiple headers_ options when
1078 they may be empty; requires that headers_remove items with embedded
1079 colons must have them doubled (or the list-separator changed).
1081 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
1082 view the policy declared in the DMARC record. Currently, $dmarc_status
1083 is a combined value of both the record presence and the result of the
1086 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
1088 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
1089 dnslookup router and the smtp transport (applying to the forward
1092 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
1093 of ldap servers used for a specific lookup. Patch provided by Heiko
1096 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
1097 New variable $lookup_dnssec_authenticated for observability.
1099 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
1100 Patch submitted by Lars Timman.
1102 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
1104 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
1105 Requires trusted mode and valid format message id, aborts otherwise.
1106 Patch contributed by Heiko Schlichting.
1108 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
1109 certextract with support for various fields. Bug 1358.
1111 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
1112 is requested by default, modifiable by smtp transport option
1115 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
1116 operate on certificate variables to give certificate fingerprints
1117 Also new ${sha256:cert_variable}.
1119 JH/23 The PRDR feature is moved from being Experimental into the mainline.
1121 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
1122 Christian Aistleitner.
1124 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
1126 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
1127 file. Patch from Wolfgang Breyha.
1129 JH/25 Expand the coverage of the delivery $host and $host_address to
1130 client authenticators run in verify callout. Bug 1476.
1132 JH/26 Port service names are now accepted for tls_on_connect_ports, to
1133 align with daemon_smtp_ports. Bug 72.
1135 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
1136 support and error reporting did not work properly.
1138 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
1139 and is readable. Patch from Andrew Colin Kissa.
1141 TL/14 Enhance documentation of ${run expansion and how it parses the
1142 commandline after expansion, particularly in the case when an
1143 unquoted variable expansion results in an empty value.
1145 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
1147 PP/02 Fix internal collision of T_APL on systems which support RFC3123
1148 by renaming away from it. Addresses GH issue 15, reported by
1151 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
1153 TL/15 SECURITY: prevent double expansion in math comparison functions
1154 (can expand unsanitized data). Not remotely exploitable.
1161 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
1163 PP/02 Make -n do something, by making it not do something.
1164 When combined with -bP, the name of an option is not output.
1166 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
1169 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
1170 $sender_host_name and config options to manage this, and basic check
1173 PP/05 DSCP support for outbound connections and control modifier for inbound.
1175 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
1176 (Only plugin which currently uses this is kerberos4, which nobody should
1177 be using, but we should make it available and other future plugins might
1178 conceivably use it, even though it would break NAT; stuff *should* be
1179 using channel bindings instead).
1181 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
1182 name; added for Sendmail compatibility; requires admin caller.
1183 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
1184 just ignore it); requires trusted caller.
1185 Also parse but ignore: -Ac -Am -X<logfile>
1188 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
1190 TL/02 Add +smtp_confirmation as a default logging option.
1192 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
1193 Patch by Magnus Holmgren from 2007-02-20.
1195 TL/04 Bugzilla 1281 - Spec typo.
1196 Bugzilla 1283 - Spec typo.
1197 Bugzilla 1290 - Spec grammar fixes.
1199 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
1201 TL/06 Add Experimental DMARC support using libopendmarc libraries.
1203 TL/07 Fix an out of order global option causing a segfault. Reported to dev
1204 mailing list by by Dmitry Isaikin.
1206 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
1208 JH/02 Support "G" suffix to numbers in ${if comparisons.
1210 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
1212 NM/01 Bugzilla 1197 - Spec typo
1213 Bugzilla 1196 - Spec examples corrections
1215 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
1217 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
1218 gnutls_enable_pkcs11, but renamed to more accurately indicate its
1221 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
1222 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
1224 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
1225 "acl {{name}{arg}...}", and optional args on acl condition
1228 JH/05 Permit multiple router/transport headers_add/remove lines.
1230 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
1232 JH/07 Avoid using a waiting database for a single-message-only transport.
1233 Performance patch from Paul Fisher. Bugzilla 1262.
1235 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
1238 JH/09 Add $headers_added variable, with content from use of ACL modifier
1239 add_header (but not yet added to the message). Bugzilla 199.
1241 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
1242 Pulled from Bugzilla 817 by Wolfgang Breyha.
1244 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
1246 (nb: this is the same fix as in Exim 4.80.1)
1248 JH/11 Add A= logging on delivery lines, and a client_set_id option on
1251 JH/12 Add optional authenticated_sender logging to A= and a log_selector
1254 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
1256 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
1257 advertise SMTP AUTH mechanism to us, instead of a generic
1258 protocol violation error. Also, make Exim more robust to bad
1259 data from the Dovecot auth socket.
1261 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
1263 When a queue runner is handling a message, Exim first routes the
1264 recipient addresses, during which it prunes them based on the retry
1265 hints database. After that it attempts to deliver the message to
1266 any remaining recipients. It then updates the hints database using
1269 So if a recipient address works intermittently, it can get repeatedly
1270 deferred at routing time. The retry hints record remains fresh so the
1271 address never reaches the final cutoff time.
1273 This is a fairly common occurrence when a user is bumping up against
1274 their storage quota. Exim had some logic in its local delivery code
1275 to deal with this. However it did not apply to per-recipient defers
1276 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
1278 This change adds a proper retry rule check during routing so that the
1279 final cutoff time is checked against the message's age. We only do
1280 this check if there is an address retry record and there is not a
1281 domain retry record; this implies that previous attempts to handle
1282 the address had the retry_use_local_parts option turned on. We use
1283 this as an approximation for the destination being like a local
1284 delivery, as in LMTP.
1286 I suspect this new check makes the old local delivery cutoff check
1287 redundant, but I have not verified this so I left the code in place.
1289 TF/02 Correct gecos expansion when From: is a prefix of the username.
1291 Test 0254 submits a message to Exim with the header
1295 When I ran the test suite under the user fanf2, Exim expanded
1296 the header to contain my full name, whereas it should have added
1297 a Resent-Sender: header. It erroneously treats any prefix of the
1298 username as equal to the username.
1300 This change corrects that bug.
1302 GF/01 DCC debug and logging tidyup
1303 Error conditions log to paniclog rather than rejectlog.
1304 Debug lines prefixed by "DCC: " to remove any ambiguity.
1306 TF/03 Avoid unnecessary rebuilds of lookup-related code.
1308 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
1309 Bug spotted by Jeremy Harris; was flawed since initial commit.
1310 Would have resulted in OCSP responses post-SNI triggering an Exim
1311 NULL dereference and crash.
1313 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
1315 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
1316 Bug detection, analysis and fix by Samuel Thibault.
1317 Bugzilla 1331, Debian bug #698092.
1319 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
1321 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
1322 Server implementation by Todd Lyons, client by JH.
1323 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
1324 config variable "prdr_enable" controls whether the server
1325 advertises the facility. If the client requests PRDR a new
1326 acl_data_smtp_prdr ACL is called once for each recipient, after
1327 the body content is received and before the acl_smtp_data ACL.
1328 The client is controlled by both of: a hosts_try_prdr option
1329 on the smtp transport, and the server advertisement.
1330 Default client logging of deliveries and rejections involving
1331 PRDR are flagged with the string "PRDR".
1333 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
1334 fclose(). Diagnosis by Todd Lyons.
1336 PP/17 Update configure.default to handle IPv6 localhost better.
1337 Patch by Alain Williams (plus minor tweaks).
1340 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
1341 This is now consistent with GnuTLS, and is now documented: the
1342 previous undocumented portable approach to treating the option as
1343 unset was to force an expansion failure. That still works, and
1344 an empty string is now equivalent.
1346 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
1347 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
1348 not performing validation itself.
1350 PP/20 Added force_command boolean option to pipe transport.
1351 Patch from Nick Koston, of cPanel Inc.
1353 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
1356 TF/04 Added udpsend ACL modifier and hexquote expansion operator
1358 PP/21 Fix eximon continuous updating with timestamped log-files.
1359 Broken in a format-string cleanup in 4.80, missed when I repaired the
1360 other false fix of the same issue.
1361 Report and fix from Heiko Schlichting.
1364 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
1365 Report from Prashanth Katuri.
1367 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
1368 It's SecureTransport, so affects any MacOS clients which use the
1369 system-integrated TLS libraries, including email clients.
1371 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
1372 using a MIME ACL for non-SMTP local injection.
1373 Report and assistance in diagnosis by Warren Baker.
1375 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
1377 JH/16 Fix comparisons for 64b. Bugzilla 1385.
1379 TL/09 Add expansion variable $authenticated_fail_id to keep track of
1380 last id that failed so it may be referenced in subsequent ACL's.
1382 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
1385 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
1386 ldap library initialization, allowing self-signed CA's to be
1387 used. Also properly sets require_cert option later in code by
1388 using NULL (global ldap config) instead of ldap handle (per
1389 session). Bug diagnosis and testing by alxgomz.
1391 TL/12 Enhanced documentation in the ratelimit.pl script provided in
1392 the src/util/ subdirectory.
1394 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
1395 renamed to Transport Post Delivery Action by Jeremy Harris, as
1398 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
1399 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
1400 redis_servers = needs to be configured which will be used by the redis
1401 lookup. Patch from Warren Baker, of The Packet Hub.
1403 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
1405 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
1406 hostname or reverse DNS when processing a host list. Used suggestions
1407 from multiple comments on this bug.
1409 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
1411 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
1412 Missed a few lines, added it to make the runtest require no keyboard
1415 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
1416 contains upper case chars. Make router use caseful_local_part.
1418 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
1419 support when GnuTLS has been built with p11-kit.
1425 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
1427 This, or similar/improved, will also be change PP/11 of 4.82.
1433 PP/01 Handle short writes when writing local log-files.
1434 In practice, only affects FreeBSD (8 onwards).
1435 Bugzilla 1053, with thanks to Dmitry Isaikin.
1437 NM/01 Bugzilla 949 - Documentation tweak
1439 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
1442 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
1444 PP/02 Implemented gsasl authenticator.
1446 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
1448 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
1449 `pkg-config foo` for cflags/libs.
1451 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
1452 with rest of GSASL and with heimdal_gssapi.
1454 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
1455 `pkg-config foo` for cflags/libs for the TLS implementation.
1457 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
1458 properties get this fed in as external SSF. A number of robustness
1459 and debugging improvements to the cyrus_sasl authenticator.
1461 PP/08 cyrus_sasl server now expands the server_realm option.
1463 PP/09 Bugzilla 1214 - Log authentication information in reject log.
1464 Patch by Jeremy Harris.
1466 PP/10 Added dbmjz lookup type.
1468 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
1470 PP/12 MAIL args handles TAB as well as SP, for better interop with
1471 non-compliant senders.
1472 Analysis and variant patch by Todd Lyons.
1474 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
1475 Bug report from Lars Müller <lars@samba.org> (via SUSE),
1476 Patch from Dirk Mueller <dmueller@suse.com>
1478 PP/13 tls_peerdn now print-escaped for spool files.
1479 Observed some $tls_peerdn in wild which contained \n, which resulted
1480 in spool file corruption.
1482 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
1483 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
1484 or write after TLS renegotiation, which otherwise led to messages
1487 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
1488 as a tracking header (ie: a signed header comes before the signature).
1489 Patch from Wolfgang Breyha.
1491 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
1492 comma-sep list; embedded commas doubled.
1494 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
1496 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
1498 Report and patch from Dmitry Banschikov.
1500 PP/16 Removed "dont_insert_empty_fragments" from "openssl_options".
1501 Removed SSL_clear() after SSL_new() which led to protocol negotiation
1502 failures. We appear to now support TLS1.1+ with Exim.
1504 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
1505 lets Exim select keys and certificates based upon TLS SNI from client.
1506 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
1507 before an outbound SMTP session. New log_selector, +tls_sni.
1509 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
1510 NULL dereference. Report and patch from Alun Jones.
1512 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
1513 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
1514 Not seeing resolver debug output on NetBSD, but suspect this is a
1515 resolver implementation change.
1517 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
1518 Left warnings. Added "eximon gdb" invocation mode.
1520 PP/21 Defaulting "accept_8bitmime" to true, not false.
1522 PP/22 Added -bw for inetd wait mode support.
1524 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
1525 locate the relevant includes and libraries. Made this the default.
1527 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
1528 Bugzilla 1246, report and most of solution from Tomasz Kusy.
1530 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
1531 This may cause build issues on older platforms.
1533 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
1534 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
1535 gnutls_require_mac & gnutls_require_protocols (no longer supported).
1536 Added SNI support via GnuTLS too.
1537 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
1539 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
1541 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
1542 Applied second patch from Janne, implementing suggestion to default
1543 multiple-strings-in-record handling to match SPF spec.
1545 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
1547 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
1548 read-only, out of scope).
1549 Patch from Wolfgang Breyha, report from Stuart Northfield.
1551 PP/29 Fix three issues highlighted by clang analyser static analysis.
1552 Only crash-plausible issue would require the Cambridge-specific
1553 iplookup router and a misconfiguration.
1554 Report from Marcin Mirosław.
1556 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
1558 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
1559 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
1560 As part of this, removing so much warning spew let me fix some minor
1561 real issues in debug logging.
1563 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
1564 assignment on my part. Fixed.
1566 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
1567 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
1568 Janne Snabb (who went above and beyond: thank you).
1570 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
1571 string otherwise requires a connection and a bunch more work and it's
1572 relatively easy to get wrong. Should also expose TLS library linkage
1575 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
1576 64-bit ${eval} (JH/03).
1578 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
1579 GNU libc to support some of the 64-bit stuff, should not lead to
1580 conflicts. Defined before os.h is pulled in, so if a given platform
1581 needs to override this, it can.
1583 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
1584 protection layer was required, which is not implemented.
1585 Bugzilla 1254, patch from Wolfgang Breyha.
1587 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
1588 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
1589 tls_dhparam take prime identifiers. Also unbreak combination of
1590 OpenSSL+DH_params+TLSSNI.
1592 PP/39 Disable SSLv2 by default in OpenSSL support.
1598 PP/01 Solaris build fix for Oracle's LDAP libraries.
1599 Bugzilla 1109, patch from Stephen Usher.
1601 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
1603 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
1606 TF/02 Fix a couple more cases where we did not log the error message
1607 when unlink() failed. See also change 4.74-TF/03.
1609 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
1610 lock up or crash if it happened to be inside a call to libc when it
1611 got a SIGUSR1 from exiwhat.
1613 The SIGUSR1 handler appends the current process status to the process
1614 log which is later printed by exiwhat. It used to use the general
1615 purpose logging code to do this, but several functions it calls are
1616 not safe for signals.
1618 The new output code in the SIGUSR1 handler is specific to the process
1619 log, and simple enough that it's easy to inspect for signal safety.
1620 Removing some special cases also simplifies the general logging code.
1621 Removing the spurious timestamps from the process log simplifies
1624 TF/04 Improved ratelimit ACL condition.
1626 The /noupdate option has been deprecated in favour of /readonly which
1627 has clearer semantics. The /leaky, /strict, and /readonly update modes
1628 are mutually exclusive. The update mode is no longer included in the
1629 database key; it just determines when the database is updated. (This
1630 means that when you upgrade Exim will forget old rate measurements.)
1632 Exim now checks that the per_* options are used with an update mode that
1633 makes sense for the current ACL. For example, when Exim is processing a
1634 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
1635 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
1636 must specify per_mail/readonly. If you omit the update mode it defaults to
1637 /leaky where that makes sense (as before) or /readonly where required.
1639 The /noupdate option is now undocumented but still supported for
1640 backwards compatibility. It is equivalent to /readonly except that in
1641 ACLs where /readonly is required you may specify /leaky/noupdate or
1642 /strict/noupdate which are treated the same as /readonly.
1644 A useful new feature is the /count= option. This is a generalization
1645 of the per_byte option, so that you can measure the throughput of other
1646 aggregate values. For example, the per_byte option is now equivalent
1647 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
1649 The per_rcpt option has been generalized using the /count= mechanism
1650 (though it's more complicated than the per_byte equivalence). When it is
1651 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
1652 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
1653 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
1654 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
1655 Note that using per_rcpt with a non-readonly update mode in more than
1656 one ACL will cause the recipients to be double-counted. (The per_mail
1657 and per_byte options don't have this problem.)
1659 The handling of very low rates has changed slightly. If the computed rate
1660 is less than the event's count (usually one) then this event is the first
1661 after a long gap. In this case the rate is set to the same as this event's
1662 count, so that the first message of a spam run is counted properly.
1664 The major new feature is a mechanism for counting the rate of unique
1665 events. The new per_addr option counts the number of different
1666 recipients that someone has sent messages to in the last time period. It
1667 behaves like per_rcpt if all the recipient addresses are different, but
1668 duplicate recipient addresses do not increase the measured rate. Like
1669 the /count= option this is a general mechanism, so the per_addr option
1670 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
1671 example, measure the rate that a client uses different sender addresses
1672 with the options per_mail/unique=$sender_address. There are further
1673 details in the main documentation.
1675 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
1677 TF/06 Removed a few PCRE remnants.
1679 TF/07 Automatically extract Exim's version number from tags in the git
1680 repository when doing development or release builds.
1682 PP/02 Raise smtp_cmd_buffer_size to 16kB.
1683 Bugzilla 879. Patch from Paul Fisher.
1685 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
1686 Heavily based on revision 40f9a89a from Simon Arlott's tree.
1689 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
1691 PP/05 Variable $av_failed, true if the AV scanner deferred.
1692 Bugzilla 1078. Patch from John Horne.
1694 PP/06 Stop make process more reliably on build failure.
1695 Bugzilla 1087. Patch from Heiko Schlittermann.
1697 PP/07 Make maildir_use_size_file an _expandable_ boolean.
1698 Bugzilla 1089. Patch from Heiko Schlittermann.
1700 PP/08 Handle ${run} returning more data than OS pipe buffer size.
1701 Bugzilla 1131. Patch from Holger Weiß.
1703 PP/09 Handle IPv6 addresses with SPF.
1704 Bugzilla 860. Patch from Wolfgang Breyha.
1706 PP/10 GnuTLS: support TLS 1.2 & 1.1.
1708 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
1711 PP/11 match_* no longer expand right-hand-side by default.
1712 New compile-time build option, EXPAND_LISTMATCH_RHS.
1713 New expansion conditions, "inlist", "inlisti".
1715 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
1717 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
1719 PP/14 fix log_write() format string regression from TF/03.
1720 Bugzilla 1152. Patch from Dmitry Isaikin.
1726 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
1728 PP/02 Harmonised TLS library version reporting; only show if debugging.
1729 Layout now matches that introduced for other libraries in 4.74 PP/03.
1731 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
1733 PP/04 New "dns_use_edns0" global option.
1735 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
1738 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
1739 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
1741 TK/01 Updated PolarSSL code to 0.14.2.
1742 Bugzilla 1097. Patch from Andreas Metzler.
1744 PP/07 Catch divide-by-zero in ${eval:...}.
1745 Fixes bugzilla 1102.
1747 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
1750 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
1751 format-string attack -- SECURITY: remote arbitrary code execution.
1753 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
1754 time unintentionally subject to list matching rules, letting the header
1755 cause arbitrary Exim lookups (of items which can occur in lists, *not*
1756 arbitrary string expansion). This allowed for information disclosure.
1758 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
1759 INT_MIN/-1 -- value coerced to INT_MAX.
1765 NM/01 Workaround for PCRE version dependency in version reporting
1768 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
1769 This fixes portability to compilers other than gcc, notably
1770 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
1772 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
1773 makefiles for portability to HP-UX and POSIX correctness.
1775 PP/01 Permit LOOKUP_foo enabling on the make command-line.
1776 Also via indented variable definition in the Makefile.
1777 (Debugging by Oliver Heesakkers).
1779 PP/02 Restore caching of spamd results with expanded spamd_address.
1780 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
1782 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
1783 Improves build reliability. Fix from: Frank Elsner
1785 NM/02 Fix wide character breakage in the rfc2047 coding
1786 Fixes bug 1064. Patch from Andrey N. Oktyabrski
1788 NM/03 Allow underscore in dnslist lookups
1789 Fixes bug 1026. Patch from Graeme Fowler
1791 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
1792 Code patches from Adam Ciarcinski of NetBSD.
1794 NM/04 Fixed exiqgrep to cope with mailq missing size issue
1797 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
1798 is logged, to avoid truncation. Patch from John Horne.
1800 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
1801 Patch from Jakob Hirsch.
1803 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
1804 SQL string expansion failure details.
1805 Patch from Andrey Oktyabrski.
1807 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
1808 Patch from Simon Arlott.
1810 PP/09 New lookups functionality failed to compile on old gcc which rejects
1811 extern declarations in function scope.
1812 Patch from Oliver Fleischmann
1814 PP/10 Use sig_atomic_t for flags set from signal handlers.
1815 Check getgroups() return and improve debugging.
1816 Fixed developed for diagnosis in bug 927 (which turned out to be
1819 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
1820 Patch from Mark Zealey.
1822 PP/12 Bugzilla 1056: Improved spamd server selection.
1823 Patch from Mark Zealey.
1825 PP/13 Bugzilla 1086: Deal with maildir quota file races.
1826 Based on patch from Heiko Schlittermann.
1828 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
1829 Patch from Uwe Doering, sign-off by Michael Haardt.
1831 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
1832 variable declaration deep within a block. Bug and patch from
1835 PP/15 lookups-Makefile IRIX compatibility coercion.
1837 PP/16 Make DISABLE_DKIM build knob functional.
1839 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
1840 Patch by Simon Arlott
1842 TF/03 Fix valgrind.h portability to C89 compilers that do not support
1843 variable argument macros. Our copy now differs from upstream.
1849 TF/01 Failure to get a lock on a hints database can have serious
1850 consequences so log it to the panic log.
1852 TF/02 Log LMTP confirmation messages in the same way as SMTP,
1853 controlled using the smtp_confirmation log selector.
1855 TF/03 Include the error message when we fail to unlink a spool file.
1857 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
1858 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
1859 for maintaining out-of-tree patches for some time.
1861 PP/01 Bugzilla 139: Documentation and portability issues.
1862 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
1863 Handle per-OS dynamic-module compilation flags.
1865 PP/02 Let /dev/null have normal permissions.
1866 The 4.73 fixes were a little too stringent and complained about the
1867 permissions on /dev/null. Exempt it from some checks.
1868 Reported by Andreas M. Kirchwitz.
1870 PP/03 Report version information for many libraries, including
1871 Exim version information for dynamically loaded libraries. Created
1872 version.h, now support a version extension string for distributors
1873 who patch heavily. Dynamic module ABI change.
1875 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
1876 privilege escalation vulnerability whereby the Exim run-time user
1877 can cause root to append content of the attacker's choosing to
1880 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
1883 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
1884 If dropping privileges for untrusted macros, we disabled normal logging
1885 on the basis that it would fail; for the Exim run-time user, this is not
1886 the case, and it resulted in successful deliveries going unlogged.
1887 Fixed. Reported by Andreas Metzler.
1893 PP/01 Date: & Message-Id: revert to normally being appended to a message,
1894 only prepend for the Resent-* case. Fixes regression introduced in
1895 Exim 4.70 by NM/22 for Bugzilla 607.
1897 PP/02 Include check_rfc2047_length in configure.default because we're seeing
1898 increasing numbers of administrators be bitten by this.
1900 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
1902 PP/03 Bugzilla 994: added openssl_options main configuration option.
1904 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
1906 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
1908 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
1910 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
1911 without a peer certificate, leading to a segfault because of an
1912 assumption that peers always have certificates. Be a little more
1913 paranoid. Problem reported by Martin Tscholak.
1915 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
1916 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
1917 NB: ClamAV planning to remove STREAM in "middle of 2010".
1918 CL also introduces -bmalware, various -d+acl logging additions and
1919 more caution in buffer sizes.
1921 PP/09 Implemented reverse_ip expansion operator.
1923 PP/10 Bugzilla 937: provide a "debug" ACL control.
1925 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
1927 PP/12 Bugzilla 973: Implement --version.
1929 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
1931 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
1933 PP/15 Bugzilla 816: support multiple condition rules on Routers.
1935 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
1936 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
1937 ignore trailing whitespace.
1939 JJ/02 prevent non-panic DKIM error from being sent to paniclog
1941 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
1944 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
1945 Notification from Dr Andrew Aitchison.
1947 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
1948 ExtendedDetectionInfo response format.
1949 Notification from John Horne.
1951 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
1954 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
1955 XSL and documented dependency on system catalogs, with examples of how
1958 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
1961 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
1962 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
1963 configuration file which is writeable by the Exim user or group.
1965 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
1966 of configuration files to cover files specified with the -C option if
1967 they are going to be used with root privileges, not just the default
1970 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
1971 option (effectively making it always true).
1973 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
1974 files to be used while preserving root privileges.
1976 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
1977 that rogue child processes cannot use them.
1979 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
1980 run-time user, instead of root.
1982 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
1983 Exim run-time user without dropping privileges.
1985 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
1986 result string, instead of calling string_vformat() twice with the same
1989 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
1990 for other users. Others should always drop root privileges if they use
1991 -C on the command line, even for a whitelisted configure file.
1993 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
1995 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
2001 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
2002 $data_path, and $header_path variables; fixed documentation bugs and
2005 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
2006 exipick to access non-standard spools, including the "frozen" queue
2009 NM/01 Bugzilla 965: Support mysql stored procedures.
2010 Patch from Alain Williams
2012 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
2014 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
2015 Patch from Andreas Metzler
2017 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
2018 Patch from Kirill Miazine
2020 NM/05 Bugzilla 671: Added umask to procmail example.
2022 JJ/03 installed exipick 20100323.0, fixing doc bug
2024 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
2025 directory. Notification and patch from Dan Rosenberg.
2027 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
2029 TK/02 Improve log output when DKIM signing operation fails.
2031 MH/01 Treat the transport option dkim_domain as a colon separated
2032 list, not as a single string, and sign the message with each element,
2033 omitting multiple occurences of the same signer.
2035 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
2036 Bugzilla 985, 986. Patch by Simon Arlott
2038 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
2039 Patch by Simon Arlott
2041 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
2042 MBX locking. Notification from Dan Rosenberg.
2048 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
2050 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
2052 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
2054 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
2056 NM/04 Bugzilla 915: Flag broken perl installation during build.
2062 TK/01 Added patch by Johannes Berg that expands the main option
2063 "spamd_address" if it starts with a dollar sign.
2065 TK/02 Write list of recipients to X-Envelope-Sender header when building
2066 the mbox-format spool file for content scanning (suggested by Jakob
2069 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
2070 (http://www.dcc-servers.net/) support via dccifd. Activated by
2071 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
2073 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
2074 by Mark Daniel Reidel <mr@df.eu>.
2076 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
2077 When building exim an external PCRE library is now needed -
2078 PCRE is a system library on the majority of modern systems.
2079 See entry on PCRE_LIBS in EDITME file.
2081 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
2082 conversation. Added nologin parameter to request.
2083 Patch contributed by Kirill Miazine.
2085 TF/01 Do not log submission mode rewrites if they do not change the address.
2087 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
2089 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
2090 log files in place. Contributed by Roberto Lima.
2092 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
2094 TF/03 Bugzilla 615: When checking the local_parts router precondition
2095 after a local_part_suffix or local_part_prefix option, Exim now
2096 does not use the address's named list lookup cache, since this
2097 contains cached lookups for the whole local part.
2099 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
2100 Robert Millan. Documentation is in experimental-spec.txt.
2102 TF/04 Bugzilla 668: Fix parallel build (make -j).
2104 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
2106 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
2107 Patch provided by Jan Srzednicki.
2109 TF/05 Leading white space used to be stripped from $spam_report which
2110 wrecked the formatting. Now it is preserved.
2112 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
2113 that they are available at delivery time.
2115 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
2117 TF/08 TLS error reporting now respects the incoming_interface and
2118 incoming_port log selectors.
2120 TF/09 Produce a more useful error message if an SMTP transport's hosts
2121 setting expands to an empty string.
2123 NM/06 Bugzilla 744: EXPN did not work under TLS.
2124 Patch provided by Phil Pennock.
2126 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
2127 Patch provided by Richard Godbee.
2129 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
2130 acl_smtp_notquit, added index entry.
2132 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
2133 Patch provided by Eugene Bujak.
2135 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
2136 accept(). Patch provided by Maxim Dounin.
2138 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
2139 Patch provided by Phil Pennock.
2141 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
2143 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
2144 Patch provided by Brad "anomie" Jorsch.
2146 NM/14 Bugzilla 622: Added timeout setting to transport filter.
2147 Patch provided by Dean Brooks.
2149 TK/05 Add native DKIM support (does not depend on external libraries).
2151 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
2152 Patch provided by Graeme Fowler.
2154 NM/16 Bugzilla 851: Documentation example syntax fix.
2156 NM/17 Changed NOTICE file to remove references to embedded PCRE.
2158 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
2161 NM/19 Bugzilla 745: TLS version reporting.
2162 Patch provided by Phil Pennock.
2164 NM/20 Bugzilla 167: bool: condition support.
2165 Patch provided by Phil Pennock.
2167 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
2168 clients. Patch provided by Phil Pennock.
2170 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
2171 Patch provided by Brad "anomie" Jorsch.
2173 NM/23 Bugzilla 687: Fix misparses in eximstats.
2174 Patch provided by Heiko Schlittermann.
2176 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
2177 Patch provided by Heiko Schlittermann.
2179 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
2180 plus update to original patch.
2182 NM/26 Bugzilla 799: Documentation correction for ratelimit.
2184 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
2185 Patch provided by David Brownlee.
2187 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
2189 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
2191 NM/30 Bugzilla 888: TLS documentation bugfixes.
2193 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
2195 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
2196 Unlike the original bugzilla I have changed all shell scripts in src tree.
2198 NM/33 Bugzilla 898: Transport filter timeout fix.
2199 Patch by Todd Rinaldo.
2201 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
2202 Patch by Serge Demonchaux.
2204 NM/35 Bugzilla 39: Base64 decode bug fixes.
2205 Patch by Jakob Hirsch.
2207 NM/36 Bugzilla 909: Correct connect() call in dcc code.
2209 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
2211 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
2213 NM/39 Bugzilla 911: Fixed MakeLinks build script.
2219 TK/01 Add preliminary DKIM support. Currently requires a forked version of
2220 ALT-N's libdkim that I have put here:
2221 http://duncanthrax.net/exim-experimental/
2223 Note to Michael Haardt: I had to rename some vars in sieve.c. They
2224 were called 'true' and it seems that C99 defines that as a reserved
2225 keyword to be used with 'bool' variable types. That means you could
2226 not include C99-style headers which use bools without triggering
2227 build errors in sieve.c.
2229 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
2230 as mailq or other aliases. Changed the --help handling significantly
2231 to do whats expected. exim_usage() emits usage/help information.
2233 SC/01 Added the -bylocaldomain option to eximstats.
2235 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
2237 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
2239 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
2245 PH/01 Another patch from the Sieve maintainer.
2247 PH/02 When an IPv6 address is converted to a string for single-key lookup
2248 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
2249 dots are used instead of colons so that keys in lsearch files need not
2250 contain colons. This was done some time before quoting was made available
2251 in lsearch files. However, iplsearch files do require colons in IPv6 keys
2252 (notated using the quote facility) so as to distinguish them from IPv4
2253 keys. This meant that lookups for IP addresses in host lists did not work
2254 for iplsearch lookups.
2256 This has been fixed by arranging for IPv6 addresses to be expressed with
2257 colons if the lookup type is iplsearch. This is not incompatible, because
2258 previously such lookups could never work.
2260 The situation is now rather anomalous, since one *can* have colons in
2261 ordinary lsearch keys. However, making the change in all cases is
2262 incompatible and would probably break a number of configurations.
2264 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
2267 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
2268 conversion specification without a maximum field width, thereby enabling
2269 a rogue spamd server to cause a buffer overflow. While nobody in their
2270 right mind would setup Exim to query an untrusted spamd server, an
2271 attacker that gains access to a server running spamd could potentially
2272 exploit this vulnerability to run arbitrary code as the Exim user.
2274 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
2275 $primary_hostname instead of what libspf2 thinks the hosts name is.
2277 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
2278 a directory entry by the name of the lookup key. Previously, if a
2279 symlink pointed to a non-existing file or a file in a directory that
2280 Exim lacked permissions to read, a lookup for a key matching that
2281 symlink would fail. Now it is enough that a matching directory entry
2282 exists, symlink or not. (Bugzilla 503.)
2284 PH/03 The body_linecount and body_zerocount variables are now exported in the
2287 PH/04 Added the $dnslist_matched variable.
2289 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
2290 This means they are set thereafter only if the connection becomes
2293 PH/06 Added the client_condition to authenticators so that some can be skipped
2294 by clients under certain conditions.
2296 PH/07 The error message for a badly-placed control=no_multiline_responses left
2297 "_responses" off the end of the name.
2299 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
2301 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
2302 (without spaces) instead of just copying the configuration text.
2304 PH/10 Added the /noupdate option to the ratelimit ACL condition.
2306 PH/11 Added $max_received_linelength.
2308 PH/12 Added +ignore_defer and +include_defer to host lists.
2310 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
2311 way in which PCRE > 7.0 is built.
2313 PH/14 Implemented queue_only_load_latch.
2315 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
2316 MAIL command. The effect was to mangle the value on 64-bit systems.
2318 PH/16 Another patch from the Sieve maintainer.
2320 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
2322 PH/18 If a system quota error occurred while trying to create the file for
2323 a maildir delivery, the message "Mailbox is full" was not appended to the
2324 bounce if the delivery eventually timed out. Change 4.67/27 below applied
2325 only to a quota excession during the actual writing of the file.
2327 PH/19 It seems that peer DN values may contain newlines (and other non-printing
2328 characters?) which causes problems in log lines. The DN values are now
2329 passed through string_printing() before being added to log lines.
2331 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
2332 and InterBase are left for another time.)
2334 PH/21 Added message_body_newlines option.
2336 PH/22 Guard against possible overflow in moan_check_errorcopy().
2338 PH/23 POSIX allows open() to be a macro; guard against that.
2340 PH/24 If the recipient of an error message contained an @ in the local part
2341 (suitably quoted, of course), incorrect values were put in $domain and
2342 $local_part during the evaluation of errors_copy.
2348 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
2349 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
2352 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
2353 issue a MAIL command.
2355 PH/02 In an ACL statement such as
2357 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
2359 if a client was not listed at all, or was listed with a value other than
2360 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
2361 the condition was not true (as it should be), so access was not denied.
2362 The bug was that the ! inversion was incorrectly passed on to the second
2363 item. This has been fixed.
2365 PH/03 Added additional dnslists conditions == and =& which are different from
2366 = and & when the dns lookup returns more than one IP address.
2368 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
2369 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
2371 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
2372 FSYNC, which compiles an option called disable_fsync that allows for
2373 bypassing fsync(). The documentation is heavily laced with warnings.
2375 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
2377 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
2378 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
2379 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
2380 including adding "make clean"; (3) Added -fPIC when compiling the test
2381 dynamically loaded module, to get rid of a warning.
2383 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
2384 message fails, move_frozen_messages = true and ignore_bounce_errors_after
2385 = 0s. The bug is otherwise harmless.
2387 PH/07 There was a bug in the dovecot authenticator such that the value of
2388 $auth1 could be overwritten, and so not correctly preserved, after a
2389 successful authentication. This usually meant that the value preserved by
2390 the server_setid option was incorrect.
2392 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
2394 PH/09 Installed PCRE release 7.0.
2396 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
2397 run for batched SMTP input. It is now run at the start of every message
2398 in the batch. While fixing this I discovered that the process information
2399 (output by running exiwhat) was not always getting set for -bs and -bS
2400 input. This is fixed, and it now also says "batched" for BSMTP.
2402 PH/11 Added control=no_pipelining.
2404 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
2405 patch, slightly modified), and move the expansion of helo_data till after
2406 the connection is made in the smtp transport (so it can use these
2409 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
2411 PH/14 Added log_selector = +pid.
2413 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
2415 PH/16 Add ${if forany and ${if forall.
2417 PH/17 Added dsn_from option to vary the From: line in DSNs.
2419 PH/18 Flush SMTP output before performing a callout, unless control =
2420 no_callout_flush is set.
2422 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
2423 was true (the default) a successful delivery failed to delete the retry
2424 item, thus causing premature timeout of the address. The bug is now
2427 PH/20 Added hosts_avoid_pipelining to the smtp transport.
2429 PH/21 Long custom messages for fakedefer and fakereject are now split up
2430 into multiline responses in the same way that messages for "deny" and
2431 other ACL rejections are.
2433 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
2434 with slight modification.
2436 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
2437 draft, changing the syntax and factoring some duplicate code".
2439 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
2440 for deliveries of the second and subsequent messages over the same SMTP
2443 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
2444 ${reduce, with only minor "tidies".
2446 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
2448 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
2449 expansion side effects.
2451 PH/27 When a message times out after an over-quota error from an Exim-imposed
2452 quota, the bounce message says "mailbox is full". This message was not
2453 being given when it was a system quota that was exceeded. It now should
2456 MH/03 Made $recipients available in local_scan(). local_scan() already has
2457 better access to the recipient list through recipients_list[], but
2458 $recipients can be useful in postmaster-provided expansion strings.
2460 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
2461 in the case of a MAIL command with additional options following the
2462 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
2463 were accidentally chopped off.
2465 PH/29 SMTP synchronization checks are implemented when a command is read -
2466 there is a check that no more input is waiting when there shouldn't be
2467 any. However, for some commands, a delay in an ACL can mean that it is
2468 some time before the response is written. In this time, more input might
2469 arrive, invalidly. So now there are extra checks after an ACL has run for
2470 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
2471 pipelining has not been advertised.
2473 PH/30 MH's patch to allow iscntrl() characters to be list separators.
2475 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
2476 returned in the SMTP response when smtp_return_error_details was false.
2477 This has been fixed.
2479 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
2480 instead of the C I/O that was originally supplied, because problems were
2481 reported on Solaris.
2483 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
2484 Exim which did not show up earlier: it was assuming that a call to
2485 SSL_CTX_set_info_callback() might give an error value. In fact, there is
2486 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
2487 was a macro that became an assignment, so it seemed to work. This has
2488 changed to a proper function call with a void return, hence the compile
2489 error. Exim's code has been fixed.
2491 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
2494 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
2496 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
2498 PH/37 If a message is not accepted after it has had an id assigned (e.g.
2499 because it turns out to be too big or there is a timeout) there is no
2500 "Completed" line in the log. When some messages of this type were
2501 selected by exigrep, they were listed as "not completed". Others were
2502 picked up by some special patterns. I have improved the selection
2503 criteria to be more general.
2505 PH/38 The host_find_failed option in the manualroute router can now be set
2506 to "ignore", to completely ignore a host whose IP address cannot be
2507 found. If all hosts are ignored, the behaviour is controlled by the new
2508 host_all_ignored option.
2510 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
2511 homing or because of multiple MX records with /mx) generated more than
2512 one IP address, and the following item turned out to be the local host,
2513 all the secondary addresses of the first item were incorrectly removed
2514 from the list, along with the local host and any following hosts (which
2515 is what is supposed to happen).
2517 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
2518 whoever called Exim into the -H file. In the case of the daemon it was
2519 behaving confusingly. When first started, it used values for whoever
2520 started the daemon, but after a SIGHUP it used the Exim user (because it
2521 calls itself on a restart). I have changed the code so that it now always
2524 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
2525 message are rejected with the same error (e.g. no authentication or bad
2526 sender address), and a DATA command is nevertheless sent (as can happen
2527 with PIPELINING or a stupid MUA), the error message that was given to the
2528 RCPT commands is included in the rejection of the DATA command. This is
2529 intended to be helpful for MUAs that show only the final error to their
2532 PH/42 Another patch from the Sieve maintainer.
2534 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
2535 Eximstats - Fixed some broken HTML links and added missing column headers
2537 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
2538 columns for Rejects, Temp Rejects, Ham, and Spam rows.
2540 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
2542 PH/43 Yet another patch from the Sieve maintainer.
2544 PH/44 I found a way to check for a TCP/IP connection going away before sending
2545 the response to the final '.' that terminates a message, but only in the
2546 case where the client has not sent further data following the '.'
2547 (unfortunately, this is allowed). However, in many cases there won't be
2548 any further data because there won't be any more messages to send. A call
2549 to select() can be used: if it shows that the input is "ready", there is
2550 either input waiting, or the socket has been closed. An attempt to read
2551 the next input character can distinguish the two cases. Previously, Exim
2552 would have sent an OK response which the client would never have see.
2553 This could lead to message repetition. This fix should cure that, at
2554 least in a lot of common cases.
2556 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
2557 advertised in response to EHLO.
2563 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
2564 fixed by 4.65/MH/01 (is this a record?) are fixed:
2566 (i) An empty string was always treated as zero by the numeric comparison
2567 operators. This behaviour has been restored.
2569 (ii) It is documented that the numeric comparison operators always treat
2570 their arguments as decimal numbers. This was broken in that numbers
2571 starting with 0 were being interpreted as octal.
2573 While fixing these problems I realized that there was another issue that
2574 hadn't been noticed. Values of message_size_limit (both the global option
2575 and the transport option) were treated as octal if they started with 0.
2576 The documentation was vague. These values are now always treated as
2577 decimal, and I will make that clear in the documentation.
2583 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
2584 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
2587 MH/01 Don't check that the operands of numeric comparison operators are
2588 integers when their expansion is in "skipping" mode (fixes bug
2589 introduced by 4.64-PH/07).
2591 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
2592 child addresses, Exim now panics and dies. Previously, because the count
2593 is held in a short int, deliveries were likely to be lost. As such a
2594 large number of recipients for a single message is ridiculous
2595 (performance will be very, very poor), I have chosen to impose a limit
2596 rather than extend the field.
2602 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
2603 leftover -K file (the existence of which was triggered by #402).
2604 While we were at it, introduced process PID as part of the -K
2605 filename. This should rule out race conditions when creating
2608 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
2609 processing considerably. Previous code took too long for large mails,
2610 triggering a timeout which in turn triggers #401.
2612 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
2613 in the DK code in transports.c. sendfile() is not really portable,
2614 hence the _LINUX specificness.
2616 TF/01 In the add_headers option to the mail command in an Exim filter,
2617 there was a bug that Exim would claim a syntax error in any
2618 header after the first one which had an odd number of characters
2621 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
2622 callout verification, Exim cached a "reject" for the entire domain. This
2623 is correct for most verifications, but it is not correct for a recipient
2624 verification with use_sender or use_postmaster set, because in that case
2625 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
2626 case of MAIL FROM:<> rejection from other early rejections (e.g.
2627 rejection of HELO). When verifying a recipient using a non-null MAIL
2628 address, the cache is ignored if it shows MAIL FROM:<> rejection.
2629 Whatever the result of the callout, the value of the domain cache is
2630 left unchanged (for any other kind of callout, getting as far as trying
2631 RCPT means that the domain itself is ok).
2633 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
2636 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
2637 manifest itself as EPIPE rather than ECONNECT. When tidying away a
2638 session, the daemon ignores ECONNECT errors and logs others; it now
2639 ignores EPIPE as well.
2641 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
2642 (quoted-printable decoding).
2644 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
2645 later the small subsequent patch to fix an introduced bug.
2647 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
2649 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
2651 PH/08 An error is now given if message_size_limit is specified negative.
2653 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
2654 to be given (somewhat) arbitrary names.
2656 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
2659 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
2660 miscellaneous code fixes
2662 PH/10 Added the log_reject_target ACL modifier to specify where to log
2665 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
2666 hostname. This is wrong, because it relates to the incoming message (and
2667 probably the interface on which it is arriving) and not to the outgoing
2668 callout (which could be using a different interface). This has been
2669 changed to use the value of the helo_data option from the smtp transport
2670 instead - this is what is used when a message is actually being sent. If
2671 there is no remote transport (possible with a router that sets up host
2672 addresses), $smtp_active_hostname is used.
2674 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
2675 tweaks were necessary in order to get it to work (see also 21 below):
2676 (a) The code assumed that strncpy() returns a negative number on buffer
2677 overflow, which isn't the case. Replaced with Exim's string_format()
2679 (b) There were several signed/unsigned issues. I just did the minimum
2680 hacking in of casts. There is scope for a larger refactoring.
2681 (c) The code used strcasecmp() which is not a standard C function.
2682 Replaced with Exim's strcmpic() function.
2683 (d) The code set only $1; it now sets $auth1 as well.
2684 (e) A simple test gave the error "authentication client didn't specify
2685 service in request". It would seem that Dovecot has changed its
2686 interface. Fortunately there's a specification; I followed it and
2687 changed what the client sends and it appears to be working now.
2689 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
2692 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
2693 address A is aliased to B and C, where B exists and C does not. Without
2694 -v the output is "A verified" because verification stops after a
2695 successful redirection if more than one address is generated. However,
2696 with -v the child addresses are also verified. Exim was outputting "A
2697 failed to verify" and then showing the successful verification for C,
2698 with its parentage. It now outputs "B failed to verify", showing B's
2699 parentage before showing the successful verification of C.
2701 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
2702 look up a TXT record in a specific list after matching in a combined
2705 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
2706 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
2707 they consult the DNS. I had assumed they would set it the way they
2708 wanted; and indeed my experiments on Linux seem to show that in some
2709 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
2710 To be on the safe side, however, I have now made the interface to
2711 host_find_byname() similar to host_find_bydns(), with an argument
2712 containing the DNS resolver options. The host_find_byname() function now
2713 sets these options at its start, just as host_find_bydns() does. The smtp
2714 transport options dns_qualify_single and dns_search_parents are passed to
2715 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
2716 of host_find_byname() use the default settings of RES_DEFNAMES
2717 (qualify_single) but not RES_DNSRCH (search_parents).
2719 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
2720 spool_read_header() do less string testing, by means of a preliminary
2721 switch on the second character of optional "-foo" lines. (This is
2722 overdue, caused by the large number of possibilities that now exist.
2723 Originally there were few.) While I was there, I also converted the
2724 str(n)cmp tests so they don't re-test the leading "-" and the first
2725 character, in the hope this might squeeze out yet more improvement.
2727 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
2728 flag allowing group syntax was set by the header_syntax check but not
2729 turned off, possible causing trouble later; (2) The flag was not being
2730 set at all for the header_verify test, causing "group"-style headers to
2731 be rejected. I have now set it in this case, and also caused header_
2732 verify to ignore an empty address taken from a group. While doing this, I
2733 came across some other cases where the code for allowing group syntax
2734 while scanning a header line wasn't quite right (mostly, not resetting
2735 the flag correctly in the right place). These bugs could have caused
2736 trouble for malformed header lines. I hope it is now all correct.
2738 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
2739 with the "reply" argument non-NULL. The code, however (which originally
2740 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
2741 but it didn't always do it. This confused somebody who was copying the
2742 code for some other use. I have removed all the tests.
2744 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
2745 feature that was used to support insecure browsers during the U.S. crypto
2746 embargo. It requires special client support, and Exim is probably the
2747 only MTA that supported it -- and would never use it because real RSA is
2748 always available. This code has been removed, because it had the bad
2749 effect of slowing Exim down by computing (never used) parameters for the
2750 RSA_EXPORT functionality.
2752 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
2753 authenticator to fail if there's a tab character in the incoming data
2754 (there should never be unless someone is messing about, as it's supposed
2755 to be base64-encoded). Also added, on Timo's advice, the "secured" option
2756 if the connection is using TLS or if the remote IP is the same as the
2757 local IP, and the "valid-client-cert option" if a client certificate has
2760 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
2761 authenticators. This can be used for authorization after authentication
2762 succeeds. (In the case of plaintext, it servers for both authentication
2765 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
2766 if any retry times were supplied.
2768 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
2769 connection, where there is no client IP address to check. In this
2770 situation, the verify now always succeeds.
2772 PH/25 Applied John Jetmore's -Mset patch.
2774 PH/26 Added -bem to be like -Mset, but loading a message from a file.
2776 PH/27 In a string expansion for a processed (not raw) header when multiple
2777 headers of the same name were present, leading whitespace was being
2778 removed from all of them, but trailing whitespace was being removed only
2779 from the last one. Now trailing whitespace is removed from each header
2780 before concatenation. Completely empty headers in a concatenation (as
2781 before) are ignored.
2783 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
2784 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
2786 PH/29 [Removed. This was a change that I later backed out, and forgot to
2787 correct the ChangeLog entry (that I had efficiently created) before
2788 committing the later change.]
2790 PH/30 Exim was sometimes attempting to deliver messages that had suffered
2791 address errors (4xx response to RCPT) over the same connection as other
2792 messages routed to the same hosts. Such deliveries are always "forced",
2793 so retry times are not inspected. This resulted in far too many retries
2794 for the affected addresses. The effect occurred only when there were more
2795 hosts than the hosts_max_try setting in the smtp transport when it had
2796 the 4xx errors. Those hosts that it had tried were not added to the list
2797 of hosts for which the message was waiting, so if all were tried, there
2798 was no problem. Two fixes have been applied:
2800 (i) If there are any address or message errors in an SMTP delivery, none
2801 of the hosts (tried or untried) are now added to the list of hosts
2802 for which the message is waiting, so the message should not be a
2803 candidate for sending over the same connection that was used for a
2804 successful delivery of some other message. This seems entirely
2805 reasonable: after all the message is NOT "waiting for some host".
2806 This is so "obvious" that I'm not sure why it wasn't done
2807 previously. Hope I haven't missed anything, but it can't do any
2808 harm, as the worst effect is to miss an optimization.
2810 (ii) If, despite (i), such a delivery is accidentally attempted, the
2811 routing retry time is respected, so at least it doesn't keep
2812 hammering the server.
2814 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
2815 in ${readsocket because some servers need this prod.
2817 PH/32 Added some extra debug output when updating a wait-xxx database.
2819 PH/33 The hint "could be header name not terminated by colon", which has been
2820 given for certain expansion errors for a long time, was not being given
2821 for the ${if def:h_colon_omitted{... case.
2823 PH/34 The spec says: "With one important exception, whenever a domain list is
2824 being scanned, $domain contains the subject domain." There was at least
2825 one case where this was not true.
2827 PH/35 The error "getsockname() failed: connection reset by peer" was being
2828 written to the panic log as well as the main log, but it isn't really
2829 panic-worthy as it just means the connection died rather early on. I have
2830 removed the panic log writing for the ECONNRESET error when getsockname()
2833 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
2834 runs only) independently of the message's sender address. This meant
2835 that, if the 4xx error was in fact related to the sender, a different
2836 message to the same recipient with a different sender could confuse
2837 things. In particular, this can happen when sending to a greylisting
2838 server, but other circumstances could also provoke similar problems.
2839 I have changed the default so that the retry time for these errors is now
2840 based a combination of the sender and recipient addresses. This change
2841 can be overridden by setting address_retry_include_sender=false in the
2844 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
2845 remote server are returned as part of bounce messages. This was not
2846 happening for LMTP over a pipe (the lmtp transport), but now it is the
2847 same for both kinds of LMTP.
2849 PH/38 Despite being documented as not happening, Exim was rewriting addresses
2850 in header lines that were in fact CNAMEs. This is no longer the case.
2852 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
2853 and queue runs started by the daemon processed all messages. This has
2854 been fixed so that -R and -S can now usefully be given with -q<time>.
2856 PH/40 Import PCRE release 6.7 (fixes some bugs).
2858 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
2860 PH/42 Give an error if -q is specified more than once.
2862 PH/43 Renamed the variables $interface_address and $interface_port as
2863 $received_ip_address and $received_port, to make it clear that these
2864 values apply to message reception, and not to the outgoing interface when
2865 a message is delivered. (The old names remain recognized, of course.)
2867 PH/44 There was no timeout on the connect() call when using a Unix domain
2868 socket in the ${readsocket expansion. There now is.
2870 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
2871 be meaningful with "accept".
2873 SC/01 Eximstats V1.43
2874 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
2876 SC/02 Eximstats V1.44
2877 Use a glob alias rather than an array ref in the generated
2878 parser. This improves both readability and performance.
2880 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
2881 Collect SpamAssassin and rejection statistics.
2882 Don't display local sender or destination tables unless
2883 there is data to show.
2884 Added average volumes into the top table text output.
2886 SC/04 Eximstats V1.46
2887 Collect data on the number of addresses (recipients)
2888 as well as the number of messages.
2890 SC/05 Eximstats V1.47
2891 Added 'Message too big' to the list of mail rejection
2892 reasons (thanks to Marco Gaiarin).
2894 SC/06 Eximstats V1.48
2895 Mainlog lines which have GMT offsets and are too short to
2896 have a flag are now skipped.
2898 SC/07 Eximstats V1.49 (Alain Williams)
2899 Added the -emptyok flag.
2901 SC/08 Eximstats V1.50
2902 Fixes for obtaining the IP address from reject messages.
2904 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
2905 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
2906 whitespace changes from 4.64-PH/27
2908 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
2911 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
2912 are found, allow negative numbers in numeric criteria)
2914 JJ/06 exipick.20061117.2, added new $message_body_missing variable
2916 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
2917 to match changes made in 4.64-PH/43
2919 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
2921 PH/47 Put in an explicit test for a DNS lookup of an address record where the
2922 "domain" is actually an IP address, and force a failure. This locks out
2923 those revolvers/nameservers that support "A-for-A" lookups, in
2924 contravention of the specifications.
2926 PH/48 When a host name was looked up from an IP address, and the subsequent
2927 forward lookup of the name timed out, the host name was left in
2928 $sender_host_name, contrary to the specification.
2930 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
2931 restricted to single-key lookups, Exim was not diagnosing an error if
2932 * or *@ was used with a query-style lookup.
2934 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
2936 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
2937 long ago, but noone interested enough thought of it. Let's just say that
2938 the "1.1" means that there are some new functions that weren't there at
2939 some point in the past.
2941 PH/51 Error processing for expansion failure of helo_data from an smtp
2942 transport during callout processing was broken.
2944 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
2945 tested/used via the -bh/-bhc/-bs options.
2947 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
2948 bug, fixed in subsequent PCRE releases).
2950 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
2951 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
2953 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
2959 SC/01 Use a glob alias rather than an array ref in eximstats generated
2960 parser. This improves both readability and performance.
2962 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
2963 Don't display local sender or destination tables in eximstats unless
2964 there is data to show.
2965 Added average volumes into the eximstats top table text output.
2967 SC/03 Collect data on the number of addresses (recipients) as well
2968 as the number of messages in eximstats.
2970 TF/01 Correct an error in the documentation for the redirect router. Exim
2971 does (usually) call initgroups() when daemonizing.
2973 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
2974 with consistent privilege compared to when running as a daemon.
2976 TF/03 Note in the spec that $authenticated_id is not set for local
2977 submissions from trusted users.
2979 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
2980 Thanks to Dean Brooks <dean@iglou.com> for the patch.
2982 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
2983 by adding some example configuration directives to the default
2984 configuration file. A little bit of work is required to uncomment the
2985 directives and define how usernames and passwords are checked, but
2986 there is now a framework to start from.
2988 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
2989 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
2990 without this. I don't know how relevant this is to other LDAP libraries.
2992 PH/02 Add the verb name to the "unknown ACL verb" error.
2994 PH/03 Magnus Holmgren's patch for filter_prepend_home.
2996 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
2998 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
2999 directory not expanded when it should be if an expanded home directory
3000 was set for the address (which is overridden by the transport).
3002 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
3005 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
3006 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
3007 because it is too late at that time, and has no effect.
3009 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
3010 security issue with \' (bugzilla #107). I could not use the
3011 PQescapeStringConn() function, because it needs a PGconn value as one of
3014 PH/08 When testing addresses using -bt, indicate those final addresses that
3015 are duplicates that would not cause an additional delivery. At least one
3016 person was confused, thinking that -bt output corresponded to deliveries.
3017 (Suppressing duplicates isn't a good idea as you lose the information
3018 about possibly different redirections that led to the duplicates.)
3020 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
3021 systems where poll() doesn't work, in particular OS X.
3023 PH/10 Added more information to debugging output for retry time not reached.
3025 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
3026 operations in malware.c.
3028 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
3031 PH/13 If write_rejectlog was set false when logging was sent to syslog with
3032 syslog_duplication set false, log lines that would normally be written
3033 both the the main log and to the reject log were not written to syslog at
3036 PH/14 In the default configuration, change the use of "message" in ACL warn
3037 statements to "add_header".
3039 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
3040 not followed by a command (e.g. "seen endif").
3042 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
3043 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
3046 PH/17 Added extra conditions to the default value of delay_warning_condition
3050 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
3051 { match{$h_precedence:}{(?i)bulk|list|junk} } \
3052 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
3055 The Auto-Submitted: and various List- headers are standardised, whereas I
3056 don't think Precedence: ever was.
3058 PH/18 Refactored debugging code in route_finduser() to show more information,
3059 in particular, the error code if getpwnam() issues one.
3061 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
3062 This is apparently needed in addition to the PH/07 change above to avoid
3063 any possible encoding problems.
3065 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
3066 but not after initializing Perl.
3068 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
3069 output them only if debugging. By default they are written stderr,
3070 apparently, which is not desirable.
3072 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
3075 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
3078 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
3080 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
3081 authenticated or an ident call has been made. Suppress the default
3082 values for $authenticated_id and $authenticated_sender (but permit -oMai
3083 and -oMas) when testing with -bh.
3085 PH/24 Re-jigged the order of the tests in the default configuration so that the
3086 tests for valid domains and recipients precede the DNS black list and CSA
3087 tests, on the grounds that those ones are more expensive.
3089 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
3090 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
3091 This bug exists in every version of Exim that I still have, right back to
3094 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
3095 However, an attempt to turn on case-sensitivity in a regex key by
3096 including (?-i) didn't work because the subject string was already
3097 lowercased, and the effects were non-intuitive. It turns out that a
3098 one-line patch can be used to allow (?-i) to work as expected.
3104 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
3105 other effects) broke the use of negated acl sub-conditions.
3107 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
3110 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
3111 "Deny" causes Exim to reject the incoming connection with a 554 error.
3112 Unfortunately, if there is a major crisis, such as a disk failure,
3113 tcp-wrappers gives "deny", whereas what one would like would be some
3114 kind of temporary error. A kludge has been added to help with this.
3115 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
3116 554 error is used if errno is still zero or contains ENOENT (which occurs
3117 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
3120 PH/03 Add -lutil to the default FreeBSD LIBS setting.
3122 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
3123 errors. Otherwise a message that provokes a temporary error (when other
3124 messages do not) can cause a whole host to time out.
3126 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
3127 addresses were routed directly to files or pipes from a redirect router.
3128 File deliveries just didn't batch; pipe deliveries might have suffered
3131 PH/06 A failure to get a lock for a hints database would erroneously always say
3132 "Failed to get write lock", even when it was really a read lock.
3134 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
3135 of 0600. This has been changed to use the value of the lockfile_mode
3136 option (which defaults to 0600).
3138 PH/08 Applied small patch from the Sieve maintainer.
3140 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
3141 folder from quota calculations, a direct delivery into this folder messed
3142 up the contents of the maildirsize file. This was because the regex was
3143 used only to exclude .Trash (or whatever) when the size of the mailbox
3144 was calculated. There was no check that a delivery was happening into an
3145 excluded directory. This bug has been fixed by ignoring all quota
3146 processing for deliveries into excluded directories.
3148 PH/10 Added the maildirfolder_create_regex option to appendfile.
3154 PH/01 The code for finding all the local interface addresses on a FreeBSD
3155 system running IPv6 was broken. This may well have applied to all BSD
3156 systems, as well as to others that have similar system calls. The broken
3157 code found IPv4 interfaces correctly, but gave incorrect values for the
3158 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
3159 that it would not match correctly against @[] and not recognize the IPv6
3162 PH/02 The ipliteral router was not recognizing addresses of the form user@
3163 [ipv6:....] because it didn't know about the "ipv6:" prefix.
3165 PH/03 Added disable_ipv6.
3167 PH/04 Changed $reply_address to use the raw form of the headers instead of the
3168 decoded form, because it is most often used to construct To: headers
3169 lines in autoreplies, and the decoded form may well be syntactically
3170 invalid. However, $reply_address has leading white space removed, and all
3171 newlines turned into spaces so that the autoreply transport does not
3174 PH/05 If group was specified without a user on a router, and no group or user
3175 was specified on a transport, the group from the router was ignored.
3177 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
3178 for visible compile-time settings that can be used to change these
3179 numbers, for those that want even more. Backwards compatibility with old
3180 spool files has been maintained. However, going back to a previous Exim
3181 release will lost any variables that are in spool files.
3183 PH/07 Two small changes when running in the test harness: increase delay when
3184 passing a TCP/IP connection to a new process, in case the original
3185 process has to generate a bounce, and remove special handling of
3186 127.0.0.2 (sic), which is no longer necessary.
3188 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
3189 be the same on different OS.
3191 PH/09 Moved a debug statement in filter processing to avoid a race problem when
3194 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
3195 whether --show-vars was specified or not
3197 JJ/02 exipick: Added support for new ACL variable spool format introduced
3200 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
3201 syntactically invalid From: or Reply-to: line, and a filter used this to
3202 generate an autoreply, and therefore failed to obtain an address for the
3203 autoreply, Exim could try to deliver to a non-existent relative file
3204 name, causing unrelated and misleading errors. What now happens is that
3205 it logs this as a hard delivery error, but does not attempt to create a
3208 PH/11 The exinext utility has a -C option for testing purposes, but although
3209 the given file was scanned by exinext itself; it wasn't being passed on
3210 when Exim was called.
3212 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
3213 an end-of-file indication when reading a command response.
3215 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
3216 compiled. In many other places in Exim, IPv6 addresses are always
3217 recognized, so I have changed this. It also means that IPv4 domain
3218 literals of the form [IPV4:n.n.n.n] are now always recognized.
3220 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
3221 used if the router is not running as root, for example, when verifying at
3222 ACL time, or when using -bh. The debugging output from this situation was
3223 non-existent - all you got was a failure to exec. I have made two
3226 (a) Failures to set uid/gid, the current directory, or a process leader
3227 in a subprocess such as that created by queryprogram now generate
3228 suitable debugging output when -d is set.
3230 (b) The queryprogram router detects when it is not running as root,
3231 outputs suitable debugging information if -d is set, and then runs
3232 the subprocess without attempting to change uid/gid.
3234 PH/15 Minor change to Makefile for building test_host (undocumented testing
3237 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
3238 additional section of a DNS packet that returns MX or SRV records.
3239 Instead, it always explicitly searches for A/AAAA records. This avoids
3240 major problems that occur when a DNS server includes only records of one
3241 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
3242 fixed another bug: if SRV records were looked up and the corresponding
3243 address records were *not* found in the additional section, the port
3244 values from the SRV records were lost.
3246 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
3247 using the correct key (the original address) when searching the retry
3248 rules in order to find which one to use for generating the retry hint.
3250 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
3251 adding the default one. Similarly, if it contains a Reply-To: header, the
3252 errors_reply_to option, if set, is not used.
3254 PH/19 When calculating a retry time, Exim used to measure the "time since
3255 failure" by looking at the "first failed" field in the retry record. Now
3256 it does not use this if it is later than than the arrival time of the
3257 message. Instead it uses the arrival time. This makes for better
3258 behaviour in cases where some deliveries succeed, thus re-setting the
3259 "first failed" field. An example is a quota failure for a huge message
3260 when small messages continue to be delivered. Without this change, the
3261 "time since failure" will always be short, possible causing more frequent
3262 delivery attempts for the huge message than are intended.
3263 [Note: This change was subsequently modified - see PH/04 for 4.62.]
3265 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
3266 $1, $2, $3) because the numerical variables can be reset during some
3267 expansion items (e.g. "match"), thereby losing the authentication data.
3269 PH/21 Make -bV show the size of off_t variables so that the test suite can
3270 decide whether to run tests for quotas > 2G.
3272 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
3273 mailbox_size, and mailbox_filecount in the appendfile transport. If a
3274 filecount value is greater than 2G or if a quota value is greater than 2G
3275 on a system where the size of off_t is not greater than 4, a panic error
3278 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
3279 never match. The debug and -bh output now contains an explicit error
3280 message indicating a malformed IPv4 address or mask.
3282 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
3283 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
3284 PH/23 above applies.
3286 PH/25 Do not write to syslog when running in the test harness. The only
3287 occasion when this arises is a failure to open the main or panic logs
3288 (for which there is an explicit test).
3290 PH/26 Added the /no_tell option to "control=freeze".
3292 PH/27 If a host name lookup failed very early in a connection, for example, if
3293 the IP address matched host_lookup and the reverse lookup yielded a name
3294 that did not have a forward lookup, an error message of the form "no IP
3295 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
3296 could be logged. Now it outputs the IP address instead of "NULL".
3298 PH/28 An enabling patch from MH: add new function child_open_exim2() which
3299 allows the sender and the authenticated sender to be set when
3300 submitting a message from within Exim. Since child_open_exim() is
3301 documented for local_scan(), the new function should be too.
3303 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
3304 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
3305 results in an empty string is now treated as unset.
3307 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
3309 PH/31 Added sender_verify_fail logging option.
3311 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
3312 needed by RFC 822 but not by RFC 2822 was commented out. I have now
3313 tidied the source and removed it altogether.
3315 PH/33 When a queue run was abandoned because the load average was too high, a
3316 log line was always written; now it is written only if the queue_run log
3317 selector is set. In addition, the log line for abandonment now contains
3318 information about the queue run such as the pid. This is always present
3319 in "start" and "stop" lines but was omitted from the "abandon" line.
3321 PH/34 Omit spaces between a header name and the colon in the error message that
3322 is given when verify = headers_syntax fails (if there are lots of them,
3323 the message gets confusing).
3325 PH/35 Change the default for dns_check_names_pattern to allow slashes within
3326 names, as there are now some PTR records that contain slashes. This check
3327 is only to protect against broken name servers that fall over on strange
3328 characters, so the fact that it applies to all lookups doesn't matter.
3330 PH/36 Now that the new test suite is complete, we can remove some of the
3331 special code in Exim that was needed for the old test suite. For example,
3332 sorting DNS records because real resolvers return them in an arbitrary
3333 order. The new test suite's fake resolver always returns records in the
3336 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
3337 bounces) except when queue_only is set, to avoid logging races between
3338 the different processes.
3340 PH/38 Panic-die if .include specifies a non-absolute path.
3342 PH/39 A tweak to the "H" retry rule from its user.
3344 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
3345 a label. They prevented compilation on older perls.
3347 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
3348 a warning to be raised on newish perls.
3350 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
3351 on queue. Changes to match documented behaviour of showing count of
3352 messages matching specified criteria.
3354 PH/40 Changed the default ident timeout from 30s to 5s.
3356 PH/41 Added support for the use of login_cap features, on those BSD systems
3357 that have them, for controlling the resources used by pipe deliveries.
3359 PH/42 The content-scanning code uses fopen() to create files in which to put
3360 message data. Previously it was not paying any attention to the mode of
3361 the files. Exim runs with umask(0) because the rest of the code creates
3362 files with open(), and sets the required mode explicitly. Thus, these
3363 files were ending up world-writeable. This was not a big issue, because,
3364 being within the spool directory, they were not world-accessible. I have
3365 created a function called modefopen, which takes an additional mode
3366 argument. It sets umask(777), creates the file, chmods it to the required
3367 mode, then resets the umask. All the relevant calls to fopen() in the
3368 content scanning code have been changed to use this function.
3370 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
3371 to 24 hours. This avoids potential overflow problems when processing G
3372 and H retry rules. I suspect nobody ever tinkers with this value.
3374 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
3376 PH/45 When the plaintext authenticator is running as a client, the server's
3377 challenges are checked to ensure they are valid base64 strings. By
3378 default, the authentication attempt is cancelled if an invalid string is
3379 received. Setting client_ignore_invalid_base64 true ignores these errors.
3380 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
3381 they are received. Thus, the responses can be made to depend on the
3382 challenges. If an invalid string is ignored, an empty string is placed in
3385 PH/46 Messages that are created by the autoreply transport now contains a
3386 References: header, in accordance with RFCs 2822 and 3834.
3388 PH/47 Added authenticated_sender_force to the smtp transport.
3390 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
3392 PH/49 Installed latest patch from the Sieve maintainer.
3394 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
3395 was also set, the appendfile transport was unnecessarily scanning a
3396 directory of message files (e.g. for maildir delivery) to find the count
3397 of files (along with the size), even though it did not need this
3398 information. It now does the scan only if it needs to find either the
3399 size of the count of files.
3401 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
3403 PH/52 Two bugs concerned with error handling when the smtp transport is
3406 (i) Exim was not creating retry information for temporary errors given
3407 for individual recipients after the DATA command when the smtp transport
3408 was used in LMTP mode. This meant that they could be retried too
3409 frequently, and not timed out correctly.
3411 (ii) Exim was setting the flag that allows error details to be returned
3412 for LMTP errors on RCPT commands, but not for LMTP errors for individual
3413 recipients that were returned after the DATA command.
3415 PH/53 This is related to PH/52, but is more general: for any failing address,
3416 when detailed error information was permitted to be returned to the
3417 sender, but the error was temporary, then after the final timeout, only
3418 "retry timeout exceeded" was returned. Now it returns the full error as
3419 well as "retry timeout exceeded".
3421 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
3422 do this, and (what is worse) MTAs that accept it.
3424 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
3425 will now be deprecated.
3427 PH/56 New os.c-cygwin from the Cygwin maintainer.
3429 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
3430 formats (previously only available in exim formats via -bpr, -bpru,
3431 and -bpra. Now also available in native and exiqgrep formats)
3433 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
3434 with very large, slow to parse queues
3436 JJ/08 exipick: added ! as generic prefix to negate any criteria format
3438 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
3440 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
3441 responses to authentication challenges, though it was showing the
3442 challenges; (ii) I've removed the CR characters from the debug output for
3445 PH/58 Allow for the insertion of a newline as well as a space when a string
3446 is turned into more than one encoded-word during RFC 2047 encoding. The
3447 Sieve code now uses this.
3449 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
3450 data_4xx, lost_connection, tls_required.
3452 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
3453 message was being sent as an SMTP response.
3455 PH/61 Add -l and -k options to exicyclog.
3457 PH/62 When verifying, if an address was redirected to one new address, so that
3458 verification continued, and the new address failed or deferred after
3459 having set something in $address_data, the value of $address_data was not
3460 passed back to the ACL. This was different to the case when no
3461 redirection occurred. The value is now passed back in both cases.
3463 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
3464 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
3465 use login_cap.h, so on its own it isn't the distinguishing feature. The
3466 new name refers directly to the setclassresources() function.
3468 PH/65 Added configuration files for NetBSD3.
3470 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
3472 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
3473 is preferred over IPv4.
3475 PH/68 The bounce_return_message and bounce_return_body options were not being
3476 honoured for bounces generated during the reception of non-SMTP messages.
3477 In particular, this applied to messages rejected by the ACL. This bug has
3478 been fixed. However, if bounce_return_message is true and bounce_return_
3479 body is false, the headers that are returned for a non-SMTP message
3480 include only those that have been read before the error was detected.
3481 (In the case of an ACL rejection, they have all been read.)
3483 PH/69 The HTML version of the specification is now built in a directory called
3484 spec_html instead of spec.html, because the latter looks like a path with
3485 a MIME-type, and this confuses some software.
3487 PH/70 Catch two compiler warnings in sieve.c.
3489 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
3490 function verify_get_ident() calls ip_connect() to connect a socket, but
3491 if the "connect()" function timed out, ip_connect() used to close the
3492 socket. However, verify_get_ident() also closes the socket later, and in
3493 between Exim writes to the log, which may get opened at this point. When
3494 the socket was closed in ip_connect(), the log could get the same file
3495 descriptor number as the socket. This naturally causes chaos. The fix is
3496 not to close the socket in ip_connect(); the socket should be closed by
3497 the function that creates it. There was only one place in the code where
3498 this was missing, in the iplookup router, which I don't think anybody now
3499 uses, but I've fixed it anyway.
3501 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
3502 well as to direct DNS lookups. Otherwise the handling of names in host
3503 lists is inconsistent and therefore confusing.
3509 PH/01 Two changes to the default runtime configuration:
3511 (1) Move the checks for relay_from_hosts and authenticated clients from
3512 after to before the (commented out) DNS black list checks.
3514 (2) Add control=submission to the relay_from_hosts and authenticated
3515 clients checks, on the grounds that messages accepted by these
3516 statements are most likely to be submissions.
3518 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
3520 (1) Generate an error if the third argument for the ${prvs expansion is
3523 (2) Treat a missing third argument of ${prvscheck as if it were an empty
3526 (3) Reset the variables that are obtained from the first argument of
3527 ${prvscheck and used in the second argument before leaving the code,
3528 because their memory is reclaimed, so using them afterwards may do
3531 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
3532 one (it's much easier than Tom thought :-).
3534 (5) Because of (4), we can now allow for the use of $prvscheck_result
3535 inside the third argument.
3537 PH/03 For some reason, the default setting of PATH when running a command from
3538 a pipe transport was just "/usr/bin". I have changed it to
3541 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
3542 anything to be listed in the output from -bV.
3544 PH/05 When a filter generated an autoreply, the entire To: header line was
3545 quoted in the delivery log line, like this:
3547 => >A.N.Other <ano@some.domain> <original@ddress> ...
3549 This has been changed so that it extracts the operative address. There
3550 may be more than one such address. If so, they are comma-separated, like
3553 => >ano@some.domain,ona@other.domain <original@ddress> ...
3555 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
3556 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
3557 not being looked up in the rDNS to get a host name, Exim was showing the
3558 IP address twice in Received: lines, even though the IP addresses were
3559 identical. For example:
3561 Received: from [1.2.3.4] (helo=[1.2.3.4])
3563 However, if the real host name was known, it was omitting the HELO data
3564 if it matched the actual IP address. This has been tidied up so that it
3565 doesn't show the same IP address twice.
3567 PH/07 When both +timestamp and +memory debugging was on, the value given by
3568 $tod_xxx expansions could be wrong, because the tod_stamp() function was
3569 called by the debug printing, thereby overwriting the timestamp buffer.
3570 Debugging no longer uses the tod_stamp() function when +timestamp is set.
3572 PH/08 When the original message was included in an autoreply transport, it
3573 always said "this is a copy of the message, including all the headers",
3574 even if body_only or headers_only was set. It now gives an appropriate
3577 PH/09 Applied a patch from the Sieve maintainer which:
3579 o fixes some comments
3580 o adds the (disabled) notify extension core
3581 o adds some debug output for the result of if/elsif tests
3582 o points to the current vacation draft in the documentation
3583 and documents the missing references header update
3587 o fixes a bug in processing the envelope test (when testing
3588 multiple envelope elements, the last element determined the
3591 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
3592 Electronic Mail") by including:
3594 Auto-submitted: auto-generated
3596 in the messages that it generates (bounce messages and others, such as
3597 warnings). In the case of bounce messages for non-SMTP messages, there was
3598 also a typo: it was using "Auto_submitted" (underscore instead of
3599 hyphen). Since every message generated by Exim is necessarily in response
3600 to another message, thes have all been changed to:
3602 Auto-Submitted: auto-replied
3604 in accordance with these statements in the RFC:
3606 The auto-replied keyword:
3608 - SHOULD be used on messages sent in direct response to another
3609 message by an automatic process,
3611 - MUST NOT be used on manually-generated messages,
3613 - MAY be used on Delivery Status Notifications (DSNs) and Message
3614 Disposition Notifications (MDNs),
3616 - MUST NOT be used on messages generated by automatic or periodic
3617 processes, except for messages which are automatic responses to
3620 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
3621 to the default Received: header definition.
3623 PH/12 Added log selector acl_warn_skipped (default on).
3625 PH/13 After a successful wildlsearch lookup, discard the values of numeric
3626 variables because (a) they are in the wrong storage pool and (b) even if
3627 they were copied, it wouldn't work properly because of the caching.
3629 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
3630 checking when decoding. Apparently there are clients that generate
3631 overlong encoded strings. Why am I not surprised?
3633 PH/15 If the first argument of "${if match_address" was not empty, but did not
3634 contain an "@" character, Exim crashed. Now it writes a panic log message
3635 and treats the condition as false.
3637 PH/16 In autoreply, treat an empty string for "once" the same as unset.
3639 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
3640 extension "envelope-auth". The code is finished and in agreement with
3641 other implementations, but there is no documentation so far and in fact,
3642 nobody wrote the draft yet. This extension is currently #undef'ed, thus
3643 not changing the active code.
3645 Print executed "if" and "elsif" statements when debugging is used. This
3646 helps a great deal to understand what a filter does.
3648 Document more things not specified clearly in RFC3028. I had all this
3649 sorted out, when out of a sudden new issues came to my mind. Oops."
3651 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
3654 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
3655 canonical form (as documented). However, after a host name lookup from
3656 the IP address, check_host() was doing a simple string comparison with
3657 addresses acquired from the DNS when checking that the found name did
3658 have the original IP as one of its addresses. Since any found IPv6
3659 addresses are likely to be in abbreviated form, the comparison could
3660 fail. Luckily, there already exists a function for doing the comparison
3661 by converting both addresses to binary, so now that is used instead of
3662 the text comparison.
3664 PH/20 There was another similar case to PH/19, when a complete host name was
3665 given in a host list; looking up its IP address could give an abbreviated
3666 form, whereas the current host's name might or might not be abbreviated.
3667 The same fix has been applied.
3673 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
3674 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
3677 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
3678 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
3680 PH/03 Typo: missing ".o" in src/pcre/Makefile.
3682 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
3683 header line, restrict the check to what is listed in RFCs 2369 and 2929.
3684 Also, for "Auto-Submitted", treat anything other than "no" as
3685 non-personal, in accordance with RFC 3834. (Previously it treated
3686 anything starting "auto-" as non-personal.)
3688 TF/01 The control=submission/name=... option had a problem with syntax
3689 errors if the name included a slash character. The /name= option
3690 now slurps the rest of the string, so it can include any characters
3691 but it must come last in the list of options (after /sender_retain
3694 PH/05 Some modifications to the interface to the fake nameserver for the new
3702 TK/01 Added the "success_on_redirect" address verification option. See
3703 NewStuff for rationale and an example.
3705 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
3707 PH/02 Patch to exigrep to allow it to work on syslog lines.
3709 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
3710 fread() to skip over the body file's header line, because in Cygwin the
3711 header line is locked and is inaccessible.
3713 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
3714 co-exist for some time) to make it clear that it is the Exim ID that is
3715 referenced, not the Message-ID: header line.
3717 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
3718 string_format() function, because snprintf() does not exist on all
3721 PH/06 The use of forbid_filter_existstest now also locks out the use of the
3722 ${stat: expansion item.
3724 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
3725 protocol synchronization error", to keep the pedants happy.
3727 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
3728 well as for IRIX systems, when gcc is being used. See the host.c source
3731 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
3733 PH/10 Named domain lists were not working if used in a queue_smtp_domains
3736 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
3737 transport and to the smtp transport in LMTP mode.
3739 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
3741 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
3742 run a filter in a subprocess. This could lead to confusion in subsequent
3743 lookups in the parent process. There should also be a search_tidyup() at
3744 the end of the subprocess.
3746 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
3747 only if the host matched helo_try_verify_hosts, which caused the
3748 verification to occur when the EHLO/HELO command was issued. The ACL just
3749 tested the remembered result. Now, if a previous verification attempt has
3750 not happened, "verify = helo" does it there and then.
3752 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
3754 TK/03 Fix log output including CR from clamd.
3756 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
3757 exist provoked a memory error which could cause a segfault.
3759 PH/15 Installed PCRE 6.2
3761 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
3763 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
3764 of the problem. Specifically, suggested +O2 rather than +O1 for the
3767 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
3769 PH/20 If a delivery was routed to a non-standard port by means of an SRV
3770 record, the port was not correctly logged when the outgoing_port log
3771 selector was set (it logged the transort's default port).
3773 PH/21 Added support for host-specific ports to manualroute, queryprogram,
3774 fallback_hosts, and "hosts" in the smtp transport.
3776 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
3777 host errors such as "Connection refused".
3779 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
3780 authentication with radiusclient 0.4.9:
3782 - Error returned from rc_read_config was caught wrongly
3783 - Username/password not passed on to radius server due to wrong length.
3785 The presumption is that some radiusclient API changes for 4.51/PH/17
3786 were not taken care of correctly. The code is still untested by me (my
3787 Linux distribution still has 0.3.2 of radiusclient), but it was
3788 contributed by a Radius user.
3790 PH/24 When doing a callout, the value of $domain wasn't set correctly when
3791 expanding the "port" option of the smtp transport.
3793 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
3794 while reading a MIME header. Thanks to Tom Hughes for a patch.
3796 PH/24 Include config.h inside local_scan.h so that configuration settings are
3799 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
3800 that in an ACL for RCPT (for example), you can examine exactly what was
3803 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
3804 commands, but it was not correctly comparing the address with the actual
3805 client host address. Thus, it would show the EHLO address in Received:
3806 header lines when this was not necessary.
3808 PH/27 Added the % operator to ${eval:}.
3810 PH/28 Exim tries to create and chdir to its spool directory when it starts;
3811 it should be ignoring failures (because with -C, for example, it has lost
3812 privilege). It wasn't ignoring creation failures other than "already
3815 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
3818 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
3819 because an input line was too long, either on its own, or by virtue of
3820 too many continuations, the temporary file was not being removed, and the
3821 return code was incorrect.
3823 PH/31 Missing "BOOL" in function definition in filtertest.c.
3825 PH/32 Applied Sieve patches from the maintainer.
3827 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
3829 PH/33 Added "verify = not_blind".
3831 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
3832 Local/Makefile (with some defaults set). These are used in built scripts
3833 such as exicyclog, but they have never been used in the exim_install
3834 script (though there are many overriding facilities there). I have
3835 arranged that the exim_install script now takes note of these two
3838 PH/35 Installed configuration files for Dragonfly.
3840 PH/36 When a locally submitted message by a trusted user did not contain a
3841 From: header, and the sender address was obtained from -f or from an SMTP
3842 MAIL command, and the trusted user did not use -F to supply a sender
3843 name, $originator_name was incorrectly used when constructing a From:
3844 header. Furthermore, $originator_name was used for submission mode
3845 messages from external hosts without From: headers in a similar way,
3846 which is clearly wrong.
3848 PH/37 Added control=suppress_local_fixups.
3850 PH/38 When log_selector = +received_sender was set, and the addition of the
3851 sender made the log line's construction buffer exactly full, or one byte
3852 less than full, an overflow happened when the terminating "\n" was
3855 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
3856 when the result of a list match is failure because a DNS lookup failed.
3858 PH/40 RM_COMMAND is now used in the building process.
3860 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
3861 the "build-* directories that it finds.
3863 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
3864 address in a domain literal was a prefix of an interface address.
3866 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
3867 when verifying a sender address, unless rewrite_headers is false.
3869 PH/44 Wrote a long comment about why errors_to addresses are verified as
3870 recipients, not senders.
3872 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
3873 the ratelimit ACL was added.
3875 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
3877 PH/46 Added extra information about PostgreSQL errors to the error string.
3879 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
3880 avoiding the need to install special zones in a real server. This is
3881 backwards compatible; if it can't find the fake resolver, it drops back.
3882 Thus, both old and new test suites can be run.
3884 TF/02 Added util/ratelimit.pl
3886 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
3887 clock is set back in time.
3889 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
3890 Candler <B.Candler@pobox.com>.
3892 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
3893 OK for addresses that are the result of redirections.
3895 PH/48 A number of further additions for the benefit of the new test suite,
3896 including a fake gethostbyname() that interfaces to the fake DNS resolver
3899 TF/06 The fix for widen_domains has also been applied to qualify_single and
3900 search_parents which are the other dnslookup options that can cause
3903 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
3906 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
3908 TF/07 Exim produced the error message "an SRV record indicated no SMTP
3909 service" if it encountered an MX record with an empty target hostname.
3910 The message is now "an MX or SRV record indicated no SMTP service".
3912 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
3913 if the DNS of the sending site is misconfigured. This is quite a
3914 common situation. This change restores the behaviour of treating a
3915 helo verification defer as a failure.
3917 PH/51 If self=fail was set on a router, the bounce message did not include the
3918 actual error message.
3924 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
3926 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
3927 command itself ended in error, the underlying message about the transport
3928 filter timeout was being overwritten with the pipe command error. Now the
3929 underlying error message should be appended to the second error message.
3931 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
3933 PH/02 Reduce the amount of output that "make" produces by default. Full output
3934 can still be requested.
3936 PH/03 The warning log line about a condition test deferring for a "warn" verb
3937 was being output only once per connection, rather than after each
3938 occurrence (because it was using the same function as for successful
3939 "warn" verbs). This seems wrong, so I have changed it.
3941 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
3942 it should not have, which might have caused a crash in the right
3943 circumstances, but probably never did.
3945 PH/04 Installed a modified version of Tony Finch's patch to make submission
3946 mode fix the return path as well as the Sender: header line, and to
3947 add a /name= option so that you can make the user's friendly name appear
3950 TF/03 Added the control = fakedefer ACL modifier.
3952 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
3953 Mark Lowes for thorough testing.
3955 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
3957 TK/03 Merged latest SRS patch from Miles Wilton.
3959 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
3960 with the definition in sysexits.h (which is #included earlier).
3961 Fortunately, Exim does not actually use EX_OK. The code used to try to
3962 preserve the sysexits.h value, by assuming that macro definitions were
3963 scanned for macro replacements. I have been disabused of this notion,
3964 so now the code just undefines EX_OK before #including unistd.h.
3966 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
3967 in the smtp transport. When a block could not be written in a single
3968 write() function, the timeout was being re-applied to each part-write.
3969 This seems wrong - if the receiver was accepting one byte at a time it
3970 would take for ever. The timeout is now adjusted when this happens. It
3971 doesn't have to be particularly precise.
3973 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
3974 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
3976 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
3977 without a domain if the check to <postmaster@domain> fails.
3979 SC/01 Eximstats: added -xls and the ability to specify output files
3980 (patch written by Frank Heydlauf).
3982 SC/02 Eximstats: use FileHandles for outputting results.
3984 SC/03 Eximstats: allow any combination of xls, txt, and html output.
3986 SC/04 Eximstats: fixed display of large numbers with -nvr option
3988 SC/05 Eximstats: fixed merging of reports with empty tables.
3990 SC/06 Eximstats: added the -include_original_destination flag
3992 SC/07 Eximstats: removed tabs and trailing whitespace.
3994 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
3996 TK/06 MBOX spool code: Add real "From " MBOX separator line
3997 so the .eml file is really in mbox format (even though
3998 most programs do not really care). Patch from Alex Miller.
4000 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
4001 The latter is generated from $received_to and is only set if the
4002 message has one envelope recipient. SA can use these headers,
4003 obviously out-of-the-box. Patch from Alex Miller.
4005 PH/08 The ${def test on a variable was returning false if the variable's
4006 value was "0", contrary to what the specification has always said!
4007 The result should be true unless the variable is empty.
4009 PH/09 The syntax error of a character other than { following "${if
4010 def:variable_name" (after optional whitespace) was not being diagnosed.
4011 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
4012 accidental colon was present, for example, could give incorrect results.
4014 PH/10 Tidied the code in a number of places where the st_size field of a stat()
4015 result is used (not including appendfile, where other changes are about
4018 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
4019 This involved changing a lot of size variables from int to off_t. It
4020 should work with maildirs and everything.
4022 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
4023 spamd dying while we are connected to it.
4025 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
4028 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
4029 function for BDB 4.3.
4031 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
4033 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
4034 the sources for PCRE have been re-arranged and more files are now
4037 PH/15 The code I had for printing potentially long long variables in PH/11
4038 above was not the best (it lost precision). The length of off_t variables
4039 is now inspected at build time, and an appropriate printing format (%ld
4040 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
4041 to be "long long int" or "long int". This is needed for the internal
4042 formatting function string_vformat().
4044 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
4045 the configuration file to be ":syslog", then the script "guesses" where
4046 the logs files are, rather than using the compiled in default. In our
4047 case the guess is not the same as the compiled default, so the script
4048 suddenly stopped working when I started to use syslog. The patch checks
4049 to see if log_file_path is "". If so, it attempts to read it from exim
4050 with no configuration file to get the compiled in version, before it
4051 falls back to the previous guessing code."
4053 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
4054 implementing BATV in an Exim configuration. See NewStuff for the gory
4057 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
4058 Makefile that are specific to HP-UX.
4060 PH/18 If the "use_postmaster" option was set for a recipient callout together
4061 with the "random" option, the postmaster address was used as the MAIL
4062 FROM address for the random test, but not for the subsequent recipient
4063 test. It is now used for both.
4065 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
4066 patch removes a few documentation additions to RFC 3028, because the
4067 latest draft now contains them. It adds the new en;ascii-case comparator
4068 and a new error check for 8bit text in MIME parts. Comparator and
4069 require names are now matched exactly. I enabled the subaddress
4070 extension, but it is not well tested yet (read: it works for me)."
4072 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
4073 rework some of the code of TK/09 above to avoid the hardwired use of
4074 "%lld" and "long long". Replaced the call to snprintf() with a call to
4077 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
4078 records point to non-existent hosts", "retry timeout exceeded", and
4079 "retry time not reached for any host after a long failure period".
4081 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
4082 experimental DomainKeys support:
4084 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
4085 (2) On an error such as an illegally used "control", the wrong name for
4086 the control was given.
4088 These problems did NOT occur unless DomainKeys support was compiled.
4090 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
4092 PH/24 Added ${if match_ip condition.
4094 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
4095 databases so that it will be absolutely obvious if a crash occurs in the
4096 DB library. This is a regular occurrence (often caused by mis-matched
4099 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
4100 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
4101 picked up on a user's system that detects such things. There doesn't seem
4102 to be a gcc warning option for this - only an attribute that has to be
4103 put on the function's prototype. It seems that in Fedora Core 4 they have
4104 set this on a number of new functions. No doubt there will be more in due
4107 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
4108 specify a transport. However, if an address that was verified by such a
4109 router was the subject of a callout, Exim crashed because it tried to
4110 read the rcpt_include_affixes from the non-existent transport. Now it
4111 just assumes that the setting of that option is false. This bug was
4112 introduced by 4.51/PH/31.
4114 PH/28 Changed -d+all to exclude +memory, because that information is very
4115 rarely of interest, but it makes the output a lot bigger. People tend to
4116 do -d+all out of habit.
4118 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
4119 code in os-type was giving problems when libc.so lives in lib64, like on
4122 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
4123 aren't the modern standard, and it seems that some systems' include files
4124 don't always have them. Exim was already checking for some of the newer
4125 ones like T_AAAA, and defining it itself. I've added checks for all the
4126 record types that Exim uses.
4128 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
4129 not automatically generating a new one, as it is supposed to. This
4130 prevented TLS from working. If the file did exist, but contained invalid
4131 data, a new version was generated, as expected. It was only the case of a
4132 non-existent file that was broken.
4134 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
4135 with a change in libdomainkeys > 0.64.
4137 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
4138 from DNS. If the selector record carries the flag, it now has
4139 precedence over the domain-wide flag.
4141 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
4143 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
4144 the use of a port name that isn't defined in /etc/services) occurred, the
4145 message was deferred as in a normal delivery, and thus remained on the
4146 spool, instead of being failed because of the mua_wrapper setting. This
4147 is now fixed, and I tidied up some of the mua_wrapper messages at the
4150 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
4151 the messages in a hash of arrays rather than using individual hashes.
4152 This is a bit cleaner and results in dramatic memory savings, albeit
4153 at a slight CPU cost.
4155 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
4156 as requested by Marc Sherman.
4158 SC/10 Eximstats: added histograms for user specified patterns as requested
4161 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
4163 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
4164 fopen() in the content-scanning modules that did not already have it.
4170 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
4171 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
4173 TK/02 Fix ACL "control" statement not being available in MIME ACL.
4175 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
4177 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
4178 to test Sieve filters that use "vacation".
4180 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
4181 that changes the way the GnuTLS parameters are stored in the cache file.
4182 The new format can be generated externally. For backward compatibility,
4183 if the data in the cache doesn't make sense, Exim assumes it has read an
4184 old-format file, and it generates new data and writes a new file. This
4185 means that you can't go back to an older release without removing the
4188 PH/03 A redirect router that has both "unseen" and "one_time" set does not
4189 work if there are any delivery delays because "one_time" forces the
4190 parent to be marked "delivered", so its unseen clone is never tried
4191 again. For this reason, Exim now forbids the simultaneous setting of
4194 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
4195 redirected to themselves ("homonym" addresses). Read the long ChangeLog
4196 entry if you want to know the details. The fix, however, neglected to
4197 consider the case when local delivery batching is involved. The test for
4198 "previously delivered" was not happening when checking to see if an
4199 address could be batched with a previous (undelivered) one; under
4200 certain circumstances this could lead to multiple deliveries to the same
4203 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
4204 in its include files, and this causes problems building Exim.
4206 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
4207 header_syntax) but Exim was just ignoring anything given after a slash.
4208 In particular, this caused confusion with an attempt to use "verify =
4209 reverse_host_lookup/defer_ok". An error is now given when options are
4210 supplied for verify items that do not have them. (Maybe reverse_host_
4211 lookup should have a defer_ok option, but that's a different point.)
4213 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
4214 defined by RFC 821) to 2048, because there were problems with some AUTH
4215 commands, and RFC 1869 says the size should be increased for extended
4216 SMTP commands that take arguments.
4218 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
4221 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
4222 "unknown" error; now it says that the functionality isn't in the binary.
4224 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
4225 an address' error message when a string expansion fails (syntax or
4226 whatever). Otherwise the password may appear in the log. Following change
4227 PH/42 below, there is no longer a chance of it appearing in a bounce
4230 PH/11 Installed exipick version 20050225.0 from John Jetmore.
4232 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
4233 of its addresses was ever tried. (Bugzilla bug #2.)
4235 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
4236 the result incorrectly in the debug output. (It correctly added a newline
4237 to what was transported.)
4239 TF/01 Added $received_time.
4241 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
4242 commented out examples of how to interface to a virus scanner and to
4243 SpamAssassin. Also added commented examples of av_scanner and
4244 spamd_address settings.
4246 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
4247 and controls are allowed in which ACLs. There were a couple of minor
4248 errors. Some of the entries in the conditions table (which is a table of
4249 where they are NOT allowed) were getting very unwieldy; rewrote them as a
4250 negation of where the condition IS allowed.
4252 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
4254 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
4255 header file does not have a version number, so I've had to invent a new
4256 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
4257 API. The code is untested by me (my Linux distribution still has 0.3.2 of
4258 radiusclient), but it was contributed by a Radius user.
4260 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
4261 files or directories, for OpenSSL.
4263 PH/19 When an Exim process that is running as root has to create an Exim log
4264 file, it does so in a subprocess that runs as exim:exim so as to get the
4265 ownership right at creation (otherwise, other Exim processes might see
4266 the file with the wrong ownership). There was no test for failure of this
4267 fork() call, which would lead to the process getting stuck as it waited
4268 for a non-existent subprocess. Forks do occasionally fail when resources
4269 run out. I reviewed all the other calls to fork(); they all seem to check
4272 PH/20 When checking for unexpected SMTP input at connect time (before writing
4273 the banner), Exim was not dealing correctly with a non-positive return
4274 from the read() function. If the client had disconnected by this time,
4275 the result was a log entry for a synchronization error with an empty
4276 string after "input=" when read() returned zero. If read() returned -1
4277 (an event I could not check), uninitialized data bytes were printed.
4278 There were reports of junk text (parts of files, etc) appearing after
4281 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
4283 PH/22 Added support for macro redefinition, and (re)definition in between
4284 driver and ACL definitions.
4286 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
4287 forgetting to use the resulting value; it was using the unexpanded value.
4289 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
4290 hadn't been configured. The fix is from Juergen Kreileder, who
4291 understands it better than I do:
4293 "Here's what I see happening with three configured cyrus_sasl
4294 authenticators configured (plain, login, cram-md5):
4296 On startup auth_cyrus_sasl_init() gets called for each of these.
4297 This means three calls to sasl_listmech() without a specified mech_list.
4298 => SASL tests which mechs of all available mechs actually work
4299 => three warnings about OTP not working
4300 => the returned list contains: plain, login, cram-md5, digest-md5, ...
4302 With the patch, sasl_listmech() also gets called three times. But now
4303 SASL's mech_list option is set to the server_mech specified in the the
4304 authenticator. Or in other words, the answer from sasl_listmech()
4305 gets limited to just the mech you're testing for (which is different
4307 => the return list contains just 'plain' or 'login', 'cram-md5' or
4308 nothing depending on the value of ob->server_mech.
4310 I've just tested the patch: Authentication still works fine,
4311 unavailable mechs specified in the exim configuration are still
4312 caught, and the auth.log warnings about OTP are gone."
4314 PH/25 When debugging is enabled, the contents of the command line are added
4315 to the debugging output, even when log_selector=+arguments is not
4318 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
4319 answer is "GNU", and only if the return is "GNU/something" is the answer
4322 PH/27 $acl_verify_message is now set immediately after the failure of a
4323 verification in an ACL, and so is available in subsequent modifiers. In
4324 particular, the message can be preserved by coding like this:
4326 warn !verify = sender
4327 set acl_m0 = $acl_verify_message
4329 Previously, $acl_verify_message was set only while expanding "message"
4330 and "log_message" when a very denied access.
4332 PH/28 Modified OS/os.c-Linux with
4334 -#ifndef OS_LOAD_AVERAGE
4335 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
4337 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
4338 nomenclature these days.)
4340 PH/29 Installed patch from the Sieve maintainer that adds the options
4341 sieve_useraddress and sieve_subaddress to the redirect router.
4343 PH/30 In these circumstances:
4344 . Two addresses routed to the same list of hosts;
4345 . First host does not offer TLS;
4346 . First host accepts first address;
4347 . First host gives temporary error to second address;
4348 . Second host offers TLS and a TLS session is established;
4349 . Second host accepts second address.
4350 Exim incorrectly logged both deliveries with the TLS parameters (cipher
4351 and peerdn, if requested) that were in fact used only for the second
4354 PH/31 When doing a callout as part of verifying an address, Exim was not paying
4355 attention to any local part prefix or suffix that was matched by the
4356 router that accepted the address. It now behaves in the same way as it
4357 does for delivery: the affixes are removed from the local part unless
4358 rcpt_include_affixes is set on the transport.
4360 PH/32 Add the sender address, as F=<...>, to the log line when logging a
4361 timeout during the DATA phase of an incoming message.
4363 PH/33 Sieve envelope tests were broken for match types other than :is. I have
4364 applied a patch sanctioned by the Sieve maintainer.
4366 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
4367 the uid or gid is negative. A case of a negative gid caused this to be
4368 noticed. The fix allows for either to be negative.
4370 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
4371 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
4374 PH/36 The change PH/12 above was broken. Fixed it.
4376 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
4377 the grounds that routing the same address twice would always produce the
4378 same answer. This might have been true once, but it is certainly no
4379 longer true now. Routing a child address may depend on the previous
4380 routing that produced that child. Some complicated redirection strategies
4381 went wrong when messages had multiple recipients, and made Exim's
4382 behaviour dependent on the order in which the addresses were given.
4384 I have moved the duplicate checking until after the routing is complete.
4385 Exim scans the addresses that are assigned to local and remote
4386 transports, and removes any duplicates. This means that more work will be
4387 done, as duplicates will always all be routed, but duplicates are
4388 presumably rare, so I don't expect this is of any significance.
4390 For deliveries to pipes, files, and autoreplies, the duplicate checking
4391 still happens during the routing process, since they are not going to be
4394 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
4395 It corrects a timeout issue with spamd. This is Ian's comment: "The
4396 background is that sometimes spamd either never reads data from a
4397 connection it has accepted, or it never writes response data. The exiscan
4398 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
4399 blindly assumes that writes won't block so it may never time out."
4401 PH/39 Allow G after quota size as well as K and M.
4403 PH/40 The value set for $authenticated_id in an authenticator may not contain
4404 binary zeroes or newlines because the value is written to log lines and
4405 to spool files. There was no check on this. Now the value is run through
4406 the string_printing() function so that such characters are converted to
4407 printable escape sequences.
4409 PH/41 $message_linecount is a new variable that contains the total number of
4410 lines in the message. Compare $body_linecount, which is the count for the
4413 PH/42 Exim no longer gives details of delivery errors for specific addresses in
4414 bounce and delay warning messages, except in certain special cases, which
4417 (a) An SMTP error message from a remote host;
4418 (b) A message specified in a :fail: redirection;
4419 (c) A message specified in a "fail" command in a system filter;
4420 (d) A message specified in a FAIL return from the queryprogram router;
4421 (e) A message specified by the cannot_route_message router option.
4423 In these cases only, Exim does include the error details in bounce and
4424 warning messages. There are also a few cases where bland messages such
4425 as "unrouteable address" or "local delivery error" are given.
4427 PH/43 $value is now also set for the "else" part of a ${run expansion.
4429 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
4430 being worked on, but at least Exim now implements the latest version to
4433 PH/45 In a pipe transport, although a timeout while waiting for the pipe
4434 process to complete was treated as a delivery failure, a timeout while
4435 writing the message to the pipe was logged, but erroneously treated as a
4436 successful delivery. Such timeouts include transport filter timeouts. For
4437 consistency with the overall process timeout, these timeouts are now
4438 treated as errors, giving rise to delivery failures by default. However,
4439 there is now a new Boolean option for the pipe transport called
4440 timeout_defer, which, if set TRUE, converts the failures into defers for
4441 both kinds of timeout. A transport filter timeout is now identified in
4444 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
4445 systems where "make" and "gmake" are different, calling "gmake" at top
4446 level broke things. I've arranged for the value of $(MAKE) to be passed
4447 from the Makefile to this script so that it can call the same version of
4451 A note about Exim versions 4.44 and 4.50
4452 ----------------------------------------
4454 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
4455 changes of various kinds. As a consequence, a big documentation update was
4456 needed. This delayed the release for rather longer than seemed good, especially
4457 in the light of a couple of (minor) security issues. Therefore, the changes
4458 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
4459 release. So 4.44 and 4.50 are in effect two different branches that both start
4462 I have left the 4.50 change log unchanged; it contains all the changes since
4463 4.43. The change log for 4.44 is below; many of its items are identical to
4464 those for 4.50. This seems to be the most sensible way to preserve the
4465 historical information.
4471 1. Minor wording change to the doc/README.SIEVE file.
4473 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
4474 computation of the current number of files was incorrect.
4476 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4477 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4480 4. Give more explanation in the error message when the command for a transport
4481 filter fails to execute.
4483 5. There are several places where Exim runs a non-Exim command in a
4484 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4485 was being done only for the command run by the queryprogram router. It is
4486 now done for all such subprocesses. The other cases are: ${run, transport
4487 filters, and the commands run by the lmtp and pipe transports.
4489 6. Added CONFIGURE_GROUP build-time option.
4491 7. Some older OS have a limit of 256 on the maximum number of file
4492 descriptors. Exim was using setrlimit() to set 1000 as a large value
4493 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4494 systems. I've change it so that if it can't get 1000, it tries for 256.
4496 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4497 was an oversight, and furthermore, ever since the addition of extra
4498 controls (e.g. 4.43/32), the checks on when to allow different forms of
4499 "control" were broken. There should now be diagnostics for all cases when a
4500 control that does not make sense is encountered.
4502 9. Added the /retain_sender option to "control=submission".
4504 10. $recipients is now available in the predata ACL (oversight).
4506 11. Tidy the search cache before the fork to do a delivery from a message
4507 received from the command line. Otherwise the child will trigger a lookup
4508 failure and thereby defer the delivery if it tries to use (for example) a
4509 cached ldap connection that the parent has called unbind on.
4511 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4512 of $address_data from the recipient verification was clobbered by the
4513 sender verification.
4515 13. The value of address_data from a sender verification is now available in
4516 $sender_address_data in subsequent conditions in the ACL statement.
4518 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
4520 15. Added a new option "connect=<time>" to callout options, to set a different
4523 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4524 was its contents. (It was OK if the option was not defined at all.)
4526 17. A "Completed" log line is now written for messages that are removed from
4527 the spool by the -Mrm option.
4529 18. New variables $sender_verify_failure and $recipient_verify_failure contain
4530 information about exactly what failed.
4532 19. Added -dd to debug only the daemon process.
4534 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
4535 handles timeouts, both on the server side and network timeouts. Renamed the
4536 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
4538 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
4539 It is now set to "smtps".
4541 22. $host_address is now set to the target address during the checking of
4542 ignore_target_hosts.
4544 23. When checking ignore_target_hosts for an ipliteral router, no host name was
4545 being passed; this would have caused $sender_host_name to have been used if
4546 matching the list had actually called for a host name (not very likely,
4547 since this list is usually IP addresses). A host name is now passed as
4550 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4551 code that specifies a non-restarting handler (typically sigaction() in
4552 modern systems) in an attempt to fix a rare and obscure crash bug.
4554 25. Narrowed the window for a race in the daemon that could cause it to ignore
4555 SIGCHLD signals. This is not a major problem, because they are used only to
4556 wake it up if nothing else does.
4558 26. A malformed maildirsize file could cause Exim to calculate negative values
4559 for the mailbox size or file count. Odd effects could occur as a result.
4560 The maildirsize information is now recalculated if the size or filecount
4563 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4564 support for a long time. Removed HAVE_SYS_VFS_H.
4566 28. Installed the latest version of exipick from John Jetmore.
4568 29. In an address list, if the pattern was not a regular expression, an empty
4569 subject address (from a bounce message) matched only if the pattern was an
4570 empty string. Non-empty patterns were not even tested. This was the wrong
4571 because it is perfectly reasonable to use an empty address as part of a
4572 database query. An empty address is now tested by patterns that are
4573 lookups. However, all the other forms of pattern expect the subject to
4574 contain a local part and a domain, and therefore, for them, an empty
4575 address still always fails if the pattern is not itself empty.
4577 30. Exim went into a mad DNS loop when attempting to do a callout where the
4578 host was specified on an smtp transport, and looking it up yielded more
4579 than one IP address.
4581 31. Re-factored the code for checking spool and log partition space into a
4582 function that finds that data and another that does the check. The former
4583 is then used to implement four new variables: $spool_space, $log_space,
4584 $spool_inodes, and $log_inodes.
4586 32. The RFC2047 encoding function was originally intended for short strings
4587 such as real names; it was not keeping to the 75-character limit for
4588 encoded words that the RFC imposes. It now respects the limit, and
4589 generates multiple encoded words if necessary. To be on the safe side, I
4590 have increased the buffer size for the ${rfc2047: expansion operator from
4593 33. It is now permitted to omit both strings after an "if" condition; if the
4594 condition is true, the result is "true". As before, when the second string
4595 is omitted, a false condition yields an empty string. This makes it less
4596 cumbersome to write custom ACL and router conditions.
4598 34. Failure to deliver a bounce message always caused it to be frozen, even if
4599 there was an errors_to setting on the router. The errors_to setting is now
4602 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4603 canonical form (fully expanded) before being placed in
4604 $sender_host_address.
4606 36. The table in the code that translates DNS record types into text (T_A to
4607 "A" for instance) was missing entries for NS and CNAME. It is just possible
4608 that this could have caused confusion if both these types were looked up
4609 for the same domain, because the text type is used as part of Exim's
4610 per-process caching. But the chance of anyone hitting this buglet seems
4613 37. The dnsdb lookup has been extended in a number of ways.
4615 (1) There is a new type, "zns", which walks up the domain tree until it
4616 finds some nameserver records. It should be used with care.
4618 (2) There is a new type, "mxh", which is like "mx" except that it returns
4619 just the host names, not the priorities.
4621 (3) It is now possible to give a list of domains (or IP addresses) to be
4622 looked up. The behaviour when one of the lookups defers can be
4623 controlled by a keyword.
4625 (4) It is now possible to specify the separator character for use when
4626 multiple records are returned.
4628 38. The dnslists ACL condition has been extended: it is now possible to supply
4629 a list of IP addresses and/or domains to be looked up in a particular DNS
4632 39. Added log_selector=+queue_time_overall.
4634 40. When running the queue in the test harness, wait just a tad after forking a
4635 delivery process, to get repeatability of debugging output.
4637 41. Include certificate and key file names in error message when GnuTLS fails
4638 to set them up, because the GnuTLS error message doesn't include the name
4639 of the failing file when there is a problem reading it.
4641 42. Allow both -bf and -bF in the same test run.
4643 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
4645 44. The "Exiscan patch" is now merged into the mainline Exim source.
4647 45. Sometimes the final signoff response after QUIT could fail to get
4648 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4649 < 0 before doing a fflush(). This bug looks as though it goes back to the
4650 introduction of TLS in release 3.20, but "sometimes" must have been rare
4651 because the tests only now provoked it.
4653 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
4654 (this can affect the format of dates).
4656 47. exim_tidydb, when checking for the continued existence of a message for
4657 which it has found a message-specific retry record, was not finding
4658 messages that were in split spool directories. Consequently, it was
4659 deleting retry records that should have stayed in existence.
4661 48. Steve fixed some bugs in eximstats.
4663 49. The SPA authentication driver was not abandoning authentication and moving
4664 on to the next authenticator when an expansion was forced to fail,
4665 contradicting the general specification for all authenticators. Instead it
4666 was generating a temporary error. It now behaves as specified.
4668 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
4669 (the order specifies the preference for clients). The order is now AES256,
4670 AES128, 3DES, ARCFOUR128.
4672 51. Small patch to Sieve code - explicitly set From: when generating an
4675 52. Exim crashed if a remote delivery caused a very long error message to be
4676 recorded - for instance if somebody sent an entire SpamAssassin report back
4677 as a large number of 550 error lines. This bug was coincidentally fixed by
4678 increasing the size of one of Exim's internal buffers (big_buffer) that
4679 happened as part of the Exiscan merge. However, to be on the safe side, I
4680 have made the code more robust (and fixed the comments that describe what
4683 53. Now that there can be additional text after "Completed" in log lines (if
4684 the queue_time_overall log selector is set), a one-byte patch to exigrep
4685 was needed to allow it to recognize "Completed" as not the last thing in
4688 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
4689 patch that reportedly fixes this has been added. I am not expert enough to
4690 create a test for it. This is what the patch creator wrote:
4692 "I found a little strange behaviour of ldap code when working with
4693 Windows 2003 AD Domain, where users was placed in more than one
4694 Organization Units. When I tried to give exim partial DN, the exit code
4695 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
4696 But simultaneously result of request was absolutely normal ldap result,
4697 so I produce this patch..."
4699 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
4700 REFERENCE, so I have modified the code to exclude the patch when that macro
4703 55. Some experimental protocols are using DNS PTR records for new purposes. The
4704 keys for these records are domain names, not reversed IP addresses. The
4705 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4706 leaves it alone. Component reversal etc. now happens only for IP addresses.
4709 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
4711 57. Double the size of the debug message buffer (to 2048) so that more of very
4712 long debug lines gets shown.
4714 58. The exicyclog utility now does better if the number of log files to keep
4715 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
4717 59. Two changes related to the smtp_active_hostname option:
4719 (1) $smtp_active_hostname is now available as a variable.
4720 (2) The default for smtp_banner uses $smtp_active_hostname instead
4721 of $primary_hostname.
4723 60. The host_aton() function is supposed to be passed a string that is known
4724 to be a valid IP address. However, in the case of IPv6 addresses, it was
4725 not checking this. This is a hostage to fortune. Exim now panics and dies
4726 if the condition is not met. A case was found where this could be provoked
4727 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4728 components; fortuitously, this particular loophole had already been fixed
4729 by change 4.50/55 above.
4731 If there are any other similar loopholes, the new check in host_aton()
4732 itself should stop them being exploited. The report I received stated that
4733 data on the command line could provoke the exploit when Exim was running as
4734 exim, but did not say which command line option was involved. All I could
4735 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4736 running as the user.
4739 61. There was a buffer overflow vulnerability in the SPA authentication code
4740 (which came originally from the Samba project). I have added a test to the
4741 spa_base64_to_bits() function which I hope fixes it.
4744 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
4745 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
4747 63. The daemon start-up calls getloadavg() while still root for those OS that
4748 need the first call to be done as root, but it missed one case: when
4749 deliver_queue_load_max is set with deliver_drop_privilege. This is
4750 necessary for the benefit of the queue runner, because there is no re-exec
4751 when deliver_drop_privilege is set.
4753 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
4754 This has been fixed.
4756 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4757 was in use, was not putting the data itself into the right store pool;
4758 consequently, it could be overwritten for a subsequent message in the same
4759 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4762 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
4764 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
4765 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
4766 the function were treating the return as a boolean value, which happened to
4767 work because 0=false and not-0=true, but is not correct code.
4769 68. The host_aton() function was not handling scoped IPv6 addresses (those
4770 with, for example, "%eth0" on the end) correctly.
4772 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
4773 negated items (that is, ~something) in unsigned ints. Some compilers
4774 apparently mutter when there is no cast.
4776 70. If an address verification called from an ACL failed, and did not produce a
4777 user-specific message (i.e. there was only a "system" message), nothing was
4778 put in $acl_verify_message. In this situation, it now puts the system
4781 71. Change 4.23/11 added synchronization checking at the start of an SMTP
4782 session; change 4.31/43 added the unwanted input to the log line - except
4783 that it did not do this in the start of session case. It now does.
4785 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
4786 This is wrong and can cause the other end to generate a synchronization
4787 error if it is another Exim or anything else that does the synchronization
4788 check. A QUIT command is no longer sent after a timeout.
4790 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
4791 during host lookups.
4793 74. The defer_ok option of callout verification was not working if it was used
4794 when verifying addresses in header lines, that is, for this case:
4796 verify = header_sender/callout=defer_ok
4798 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
4799 those file descriptors could be used for SMTP connections. If anything
4800 wrote to stderr (the example that came up was "warn" in embedded Perl), it
4801 could be sent to the SMTP client, causing chaos. The daemon now opens
4802 stdin, stdout, and stderr to /dev/null when it puts itself into the
4805 76. Arrange for output from Perl's "warn" command to be written to Exim's main
4806 log by default. The user can override this with suitable Perl magic.
4808 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
4809 the log message when discard triggers, was not working for the DATA ACL or
4810 for the non-SMTP ACL.
4812 78. Error message wording change in sieve.c.
4814 79. If smtp_accept_max_per_host was set, the number of connections could be
4815 restricted to fewer than expected, because the daemon was trying to set up
4816 a new connection before checking whether the processes handling previous
4817 connections had finished. The check for completed processes is now done
4818 earlier. On busy systems, this bug wouldn't be noticed because something
4819 else would have woken the daemon, and it would have reaped the completed
4822 80. If a message was submitted locally by a user whose login name contained one
4823 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
4824 It caused a spool format error. I have fixed the spool reading code. A
4825 related problem was that the "from" clause in the Received: line became
4826 illegal because of the space(s). It is now covered by ${quote_local_part.
4828 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
4831 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
4833 83. Patch from Sieve maintainer to fix unterminated string problem in
4834 "vacation" handling.
4836 84. Some minor changes to the Linux configuration files to help with other
4837 OS variants using glibc.
4839 85. One more patch for Sieve to update vacation handling to latest spec.
4842 ----------------------------------------------------
4843 See the note above about the 4.44 and 4.50 releases.
4844 ----------------------------------------------------
4850 1. Change 4.43/35 introduced a bug that caused file counts to be
4851 incorrectly computed when quota_filecount was set in an appendfile
4854 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4855 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4858 3. Give more explanation in the error message when the command for a transport
4859 filter fails to execute.
4861 4. There are several places where Exim runs a non-Exim command in a
4862 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4863 was being done only for the command run by the queryprogram router. It is
4864 now done for all such subprocesses. The other cases are: ${run, transport
4865 filters, and the commands run by the lmtp and pipe transports.
4867 5. Some older OS have a limit of 256 on the maximum number of file
4868 descriptors. Exim was using setrlimit() to set 1000 as a large value
4869 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4870 systems. I've change it so that if it can't get 1000, it tries for 256.
4872 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4873 was an oversight, and furthermore, ever since the addition of extra
4874 controls (e.g. 4.43/32), the checks on when to allow different forms of
4875 "control" were broken. There should now be diagnostics for all cases when a
4876 control that does not make sense is encountered.
4878 7. $recipients is now available in the predata ACL (oversight).
4880 8. Tidy the search cache before the fork to do a delivery from a message
4881 received from the command line. Otherwise the child will trigger a lookup
4882 failure and thereby defer the delivery if it tries to use (for example) a
4883 cached ldap connection that the parent has called unbind on.
4885 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4886 of $address_data from the recipient verification was clobbered by the
4887 sender verification.
4889 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4890 was its contents. (It was OK if the option was not defined at all.)
4892 11. A "Completed" log line is now written for messages that are removed from
4893 the spool by the -Mrm option.
4895 12. $host_address is now set to the target address during the checking of
4896 ignore_target_hosts.
4898 13. When checking ignore_target_hosts for an ipliteral router, no host name was
4899 being passed; this would have caused $sender_host_name to have been used if
4900 matching the list had actually called for a host name (not very likely,
4901 since this list is usually IP addresses). A host name is now passed as
4904 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4905 code that specifies a non-restarting handler (typically sigaction() in
4906 modern systems) in an attempt to fix a rare and obscure crash bug.
4908 15. Narrowed the window for a race in the daemon that could cause it to ignore
4909 SIGCHLD signals. This is not a major problem, because they are used only to
4910 wake it up if nothing else does.
4912 16. A malformed maildirsize file could cause Exim to calculate negative values
4913 for the mailbox size or file count. Odd effects could occur as a result.
4914 The maildirsize information is now recalculated if the size or filecount
4917 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4918 support for a long time. Removed HAVE_SYS_VFS_H.
4920 18. Updated exipick to current release from John Jetmore.
4922 19. Allow an empty sender to be matched against a lookup in an address list.
4923 Previously the only cases considered were a regular expression, or an
4926 20. Exim went into a mad DNS lookup loop when doing a callout where the
4927 host was specified on the transport, if the DNS lookup yielded more than
4930 21. The RFC2047 encoding function was originally intended for short strings
4931 such as real names; it was not keeping to the 75-character limit for
4932 encoded words that the RFC imposes. It now respects the limit, and
4933 generates multiple encoded words if necessary. To be on the safe side, I
4934 have increased the buffer size for the ${rfc2047: expansion operator from
4937 22. Failure to deliver a bounce message always caused it to be frozen, even if
4938 there was an errors_to setting on the router. The errors_to setting is now
4941 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4942 canonical form (fully expanded) before being placed in
4943 $sender_host_address.
4945 24. Updated eximstats to version 1.33
4947 25. Include certificate and key file names in error message when GnuTLS fails
4948 to set them up, because the GnuTLS error message doesn't include the name
4949 of the failing file when there is a problem reading it.
4951 26. Expand error message when OpenSSL has problems setting up cert/key files.
4954 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
4955 (this can affect the format of dates).
4957 28. exim_tidydb, when checking for the continued existence of a message for
4958 which it has found a message-specific retry record, was not finding
4959 messages that were in split spool directories. Consequently, it was
4960 deleting retry records that should have stayed in existence.
4962 29. eximstats updated to version 1.35
4963 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
4964 1.35 - bugfix such that pie charts by volume are generated correctly
4966 30. The SPA authentication driver was not abandoning authentication and moving
4967 on to the next authenticator when an expansion was forced to fail,
4968 contradicting the general specification for all authenticators. Instead it
4969 was generating a temporary error. It now behaves as specified.
4971 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
4972 (the order specifies the preference for clients). The order is now AES256,
4973 AES128, 3DES, ARCFOUR128.
4975 31. Small patch to Sieve code - explicitly set From: when generating an
4978 32. Exim crashed if a remote delivery caused a very long error message to be
4979 recorded - for instance if somebody sent an entire SpamAssassin report back
4980 as a large number of 550 error lines. This bug was coincidentally fixed by
4981 increasing the size of one of Exim's internal buffers (big_buffer) that
4982 happened as part of the Exiscan merge. However, to be on the safe side, I
4983 have made the code more robust (and fixed the comments that describe what
4986 33. Some experimental protocols are using DNS PTR records for new purposes. The
4987 keys for these records are domain names, not reversed IP addresses. The
4988 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4989 leaves it alone. Component reversal etc. now happens only for IP addresses.
4992 34. The host_aton() function is supposed to be passed a string that is known
4993 to be a valid IP address. However, in the case of IPv6 addresses, it was
4994 not checking this. This is a hostage to fortune. Exim now panics and dies
4995 if the condition is not met. A case was found where this could be provoked
4996 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4997 components; fortuitously, this particular loophole had already been fixed
4998 by change 4.50/55 or 4.44/33 above.
5000 If there are any other similar loopholes, the new check in host_aton()
5001 itself should stop them being exploited. The report I received stated that
5002 data on the command line could provoke the exploit when Exim was running as
5003 exim, but did not say which command line option was involved. All I could
5004 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
5005 running as the user.
5008 35. There was a buffer overflow vulnerability in the SPA authentication code
5009 (which came originally from the Samba project). I have added a test to the
5010 spa_base64_to_bits() function which I hope fixes it.
5013 36. The daemon start-up calls getloadavg() while still root for those OS that
5014 need the first call to be done as root, but it missed one case: when
5015 deliver_queue_load_max is set with deliver_drop_privilege. This is
5016 necessary for the benefit of the queue runner, because there is no re-exec
5017 when deliver_drop_privilege is set.
5019 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
5020 was in use, was not putting the data itself into the right store pool;
5021 consequently, it could be overwritten for a subsequent message in the same
5022 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
5025 38. Sometimes the final signoff response after QUIT could fail to get
5026 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
5027 < 0 before doing a fflush(). This bug looks as though it goes back to the
5028 introduction of TLS in release 3.20, but "sometimes" must have been rare
5029 because the tests only now provoked it.
5035 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
5036 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
5037 Now it returns an integer. A number of calls were still expecting a T/F
5038 return. Fortuitously, in all cases, the tests worked in OK situations,
5039 which is the norm. However, things would have gone wrong on any write
5040 failures on the smtp file descriptor. This function is used when sending
5041 messages over SMTP and also when doing verify callouts.
5043 2. When Exim is called to do synchronous delivery of a locally submitted
5044 message (the -odf or -odi options), it no longer closes stderr before doing
5047 3. Implemented the mua_wrapper option.
5049 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
5051 5. Implemented the functions header_remove(), header_testname(),
5052 header_add_at_position(), and receive_remove_recipient(), and exported them
5055 6. If an ACL "warn" statement specified the addition of headers, Exim already
5056 inserted X-ACL-Warn: at the start if there was no header name. However, it
5057 was not making this test for the second and subsequent header lines if
5058 there were newlines in the string. This meant that an invalid header could
5059 be inserted if Exim was badly configured.
5061 7. Allow an ACL "warn" statement to add header lines at the start or after all
5062 the Received: headers, as well as at the end.
5064 8. Added the rcpt_4xx retry error code.
5066 9. Added postmaster_mailfrom=xxx to callout verification option.
5068 10. Added mailfrom=xxxx to the callout verification option, for verify=
5071 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
5072 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
5074 12. Inserted some casts to stop certain compilers warning when using pointer
5075 differences as field lengths or precisions in printf-type calls (mostly
5076 affecting debugging statements).
5078 13. Added optional readline() support for -be (dynamically loaded).
5080 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
5081 same clock tick as a message's arrival, so that its received time was the
5082 same as the "first fail" time on the retry record, and that message
5083 remained on the queue past the ultimate address timeout, every queue runner
5084 would try a delivery (because it was past the ultimate address timeout) but
5085 after another failure, the ultimate address timeout, which should have then
5086 bounced the address, did not kick in. This was a "< instead of <=" error;
5087 in most cases the first failure would have been in the next clock tick
5088 after the received time, and all would be well.
5090 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
5091 being recognized when the domain list was tested by the match_domain
5092 condition in an expansion string.
5094 16. Added the ${str2b64: operator.
5096 17. Exim was always calling setrlimit() to set a large limit for the number of
5097 processes, without checking whether the existing limit was already
5098 adequate. (It did check for the limit on file descriptors.) Furthermore,
5099 errors from getrlimit() and setrlimit() were being ignored. Now they are
5100 logged to the main and panic logs, but Exim does carry on, to try to do its
5101 job under whatever limits there are.
5103 18. Imported PCRE 5.0.
5105 19. Trivial typo in log message " temporarily refused connection" (the leading
5108 20. If the log selector return_path_on_delivery was set and an address was
5109 redirected to /dev/null, the delivery process crashed because it assumed
5110 that a return path would always be set for a "successful" delivery. In this
5111 case, the whole delivery is bypassed as an optimization, and therefore no
5114 21. Internal re-arrangement: the function for sending a challenge and reading
5115 a response while authentication was assuming a zero-terminated challenge
5116 string. It's now changed to take a pointer and a length, to allow for
5117 binary data in such strings.
5119 22. Added the cyrus_sasl authenticator (code supplied by MBM).
5121 23. Exim was not respecting finduser_retries when seeking the login of the
5122 uid under which it was called; it was always trying 10 times. (The default
5123 setting of finduser_retries is zero.) Also, it was sleeping after the final
5124 failure, which is pointless.
5126 24. Implemented tls_on_connect_ports.
5128 25. Implemented acl_smtp_predata.
5130 26. If the domain in control=submission is set empty, Exim assumes that the
5131 authenticated id is a complete email address when it generates From: or
5132 Sender: header lines.
5134 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
5135 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
5136 chown and chgrp in /bin and hostname in /usr/bin.
5138 28. Exim was keeping the "process log" file open after each use, just as it
5139 does for the main log. This opens the possibility of it remaining open for
5140 long periods when the USR1 signal hits a daemon. Occasional processlog
5141 errors were reported, that could have been caused by this. Anyway, it seems
5142 much more sensible not to leave this file open at all, so that is what now
5145 29. The long-running daemon process does not normally write to the log once it
5146 has entered its main loop, and it closes the log before doing so. This is
5147 so that log files can straightforwardly be renamed and moved. However,
5148 there are a couple of unusual error situations where the daemon does write
5149 log entries, and I had neglected to close the log afterwards.
5151 30. The text of an SMTP error response that was received during a remote
5152 delivery was being truncated at 512 bytes. This is too short for some of
5153 the long messages that one sometimes sees. I've increased the limit to
5156 31. It is now possible to make retry rules that apply only when a message has a
5157 specific sender, in particular, an empty sender.
5159 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
5160 it possible to be selective about when SMTP synchronization is enforced.
5162 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
5164 32. Implemented hosts_connection_nolog.
5166 33. Added an ACL for QUIT.
5168 34. Setting "delay_warning=" to disable warnings was not working; it gave a
5171 35. Added mailbox_size and mailbox_filecount to appendfile.
5173 36. Added control = no_multiline_responses to ACLs.
5175 37. There was a bug in the logic of the code that waits for the clock to tick
5176 in the case where the clock went backwards by a substantial amount such
5177 that the microsecond fraction of "now" was more than the microsecond
5178 fraction of "then" (but the whole seconds number was less).
5180 38. Added support for the libradius Radius client library this is found on
5181 FreeBSD (previously only the radiusclient library was supported).
5187 1. When certain lookups returned multiple values in the form name=value, the
5188 quoting of the values was not always being done properly. Specifically:
5189 (a) If the value started with a double quote, but contained no whitespace,
5191 (b) If the value contained whitespace other than a space character (i.e.
5192 tabs or newlines or carriage returns) it was not quoted.
5193 This fix has been applied to the mysql and pgsql lookups by writing a
5194 separate quoting function and calling it from the lookup code. The fix
5195 should probably also be applied to nisplus, ibase and oracle lookups, but
5196 since I cannot test any of those, I have not disturbed their existing code.
5198 2. A hit in the callout cache for a specific address caused a log line with no
5199 reason for rejecting RCPT. Now it says "Previous (cached) callout
5200 verification failure".
5202 3. There was an off-by-one bug in the queryprogram router. An over-long
5203 return line was truncated at 256 instead of 255 characters, thereby
5204 overflowing its buffer with the terminating zero. As well as fixing this, I
5205 have increased the buffer size to 1024 (and made a note to document this).
5207 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
5208 when Exim is waiting for an SMTP response from a remote server, Exim
5209 restarts its select() call on the socket, thereby resetting its timeout.
5210 This is not a problem when such interrupts are rare. Somebody set up a cron
5211 job to run exiwhat every 2 minutes, which is less than the normal select()
5212 timeout (5 or 10 minutes). This meant that the select() timeout never
5213 kicked in because it was always reset. I have fixed this by comparing the
5214 time when an interrupt arrives with the time at the start of the first call
5215 to select(). If more time than the timeout has elapsed, the interrupt is
5216 treated as a timeout.
5218 5. Some internal re-factoring in preparation for the addition of Sieve
5219 extensions (by MH). In particular, the "personal" test is moved to a
5220 separate function, and given an option for scanning Cc: and Bcc: (which is
5221 not set for Exim filters).
5223 6. When Exim created an email address using the login of the caller as the
5224 local part (e.g. when creating a From: or Sender: header line), it was not
5225 quoting the local part when it contained special characters such as @.
5227 7. Installed new OpenBSD configuration files.
5229 8. Reworded some messages for syntax errors in "and" and "or" conditions to
5230 try to make them clearer.
5232 9. Callout options, other than the timeout value, were being ignored when
5233 verifying sender addresses in header lines. For example, when using
5235 verify = header_sender/callout=no_cache
5237 the cache was (incorrectly) being used.
5239 10. Added a missing instance of ${EXE} to the exim_install script; this affects
5240 only the Cygwin environment.
5242 11. When return_path_on_delivery was set as a log selector, if different remote
5243 addresses in the same message used different return paths and parallel
5244 remote delivery occurred, the wrong values would sometimes be logged.
5245 (Whenever a remote delivery process finished, the return path value from
5246 the most recently started remote delivery process was logged.)
5248 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
5249 lines when AUTH and/or TLS are in use. This is the "received protocol"
5250 field. Exim used to use "asmtp" for authenticated SMTP, without any
5251 indication (in the protocol name) for TLS use. Now it follows the RFC and
5252 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
5253 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
5254 names appear in log lines as well as in Received: header lines.
5256 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
5257 extensions, and comparison tests, and to fix some bugs.
5259 14. Changes to the "personal" filter test:
5261 (1) The test was buggy in that it was just doing the equivalent of
5262 "contains" tests on header lines. For example, if a user's address was
5263 anne@some.where, the "personal" test would incorrectly be true for
5265 To: susanne@some.where
5267 This test is now done by extracting each address from the header in turn,
5268 and checking the entire address. Other tests that are part of "personal"
5269 are now done using regular expressions (for example, to check local parts
5270 of addresses in From: header lines).
5272 (2) The list of non-personal local parts in From: addresses has been
5273 extended to include "listserv", "majordomo", "*-request", and "owner-*",
5274 taken from the Sieve specification recommendations.
5276 (3) If the message contains any header line starting with "List-" it is
5277 treated as non-personal.
5279 (4) The test for "circular" in the Subject: header line has been removed
5280 because it now seems ill-conceived.
5282 15. Minor typos in src/EDITME comments corrected.
5284 16. Installed latest exipick from John Jetmore.
5286 17. If headers_add on a router specified a text string that was too long for
5287 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
5288 of string_sprintf() is now avoided.
5290 18. $message_body_size was not set (it was always zero) when running the DATA
5291 ACL and the local_scan() function.
5293 19. For the "mail" command in an Exim filter, no default was being set for
5294 the once_repeat time, causing a random time value to be used if "once" was
5295 specified. (If the value happened to be <= 0, no repeat happened.) The
5296 default is now 0s, meaning "never repeat". The "vacation" command was OK
5297 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
5298 (I found it when inspecting the code).
5300 20. There is now an overall timeout for performing a callout verification. It
5301 defaults to 4 times the callout timeout, which applies to individual SMTP
5302 commands during the callout. The overall timeout applies when there is more
5303 than one host that can be tried. The timeout is checked before trying the
5304 next host. This prevents very long delays if there are a large number of
5305 hosts and all are timing out (e.g. when the network connections are timing
5306 out). The value of the overall timeout can be changed by specifying an
5307 additional sub-option for "callout", called "maxwait". For example:
5309 verify = sender/callout=5s,maxwait=20s
5311 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
5312 to the end before writing, but this should make it even safer).
5314 22. Exim was forgetting that it had advertised PIPELINING for the second and
5315 subsequent messages on an SMTP connection. It was also not resetting its
5316 memory on STARTTLS and an internal HELO.
5318 23. When Exim logs an SMTP synchronization error within a session, it now
5319 records whether PIPELINING has been advertised or not.
5321 24. Added 3 instances of "(long int)" casts to time_t variables that were being
5322 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
5323 rather than long int.
5325 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
5327 26. Added the never_mail option to autoreply.
5333 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
5334 crash if the getsockname() call failed; this can happen if a connection is
5335 closed very soon after it is established. The problem was simply in the
5336 order in which certain operations were done, causing Exim to try to write
5337 to the SMTP stream before it had set up the file descriptor. The bug has
5338 been fixed by making things happen in the correct order.
5344 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
5345 before the connection was closed, thus losing the rejection response.
5347 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
5348 some early Solaris releases, but causes trouble in current releases where
5349 socklen_t is defined.
5351 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
5354 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
5357 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
5358 that releasing the top of it at the end releases what was used for sub-
5359 expansions (unless the block got too big). However, discard this block if
5360 the first thing is a variable or header, so that we can use its block when
5361 it is dynamic (useful for very large $message_headers, for example).
5363 6. Lookups now cache *every* query, not just the most recent. A new, separate
5364 store pool is used for this. It can be recovered when all lookup caches are
5365 flushed. Lookups now release memory at the end of their result strings.
5366 This has involved some general refactoring of the lookup sources.
5368 7. Some code has been added to the store_xxx() functions to reduce the amount
5369 of flapping under certain conditions.
5371 8. log_incoming_interface used to affect only the <= reception log lines. Now
5372 it causes the local interface and port to be added to several more SMTP log
5373 lines, for example "SMTP connection from", and rejection lines.
5375 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
5377 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
5379 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
5380 could be overwritten at the end of the current message (or the start of a
5381 new message if it was set in a HELO ACL). The value is now preserved for
5382 the duration of the SMTP connection.
5384 12. If a transport had a headers_rewrite setting, and a matching header line
5385 contained an unqualified address, that address was qualified, even if it
5386 did not match any rewriting rules. The underlying bug was that the values
5387 of the flags that permit the existence of unqualified sender and recipient
5388 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
5389 non-local messages, and by -bnq for local messages) were not being
5390 preserved with the message after it was received.
5392 13. When Exim was logging an SMTP synchronization error, it could sometimes log
5393 "next input=" as part of the text comprising the host identity instead of
5394 the correct text. The code was using the same buffer for two different
5395 strings. However, depending on which order the printing function evaluated
5396 its arguments, the bug did not always show up. Under Linux, for example, my
5397 test suite worked just fine.
5399 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
5400 doesn't work with some older versions of Perl. It has been changed to "my",
5401 which in any case is probably the better facility to use.
5403 15. A really picky compiler found some instances of statements for creating
5404 error messages that either had too many or two few arguments for the format
5407 16. The size of the buffer for calls to the DNS resolver has been increased
5408 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
5409 for addresses that have a lot of PTR records. This alleviates a problem; it
5410 does not fully solve it.
5412 17. A dnsdb lookup for PTR records that receives more data than will fit in the
5413 buffer now truncates the list and logs the incident, which is the same
5414 action as happens when Exim is looking up a host name and its aliases.
5415 Previously in this situation something unpredictable would happen;
5416 sometimes it was "internal error: store_reset failed".
5418 18. If a server dropped the connection unexpectedly when an Exim client was
5419 using GnuTLS and trying to read a response, the client delivery process
5420 crashed while trying to generate an error log message.
5422 19. If a "warn" verb in an ACL added multiple headers to a message in a single
5423 string, for example:
5425 warn message = H1: something\nH2: something
5427 the text was added as a single header line from Exim's point of view
5428 though it ended up OK in the delivered message. However, searching for the
5429 second and subsequent header lines using $h_h2: did not work. This has been
5430 fixed. Similarly, if a system filter added multiple headers in this way,
5431 the routers could not see them.
5433 20. Expanded the error message when iplsearch is called with an invalid key to
5434 suggest using net-iplsearch in a host list.
5436 21. When running tests using -bh, any delays imposed by "delay" modifiers in
5437 ACLs are no longer actually imposed (and a message to that effect is
5440 22. If a "gecos" field in a passwd entry contained escaped characters, in
5441 particular, if it contained a \" sequence, Exim got it wrong when building
5442 a From: or a Sender: header from that name. A second bug also caused
5443 incorrect handling when an unquoted " was present following a character
5444 that needed quoting.
5446 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
5447 was not being matched caselessly.
5449 24. Arranged for all hyphens in the exim.8 source to be escaped with
5452 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
5453 properly. Recipient callout cache records were still being keyed to include
5454 the sender, even when use_sender was set false. This led to far more
5455 callouts that were necessary. The sender is no longer included in the key
5456 when use_sender is false.
5458 26. Added "control = submission" modifier to ACLs.
5460 27. Added the ${base62d: operator to decode base 62 numbers.
5462 28. dnsdb lookups can now access SRV records.
5464 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
5465 the configuration file.
5467 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
5468 (-v) mode. This makes the output for a verbose queue run more intelligible.
5470 31. Added a use_postmaster feature to recipient callouts.
5472 32. Added the $body_zerocount variable, containing the number of binary zero
5473 bytes in the message body.
5475 33. The time of last modification of the "new" subdirectory is now used as the
5476 "mailbox time last read" when there is a quota error for a maildir
5479 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
5481 35. Added +ignore_unknown as a special item in host lists.
5483 36. Code for decoding IPv6 addresses in host lists is now included, even if
5484 IPv6 support is not being compiled. This fixes a bug in which an IPv6
5485 address was recognized as an IP address, but was then not correctly decoded
5486 into binary, causing unexpected and incorrect effects when compared with
5493 1. Very minor rewording of debugging text in manualroute to say "list of
5494 hosts" instead of "hostlist".
5496 2. If verify=header_syntax was set, and a header line with an unqualified
5497 address (no domain) and a large number of spaces between the end of the
5498 name and the colon was received, the reception process suffered a buffer
5499 overflow, and (when I tested it) crashed. This was caused by some obsolete
5500 code that should have been removed. The fix is to remove it!
5502 3. When running in the test harness, delay a bit after writing a bounce
5503 message to get a bit more predictability in the log output.
5505 4. Added a call to search_tidyup() just before forking a reception process. In
5506 theory, someone could use a lookup in the expansion of smtp_accept_max_
5507 per_host which, without the tidyup, could leave open a database connection.
5509 5. Added the variables $recipient_data and $sender_data which get set from a
5510 lookup success in an ACL "recipients" or "senders" condition, or a router
5511 "senders" option, similar to $domain_data and $local_part_data.
5513 6. Moved the writing of debug_print from before to after the "senders" test
5516 7. Change 4.31/66 (moving the time when the Received: is generated) caused
5517 problems for message scanning, either using a data ACL, or using
5518 local_scan() because the Received: header was not generated till after they
5519 were called (in order to set the time as the time of reception completion).
5520 I have revised the way this works. The header is now generated after the
5521 body is received, but before the ACL or local_scan() are called. After they
5522 are run, the timestamp in the header is updated.
5528 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
5529 before starting a queue runner without re-exec. This happened only when
5530 deliver_drop_privilege was set or when the Exim user was set to root. The
5531 effect of the bug was that timeouts during subsequent deliveries caused
5532 crashes instead of being properly handled. The handler is now left at its
5533 default (and expected) setting.
5535 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
5536 message, again when deliver_drop_privilege is set or Exim is run as root.
5537 The bug described in (1) was not present in this case, but the tidying up
5538 of the other signals was missing. I have made the two cases consistent.
5540 3. The ignore_target_hosts setting on a manualroute router was being ignored
5541 for hosts that were looked up using the /MX notation.
5543 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
5546 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
5547 operated on the sender address. After changing the $sender_address to <>
5548 for the sender address verify, Exim was re-instated it as the original
5549 (before rewriting) address, but remembering that it had rewritten it, so it
5550 wasn't rewriting it again. This bug also had the effect of breaking the
5551 sender address verification caching when the sender address was rewritten.
5553 6. The ignore_target_hosts option was being ignored by the ipliteral router.
5554 This has been changed so that if the ip literal address matches
5555 ignore_target_hosts, the router declines.
5557 7. Added expansion conditions match_domain, match_address, and match_local_
5558 part (NOT match_host).
5560 8. The placeholder for the Received: header didn't have a length field set.
5562 9. Added code to Exim itself and to exim_lock to test for a specific race
5563 condition that could lead to file corruption when using MBX delivery. The
5564 issue is with the lockfile that is created in /tmp. If this file is removed
5565 after a process has opened it but before that process has acquired a lock,
5566 there is the potential for a second process to recreate the file and also
5567 acquire a lock. This could lead to two Exim processes writing to the file
5568 at the same time. The added code performs the same test as UW imapd; it
5569 checks after acquiring the lock that its file descriptor still refers to
5570 the same named file.
5572 10. The buffer for building added header lines was of fixed size, 8192 bytes.
5573 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
5576 11. Added the smtp_active_hostname option. If used, this will typically be made
5577 to depend on the incoming interface address. Because $interface_address is
5578 not set up until the daemon has forked a reception process, error responses
5579 that can happen earlier (such as "too many connections") no longer contain
5582 12. If an expansion in a condition on a "warn" statement fails because a lookup
5583 defers, the "warn" statement is abandoned, and the next ACL statement is
5584 processed. Previously this caused the whole ACL to be aborted.
5586 13. Added the iplsearch lookup type.
5588 14. Added ident_timeout as a log selector.
5590 15. Added tls_certificate_verified as a log selector.
5592 16. Added a global option tls_require_ciphers (compare the smtp transport
5593 option of the same name). This controls incoming TLS connections.
5595 17. I finally figured out how to make tls_require_ciphers do a similar thing
5596 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
5597 before starting the TLS session.
5599 18. Tabs are now shown as \t in -bP output.
5601 19. If the log selector return_path_on_delivery was set, Exim crashed when
5602 bouncing a message because it had too many Received: header lines.
5604 20. If two routers both had headers_remove settings, and the first one included
5605 a superfluous trailing colon, the final name in the first list and the
5606 first name in the second list were incorrectly joined into one item (with a
5607 colon in the middle).
5613 1. Added -C and -D options to the exinext utility, mainly to make it easier
5614 to include in the automated testing, but these could be helpful when
5615 multiple configurations are in use.
5617 2. The exinext utility was not formatting the output nicely when there was
5618 an alternate port involved in the retry record key, nor when there was a
5619 message id as well (for retries that were specific to a specific message
5620 and a specific host). It was also confused by IPv6 addresses, because of
5621 the additional colons they contain. I have fixed the IPv4 problem, and
5622 patched it up to do a reasonable job for IPv6.
5624 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
5625 delivery, the log line now contains "pipelined" if PIPELINING was used.
5627 4. An SMTP transport process used to panic and die if the bind() call to set
5628 an explicit outgoing interface failed. This has been changed; it is now
5629 treated in the same way as a connect() failure.
5631 5. A reference to $sender_host_name in the part of a conditional expansion
5632 that was being skipped was still causing a DNS lookup. This no longer
5635 6. The def: expansion condition was not recognizing references to header lines
5636 that used bh_ and bheader_.
5638 7. Added the _cache feature to named lists.
5640 8. The code for checking quota_filecount in the appendfile transport was
5641 allowing one more file than it should have been.
5643 9. For compatibility with Sendmail, the command line option
5651 and sets the incoming protocol and host name (for trusted callers). The
5652 host name and its colon can be omitted when only the protocol is to be set.
5653 Note the Exim already has two private options, -pd and -ps, that refer to
5654 embedded Perl. It is therefore impossible to set a protocol value of "d" or
5655 "s", but I don't think that's a major issue.
5657 10. A number of refactoring changes to the code, none of which should affect
5660 (a) The number of logging options was getting close to filling up the
5661 32-bit word that was used as a bit map. I have split them into two classes:
5662 those that are passed in the argument to log_write(), and those that are
5663 only ever tested independently outside of that function. These are now in
5664 separate 32-bit words, so there is plenty of room for expansion again.
5665 There is no change in the user interface or the logging behaviour.
5667 (b) When building, for example, log lines, the code previously used a
5668 macro that called string_cat() twice, in order to add two strings. This is
5669 not really sufficiently general. Furthermore, there was one instance where
5670 it was actually wrong because one of the argument was used twice, and in
5671 one call a function was used. (As it happened, calling the function twice
5672 did not affect the overall behaviour.) The macro has been replaced by a
5673 function that can join an arbitrary number of extra strings onto a growing
5676 (c) The code for expansion conditions now uses a table and a binary chop
5677 instead of a serial search (which was left over from when there were very
5678 few conditions). Also, it now recognizes conditions like "pam" even when
5679 the relevant support is not compiled in: a suitably worded error message is
5680 given if an attempt is made to use such a condition.
5682 11. Added ${time_interval:xxxxx}.
5684 12. A bug was causing one of the ddress fields not to be passed back correctly
5685 from remote delivery subprocesses. The field in question was not being
5686 subsequently used, so this caused to problems in practice.
5688 13. Added new log selectors queue_time and deliver_time.
5690 14. Might have fixed a bug in maildirsizefile handling that threw up
5691 "unexpected character" debug warnings, and recalculated the data
5692 unnecessarily. In any case, I expanded the warning message to give more
5695 15. Added the message "Restricted characters in address" to the statements in
5696 the default ACL that block characters like @ and % in local parts.
5698 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
5699 Three changes have been made:
5701 (a) There was a serious bug; a negative response to MAIL caused the whole
5702 recipient domain to be cached as invalid, thereby blocking all messages
5703 to all local parts at the same domain, from all senders. This bug has
5704 been fixed. The domain is no longer cached after a negative response to
5705 MAIL if the sender used is not empty.
5707 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
5710 (c) A new callout option, "use_sender" has been added for people who want
5711 the modified behaviour.
5717 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
5720 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
5721 indeed breaks things for older releases.
5723 3. Added additional logging to the case where there is a problem reading data
5724 from a filter that is running in a subprocess using a pipe, in order to
5725 try to track down a specific problem.
5727 4. Testing facility fudge: when running in the test harness and attempting
5728 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
5729 getting "No route to host". Convert this to a timeout.
5731 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
5734 6. Some OS don't have socklen_t but use size_t instead. This affects the
5735 fifth argument of getsockopt() amongst other things. This is now
5736 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
5737 can be set for individual OS. I have set it for SunOS5, OSF1, and
5738 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
5739 some earlier ones do not.
5741 7. Change 4.30/15 was not doing the test caselessly.
5743 8. The standard form for an IPv6 address literal was being rejected by address
5744 parsing in, for example, MAIL and RCPT commands. An example of this kind of
5745 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5746 this, as well as the form without the "IPv6" on the front (but only when
5747 address literals are enabled, of course).
5749 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
5751 10. Exim crashed if a message with an empty sender address specified by -f
5752 encountered a router with an errors_to setting. This could be provoked only
5753 by a command such as
5757 where an empty string was supplied; "<>" did not hit this bug.
5759 11. Installed PCRE release 4.5.
5761 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
5762 remained set. It is now erased.
5764 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
5765 times from message ids (which are base 36 rather than the normal 62).
5767 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
5768 were being counted as actual protocol errors, and logged if the log
5769 selector +smtp_protocol_error was set. One cannot be perfect in this test,
5770 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
5771 and DATA following a set of rejected RCPTs do not count as protocol errors.
5772 In other words, Exim assumes they were pipelined, though this may not
5773 actually be the case. Of course, in all cases the client gets an
5774 appropriate error code.
5776 15. If a lookup fails in an ACL condition, a message about the failure may
5777 be available; it is used if testing the ACL cannot continue, because most
5778 such messages specify what the cause of the deferral is. However, some
5779 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
5780 that caused an old message to be retained and used if a later statement
5781 caused a defer, replacing the real cause of the deferral.
5783 16. If an IP address had so many PTR records that the DNS lookup buffer
5784 was not large enough to hold them, Exim could crash while trying to process
5785 the truncated data. It now detects and logs this case.
5787 17. Further to 4.21/58, another change has been made: if (and only if) the
5788 first line of a message (the first header line) ends with CRLF, a bare LF
5789 in a subsequent header line has a space inserted after it, so as not to
5790 terminate the header.
5792 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
5793 unnecessarily, used atoi() instead of strtol(), and didn't check the
5794 termination when getting file sizes from file names by regex.
5796 19. Completely re-implemented the support for maildirsize files, in the light
5797 of a number of problems with the previous contributed implementation
5798 (4.30/29). In particular:
5800 . If the quota is zero, the maildirsize file is maintained, but no quota is
5803 . If the maildir directory does not exist, it is created before any attempt
5804 to write a maildirsize file.
5806 . The quota value in the file is just a cache; if the quota is changed in
5807 the transport, the new value overrides.
5809 . A regular expression is available for excluding directories from the
5812 20. The autoreply transport checks the characters in options that define the
5813 message's headers; it allows continued headers, but it was checking with
5814 isspace() after an embedded newline instead of explicitly looking for a
5817 21. If all the "regular" hosts to which an address was routed had passed their
5818 expiry times, and had not reached their retry times, the address was
5819 bounced, even if fallback hosts were defined. Now Exim should go on to try
5822 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
5823 equivalent code in the SMTP transport. Some hosts send humungous responses
5824 to HELO/EHLO, more than 1024 it seems.
5826 23. Refactoring: code in filter.c used (void *) for "any old type" but this
5827 gives compiler warnings in some environments. I've now done it "properly",
5830 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
5831 (because of problems with the built-in one) was declared to return uschar *
5832 instead of char *, causing compiler failure.
5834 25. Fixed a file descriptor leak when processing alias/forward files.
5836 26. Fixed a minor format string issue in dbfn.c.
5838 27. Typo in exim.c: ("dmbnz" for "dbmnz").
5840 28. If a filter file refered to $h_xxx or $message_headers, and the headers
5841 contained RFC 2047 "words", Exim's memory could, under certain conditions,
5844 29. When a sender address is verified, it is cached, to save repeating the test
5845 when there is more than one recipient in a message. However, when the
5846 verification involves a callout, it is possible for different callout
5847 options to be set for different recipients. It is too complicated to keep
5848 track of this in the cache, so now Exim always runs a verification when a
5849 callout is required, relying on the callout cache for the optimization.
5850 The overhead is duplication of the address routing, but this should not be
5853 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
5854 to be verified with callout=postmaster, and the main callout worked but the
5855 postmaster check failed, the verification correctly failed. However, if a
5856 subsequent RCPT command asked for sender verification *without* the
5857 postmaster check, incorrect caching caused this verification also to fail,
5860 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
5861 it was not caching the DNS options (qualify_single, search_parents) that
5862 were used when the lookup failed. A subsequent lookup with different
5863 options therefore always gave the same answer, though there were cases
5864 where it should not have. (Example: a "domains = !$mx_any" option on a
5865 dnslookup router: the "domains" option is always processed without any
5866 widening, but the router might have qualify_single set.) Now Exim uses the
5867 cached value only when the same options are set.
5869 32. Added John Jetmore's "exipick" utility to the distribution.
5871 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
5872 than a timeout (e.g. a certificate is required, and is not provided), an
5873 Exim server now closes the connection immediately. Previously it waited for
5874 the client to close - but if the client is SSL, it seems that they each
5875 wait for each other, leading to a delay before one of them times out.
5877 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
5878 maintained 0.8.x compatibility because I don't think many are using it, and
5879 it is clearly obsolete.
5881 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
5884 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
5885 client certificate was expired. A simple patch fixes this, though I don't
5886 understand the full logic of why the verify callback is called multiple
5889 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
5890 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
5891 which causes problems with some clients (such as the Certicom SSL Plus
5892 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
5893 disables the coutermeasure allowing Eudora to connect."
5895 38. Exim was not checking that a write() to a log file succeeded. This could
5896 lead to Bad Things if a log got too big, in particular if it hit a file
5897 size limit. Exim now panics and dies if it cannot write to a log file, just
5898 as it does if it cannot open a log file.
5900 39. Modified OS/Makefile-Linux so that it now contains
5902 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
5904 The two -D definitions ensure that Exim is compiled with large file
5905 support, which makes it possible to handle log files that are bigger than
5908 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
5909 instance) a domain was checked against a named list that involved a lookup,
5910 causing $domain_data to be set, then another domain was checked against the
5911 same list, then the first domain was re-checked, the value of $domain_data
5912 after the final check could be wrong. In particular, if the second check
5913 failed, it could be set empty. This bug probably also applied to
5916 41. The strip_trailing_dot option was not being applied to the address given
5917 with the -f command-line option.
5919 42. The code for reading a message's header from the spool was incrementing
5920 $received_count, but never initializing it. This meant that the value was
5921 incorrect (doubled) while delivering a message in the same process in which
5922 it was received. In the most common configuration of Exim, this never
5923 happens - a fresh exec is done - but it can happen when
5924 deliver_drop_privilege is set.
5926 43. When Exim logs an SMTP synchronization error - client data sent too soon -
5927 it now includes up to 150 characters of the unexpected data in the log
5930 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
5931 and building data strings. The size of both of these buffers was 10 000
5932 bytes - far larger than anybody would *ever* want, thought I. Needless to
5933 say, somebody hit the limit. I have increased the maximum line length to
5934 20 000 and the maximum data length of concatenated lines to 100 000. I have
5935 also fixed two bugs, because there was no checking on these buffers. Tsk,
5936 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
5937 buffer is too small.
5939 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
5940 lsearch lookups. Now it does.
5942 46. When parsing a route_list item in a manualroute router, a fixed-length
5943 buffer was used for the list of hosts. I made this 1024 bytes long,
5944 thinking that nobody would ever have a list of hosts that long. Wrong.
5945 Somebody had a whole pile of complicated expansion conditions, and the
5946 string was silently truncated, leading to an expansion error. It turns out
5947 that it is easier to change to an unlimited length (owing to other changes
5948 that have happened since this code was originally written) than to build
5949 structure for giving a limitation error. The length of the item that
5950 expands into the list of hosts is now unlimited.
5952 47. The lsearch lookup could not handle data where the length of text line was
5953 more than 4095 characters. Such lines were truncated, leading to shortened
5954 data being returned. It should now handle lines of any length.
5956 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
5957 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
5960 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
5961 replacing strings such as BIN_DIRECTORY in a source file: the replacement
5962 no longer happens in comment lines. A list of replacements is now placed
5963 at the head of all of the source files, except those whose only change is
5964 to replace PERL_COMMAND in the very first #! line.
5966 50. Replaced the slow insertion sort in queue.c, for sorting the list of
5967 messages on the queue, with a bottom-up merge sort, using code contributed
5968 by Michael Haardt. This should make operations like -bp somewhat faster on
5969 large queues. It won't affect queue runners, except when queue_run_in_order
5972 51. Installed eximstats 1.31 in the distribution.
5974 52. Added support for SRV lookups to the dnslookup router.
5976 53. If an ACL referred to $message_body or $message_body_end, the value was not
5977 reset for any messages that followed in the same SMTP session.
5979 54. The store-handling optimization for building very long strings was not
5980 differentiating between the different store pools. I don't think this
5981 actually made any difference in practice, but I've tidied it.
5983 55. While running the routers to verify a sender address, $sender_address
5984 was still set to the sender address. This is wrong, because when routing to
5985 send a bounce to the sender, it would be empty. Therefore, I have changed
5986 it so that, while verifying a sender address, $sender_address is set to <>.
5987 (There is no change to what happens when verifying a recipient address.)
5989 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
5990 target A or AAAA records (if not already returned) without resetting the
5991 qualify_single or search_parents options of the DNS resolver. These are
5992 inappropriate in this case because the targets of MX and SRV records must
5993 be FQDNs. A broken DNS record could cause trouble if it happened to have a
5994 target that, when qualified, matched something in the local domain. These
5995 two options are now turned off when doing these lookups.
5997 57. It seems that at least some releases of Reiserfs (which does not have the
5998 concept of a fixed number of inodes) returns zero and not -1 for the
5999 number of available inodes. This interacted badly with check_spool_inodes,
6000 which assumed that -1 was the "no such thing" setting. What I have done is
6001 to check that the total number of inodes is greater than zero before doing
6002 the test of how many are available.
6004 58. When a "warn" ACL statement has a log_message modifier, the message is
6005 remembered, and not repeated. This is to avoid a lot of repetition when a
6006 message has many recipients that cause the same warning to be written.
6007 However, Exim was preserving the list of already written lines for an
6008 entire SMTP session, which doesn't seem right. The memory is now reset if a
6009 new message is started.
6011 59. The "rewrite" debugging flag was not showing the result of rewriting in the
6012 debugging output unless log_rewrite was also set.
6014 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
6015 of (int)(handle) when we know that handle contains (void *)(-1).
6017 61. The Exim daemon panic-logs an error return when it closes the incoming
6018 connection. However "connection reset by peer" seems to be common, and
6019 isn't really an error worthy of noting specially, so that particular error
6022 62. When Exim is trying to find all the local interfaces, it used to panic and
6023 die if the ioctl to get the interface flags failed. However, it seems that
6024 on at least one OS (Solaris 9) it is possible to have an interface that is
6025 included in the list of interfaces, but for which you get a failure error
6026 for this call. This happens when the interface is not "plumbed" into a
6027 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
6028 failure of the "get flags" call assumes that the interface is down.
6030 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
6031 makes life easier for people who are doing arithmetic on fields extracted
6032 from dates, where you often get leading zeros that should not be
6033 interpreted as octal.
6035 64. Added qualify_domain to the redirect router, to override the global
6038 65. If a pathologically long header line contained very many addresses (the
6039 report of this problem mentioned 10 000) and each of them was rewritten,
6040 Exim could use up a very large amount of memory. (It kept on making new
6041 copies of the header line as it rewrote, and never released the old ones.)
6042 At the expense of a bit more processing, the header rewriting function has
6043 been changed so that it no longer eats memory in this way.
6045 66. The generation of the Received: header has been moved from the time that a
6046 message starts to be received, to the time that it finishes. The timestamp
6047 in the Received: header should now be very close to that of the <= log
6048 line. There are two side-effects of this change:
6050 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
6051 logged header lines no longer include the local Received: line, because
6052 it has not yet been created. The same applies to a copy of the message
6053 that is returned to a non-SMTP sender when a message is rejected.
6055 (b) When a filter file is tested using -bf, no additional Received: header
6056 is added to the test message. After some thought, I decided that this
6059 This change does not affect the value of $received_for. It is still set
6060 after address rewriting, but before local_scan() is called.
6062 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
6064 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
6065 gave an unhelpful panic error message, and a defer error. I have managed to
6066 change this behaviour so that it now rejects any supplied certificate,
6067 which seems right, as the list of acceptable certificates is empty.
6069 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
6070 gave an unhelpful defer error. I have not managed to make this reject any
6071 supplied certificates, but the error message it gives is "no certificate
6072 supplied", which is not helpful.
6074 70. exigrep's output now also includes lines that are not associated with any
6075 message, but which match the given pattern. Implemented by a patch from
6076 Martin Sluka, which also tidied up the Perl a bit.
6078 71. Recipient callout verification, like sender verification, was using <> in
6079 the MAIL FROM command. This isn't really the right thing, since the actual
6080 sender may affect whether the remote host accepts the recipient or not. I
6081 have changed it to use the actual sender in the callout; this means that
6082 the cache record is now keyed on a recipient/sender pair, not just the
6083 recipient address. There doesn't seem to be a real danger of callout loops,
6084 since a callout by the remote host to check the sender would use <>.
6085 [SEE ABOVE: changed after hitting problems.]
6087 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
6088 temporary errors. However, in the case of such a code being given after
6089 the end of a data transmission (i.e. after ".") Exim was failing to write
6090 a retry record for the message. (Yes, there was some broken host that was
6091 actually sending 8xx at this point.)
6093 73. An unknown lookup type in a host list could cause Exim to panic-die when
6094 the list was checked. (An example that provoked this was putting <; in the
6095 middle of a list instead of at the start.) If this happened during a DATA
6096 ACL check, a -D file could be left lying around. This kind of configuration
6097 error no longer causes Exim to die; instead it causes a defer error. The
6098 incident is still logged to the main and panic logs.
6100 74. Buglet left over from Exim 3 conversion. The message "too many messages
6101 in one connection" was written to the rejectlog but not the mainlog, except
6102 when address rewriting (yes!) was being logged.
6104 75. Added write_rejectlog option.
6106 76. When a system filter was run not as root (that is, when system_filter_user
6107 was set), the values of the $n variables were not being returned to the
6108 main process; thus, they were not subsequently available in the $sn
6111 77. Added +return_path_on_delivery log selector.
6113 78. A connection timeout was being treated differently from recipients deferred
6114 when testing hosts_max_try with a message that was older than the host's
6115 retry timeout. (The host should not be counted, thus allowing all hosts to
6116 be tried at least once before bouncing.) This may have been the cause of an
6117 occasionally reported bug whereby a message would remain on the queue
6118 longer than the retry timeout, but would be bounced if a delivery was
6119 forced. I say "may" because I never totally pinned down the problem;
6120 setting up timeout/retry tests is difficult. See also the next item.
6122 79. The ultimate address timeout was not being applied to errors that involved
6123 a combination of host plus message (for example, a timeout on a MAIL
6124 command). When an address resolved to a number of possible hosts, and they
6125 were not all tried for each delivery (e.g. because of hosts_max_try), a
6126 message could remain on the queue longer than the retry timeout.
6128 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
6131 81. Fixed an obscure SMTP outgoing bug which required at least the following
6132 conditions: (a) there was another message waiting for the same server;
6133 (b) the server returned 5xx to all RCPT commands in the first message so
6134 that the message was not completed; (c) the server dropped the connection
6135 or gave a negative response to the RSET that Exim sends to abort the
6136 transaction. The observed case was a dropped connection after DATA that had
6137 been sent in pipelining mode. That is, the server had advertised PIPELINING
6138 but was not implementing it correctly. The effect of the bug was incorrect
6139 behaviour, such as trying another host, and this could lead to a crash.
6145 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
6146 and daemon.c were passed as pointers to ints; they should have been
6147 pointers to socklen_t variables (which are typically unsigned ints).
6149 2. Some signed/unsigned type warnings in the os.c file for Linux have been
6152 3. Fixed a really odd bug that affected only the testing scheme; patching a
6153 certain fixed string in the binary changed the value of another string that
6154 happened to be identical to the end of the original first string.
6156 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
6157 name", it returns that address as the IP address. On some operating
6158 systems (e.g. Solaris), it also passes back the IP address string as the
6159 "host name". However, on others (e.g. Linux), it passes back an empty
6160 string. Exim wasn't checking for this, and was changing the host name to an
6161 empty string, assuming it had been canonicalized.
6163 5. Although rare, it is permitted to have more than one PTR record for a given
6164 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
6165 all the names associated with an address, because they do in Solaris.
6166 However, it seems that they do not in Linux for data that comes from the
6167 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
6168 I found this out when I moved to a new Linux workstation and tried to run
6169 the Exim test suite.
6171 To get round this problem I have changed the code so that it now does its
6172 own call to the DNS to look up PTR records when searching for a host name.
6173 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
6174 addresses that are only in /etc/hosts are still found.
6176 This behaviour is, however, controlled by an option called host_lookup_
6177 order, which defaults to "bydns:byaddr". If people want to use the other
6178 order, or indeed, just use one or the other means of lookup, they can
6179 specify it in this variable.
6181 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
6182 some operating systems, this comes back from gethostbyaddr() as an empty
6183 string, and this is what Exim used to test for. However, it seems that in
6184 other systems, "." is yielded. Exim now tests for this case too.
6186 7. The values of check_spool_space and check_log_space are now held internally
6187 as a number of kilobytes instead of an absolute number of bytes. If a
6188 numbers is specified without 'K' or 'M', it is rounded up to the nearest
6189 kilobyte. This means that much larger values can be stored.
6191 8. Exim monitor: an attempt to get the action menu when not actually pointing
6192 at a message produces an empty menu entitled "No message selected". This
6193 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
6194 no entries in it ("Shell widget menu has zero width and/or height"). So I
6195 have added a single, blank menu entry in this case.
6197 9. Added ${quote_local_part.
6199 10. MIME decoding is now applied to the contents of Subject: header lines when
6202 11. Now that a reference to $sender_host_address automatically causes a reverse
6203 lookup to occur if necessary (4.13/18), there is no need to arrange for a
6204 host lookup before query-style lookups in lists that might use this
6205 variable. This has therefore been abolished, and the "net-" prefix is no
6206 longer necessary for query-style lookups.
6208 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
6209 have been a typo for LFLAGS, so it has been changed.
6211 13. The install script calls Exim with "-C /dev/null" in order to find the
6212 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
6213 to be output. However, since Exim outputs its version number before the
6214 error, it didn't break the script. It just looked ugly. I fixed this by
6215 always allowing "-C /dev/null" if the caller is root.
6217 14. Ignore overlarge ACL variable number when reading spool file - insurance
6218 against a later release with more variables having written the file.
6220 15. The standard form for an IPv6 address literal was being rejected by EHLO.
6221 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
6222 this, as well as the form without the "IPv6" on the front.
6224 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
6225 OS/Makefile-Darwin file.
6227 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
6228 only to LDAP libraries that do not have LDAP_OPT_DEREF.
6230 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
6231 variable. However, some OS use ints for inodes. I've added cast to long int
6232 to get rid of the compiler warning.
6234 19. I had forgotten to lock out "/../" in configuration file names when
6235 ALT_CONFIG_PREFIX was set.
6237 20. Routers used for verification do not need to specify transports. However,
6238 if such a router generated a host list, and callout was configured, Exim
6239 crashed, because it could not find a port number from the (non-existent)
6240 transport. It now assumes port 25 in this circumstance.
6242 21. Added the -t option to exigrep.
6244 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
6245 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
6246 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
6247 them is set, LOOKUP_LSEARCH is forced.
6249 23. "exim -bV" now outputs a list of lookups that are included in the binary.
6251 24. Added sender and host information to the "rejected by local_scan()" log
6252 line; previously there was no indication of these.
6254 25. Added .include_if_exists.
6256 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
6257 Linux. It turns out that not all file systems support this. Apparently some
6258 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
6259 do it.) To cope with this, the error EINVAL, which means that sync-ing is
6260 not supported on the file descriptor, is now ignored when Exim is trying to
6261 sync a directory. This applies only to Linux.
6263 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
6265 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
6266 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
6269 29. Added support for maildirsize files from supplied patch (modified a bit).
6271 30. The use of :fail: followed by an empty string could lead Exim to respond to
6272 sender verification failures with (e.g.):
6274 550 Verification failed for <xxx>
6275 550 Sender verify failed
6277 where the first response line was missing the '-' that indicates it is not
6278 the final line of the response.
6280 31. The loop for finding the name of the user that called Exim had a hardwired
6281 limit of 10; it now uses the value of finduser_retries, which is used for
6282 all other user lookups.
6284 32. Added $received_count variable, available in data and not_smtp ACLs, and at
6287 33. Exim was neglecting to zero errno before one call of strtol() when
6288 expanding a string and expecting an integer value. On some systems this
6289 resulted in spurious "integer overflow" errors. Also, it was casting the
6290 result into an int without checking.
6292 34. Testing for a connection timeout using "timeout_connect" in the retry rules
6293 did not work. The code looks as if it has *never* worked, though it appears
6294 to have been documented since at least release 1.62. I have made it work.
6296 35. The "timeout_DNS" error in retry rules, also documented since at least
6297 1.62, also never worked. As it isn't clear exactly what this means, and
6298 clearly it isn't a major issue, I have abolished the feature by treating it
6299 as "timeout", and writing a warning to the main and panic logs.
6301 36. The display of retry rules for -brt wasn't always showing the error code
6304 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
6305 timeout_connect_A, timeout_connect_MX.
6307 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
6308 to the empty sender.
6310 39. The daemon was not analysing the content of -oX till after it had closed
6311 stderr and disconnected from the controlling terminal. This meant that any
6312 syntax errors were only noted on the panic log, and the return code from
6313 the command was 0. By re-arranging the code a little, I've made the
6314 decoding happen first, so such errors now appear on stderr, and the return
6315 code is 1. However, the actual setting up of the sockets still happens in
6316 the disconnected process, so errors there are still only recorded on the
6319 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
6320 connections (as happens on some IP stacks) was logged at start up time as
6321 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
6322 it from "IPv6 and IPv4", which means that two separate sockets are being
6325 41. The debug output for gethostbyname2() or getipnodebyname() failures now
6326 says whether AF_INET or AF_INET6 was passed as an argument.
6328 42. Exiwhat output was messed up when time zones were included in log
6331 43. Exiwhat now gives more information about the daemon's listening ports,
6332 and whether -tls-on-connect was used.
6334 44. The "port" option of the smtp transport is now expanded.
6336 45. A "message" modifier in a "warn" statement in a non-message ACL was being
6337 silently ignored. Now an error message is written to the main and panic
6340 46. There's a new ACL modifier called "logwrite" which writes to a log file
6341 as soon as it is encountered.
6343 47. Added $local_user_uid and $local_user_gid at routing time.
6345 48. Exim crashed when trying to verify a sender address that was being
6348 49. Exim was recognizing only a space character after ".include". It now also
6349 recognizes a tab character.
6351 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
6352 extracting the relevant information from the specification. The man page no
6353 longer contains scrambled data for the -d option, and I've added a section
6354 at the front about calling Exim under different names.
6356 51. Added "extra_headers" argument to the "mail" command in filter files.
6358 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
6361 53. Installed eximstats 1.29.
6363 54. Added transport_filter_timeout as a generic transport option.
6365 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
6366 Cc: header lines. This was required by RFC 822, but it not required by RFC
6369 56. Exim used to add From:, Date:, and Message-Id: header lines to any
6370 incoming messages that did not have them. Now it does so only if the
6371 message originates locally, that is, if there is no associated remote host
6372 address. When Resent- header lines are present, this applies to the Resent-
6373 lines rather than the non-Resent- lines.
6375 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
6376 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
6378 58. Messages for configuration errors now include the name of the main
6379 configuration file - useful now that there may be more than one file in a
6380 list (.included file names were always shown).
6382 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
6383 for those rare installations that do not start the daemon as root or run it
6384 setuid root. I've cut out the call to initgroups() if the daemon is not
6387 60. The Exim user and group can now be bound into the binary as text strings
6388 that are looked up at the start of Exim's processing.
6390 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
6392 62. Added $mailstore_basename variable.
6394 63. Installed patch to sieve.c from Michael Haardt.
6396 64. When Exim failed to open the panic log after failing to open the main log,
6397 the original message it was trying to log was written to stderr and debug
6398 output, but if they were not available (the usual case in production), it
6399 was lost. Now it is written to syslog before the two lines that record the
6400 failures to open the logs.
6402 65. Users' Exim filters run in subprocesses under the user's uid. It is
6403 possible for a "deliver" command or an alias in a "personal" command to
6404 provoke an address rewrite. If logging of address rewriting is configured,
6405 this fails because the process is not running as root or exim. There may be
6406 a better way of dealing with this, but for the moment (because 4.30 needs
6407 to be released), I have disabled address rewrite logging when running a
6408 filter in a non-root, non-exim process.
6414 1. The buildconfig auxiliary program wasn't quoting the value set for
6415 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
6416 not defined. This bug was masked in 4.22 by the effect that was fixed in
6419 2. Some messages that were rejected after a message id was allocated were
6420 shown as "incomplete" by exigrep. It no longer does this for messages that
6421 are rejected by local_scan() or the DATA or non-SMTP ACLs.
6423 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
6424 have allow_domain_literals set, the ID did not get logged in the <= line.
6425 Domain literals are now always recognized in Message-ID: header lines.
6427 4. The first argument for a ${extract expansion item is the key name or field
6428 number. Leading and trailing spaces in this item were not being ignored,
6429 causing some misleading effects.
6431 5. When deliver_drop_privilege was set, single queue runner processes started
6432 manually (i.e. by the command "exim -q") or by the daemon (which uses the
6433 same command in the process it spins off) were not dropping privilege.
6435 6. When the daemon running as "exim" started a queue runner, it always
6436 re-executed Exim in the spun-off process. This is a waste of effort when
6437 deliver_drop_privilege is set. The new process now just calls the
6438 queue-runner function directly.
6444 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
6447 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
6448 ignored. Though the use of -oP was forcing the writing of a pid file, it
6449 was always written to the default place.
6451 3. If the message "no IP address found for host xxxx" is generated during
6452 incoming verification, it is now followed by identification of the incoming
6453 connection (so you can more easily find what provoked it).
6455 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
6457 5. Added some features to "harden" Exim a bit more against certain attacks:
6459 (a) There is now a build-time option called FIXED_NEVER_USERS that can
6460 be put in Local/Makefile. This is like the never_users runtime option,
6461 but it cannot be overridden. The default setting is "root".
6463 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
6464 prefix string with which any file named in a -C command line option
6467 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
6468 is retained for -C and -D only if the caller of Exim is root. Without
6469 it, the exim user may also use -C and -D and retain privilege.
6471 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
6472 command line option is disabled.
6474 6. Macro names set by the -D option must start with an upper case letter, just
6475 like macro names defined in the configuration file.
6477 7. Added "dereference=" facility to LDAP.
6479 8. Two instances of the typo "uknown" in the source files are fixed.
6481 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
6482 the Configure-Makefile script screwed up while processing it.
6484 10. Incorporated PCRE 4.4.
6486 11. The SMTP synchronization check was not operating right at the start of an
6487 SMTP session. For example, it could not catch a HELO sent before the client
6488 waited for the greeting. There is now a check for outstanding input at the
6489 point when the greeting is written. Because of the duplex, asynchronous
6490 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
6491 way, but not yet received, when the check is performed.
6493 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
6494 on TCP/IP sockets, because this apparently causes some broken clients to
6497 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
6498 unchanged) from the Cygwin maintainer.
6500 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
6501 was supported instead of testing for mbx. Effectively a typo.
6503 15. The spa authenticator server code was not checking that the input it
6504 received was valid base64.
6506 16. The debug output line for the "set" modifier in ACLs was not showing the
6507 name of the variable that was being set.
6509 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
6511 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
6512 lookup on its own. Something else was needed to trigger the lookup. For
6513 example, a match in host_lookup or the need for a host name in a host list.
6514 Now, if $sender_host_name is referenced and the host name has not yet been
6515 looked up, a lookup is performed. If the lookup fails, the variable remains
6516 empty, and $host_lookup_failed is set to "1".
6518 19. Added "eqi" as a case-independent comparison operator.
6520 20. The saslauthd authentication condition could segfault if neither service
6521 nor realm was specified.
6523 21. If an overflowing value such as "2048M" was set for message_size_limit, the
6524 error message that was logged was misleading, and incoming SMTP
6525 connections were dropped. The message is now more accurate, and temporary
6526 errors are given to SMTP connections.
6528 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
6529 (except RSET) with a 421 error, until QUIT is received. However, it was
6530 failing to send a response to QUIT.
6532 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
6533 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
6534 ACL is now run after the helo_try_verify_hosts code.
6536 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
6537 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
6538 case-independent, so other case variants are also recognized). Apparently
6539 some systems use these upper case variants.
6541 25. If more than two messages were waiting for the same host, and a transport
6542 filter was specified for the transport, Exim sent two messages over the
6543 same TCP/IP connection, and then failed with "socket operation on non-
6544 socket" when it tried to send the third.
6546 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
6548 27. The extern definition of crypt16() in expand.c was not being excluded when
6549 the OS had its own crypt16() function.
6551 28. Added bounce_return_body as a new option, and bounce_return_size_limit
6552 as a preferred synonym for return_size_limit, both as an option and as an
6555 29. Added LIBS=-liconv to OS/Makefile-OSF1.
6557 30. Changed the default configuration ACL to relax the local part checking rule
6558 for addresses that are not in any local domains. For these addresses,
6559 slashes and pipe symbols are allowed within local parts, but the sequence
6560 /../ is explicitly forbidden.
6562 31. SPA server authentication was not clearing the challenge buffer before
6565 32. log_message in a "warn" ACL statement was writing to the reject log as
6566 well as to the main log, which contradicts the documentation and doesn't
6567 seem right (because no rejection is happening). So I have stopped it.
6569 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
6570 However, I am unable to do any testing of this.
6572 34. Fixed an infelicity in the appendfile transport. When checking directories
6573 for a mailbox, to see if any needed to be created, it was accidentally
6574 using path names with one or more superfluous leading slashes; tracing
6575 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
6577 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
6578 contents are added to the log line that is written for every discarded
6579 recipient. (Previously a log_message setting was ignored.)
6581 36. The ${quote: operator now quotes the string if it is empty.
6583 37. The install script runs exim in order to find its version number. If for
6584 some reason other than non-existence or emptiness, which it checks, it
6585 could not run './exim', it was installing it with an empty version number,
6586 i.e. as "exim-". This error state is now caught, and the installation is
6589 38. An argument was missing from the function that creates an error message
6590 when Exim fails to connect to the socket for saslauthd authentication.
6591 This could cause Exim to crash, or give a corrupted message.
6593 39. Added isip, isip4, and isip6 to ${if conditions.
6595 40. The ACL variables $acl_xx are now saved with the message, and can be
6596 accessed later in routers, transports, and filters.
6598 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
6599 strings in the file are not string-expanded.
6601 42. If a MAIL command specified a SIZE value that was too large to fit into an
6602 int variable, the check against message_size_limit failed. Such values are
6603 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
6604 day this will have to be increased, but I don't think I want to be around
6605 when emails are that large.
6612 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
6613 iconv() is not standard in FreeBSD.
6615 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
6616 IPv6 enabled. The observed symptom was a segmentation fault on return from
6617 the function os_common_find_running_interfaces() in src/os.c.
6619 3. In the check_special_case() function in daemon.c I had used "errno" as an
6620 argument name, which causes warnings on some systems. This was basically a
6621 typo, since it was named "eno" in the comments!
6623 4. The code that waits for the clock to tick (at a resolution of some fraction
6624 of a second) so as to ensure message-id uniqueness was always waiting for
6625 at least one whole tick, when it could have waited for less. [This is
6626 almost certainly not relevant at current processor speeds, where it is
6627 unlikely to ever wait at all. But we try to future-proof.]
6629 5. The function that sleeps for a time interval that includes fractions of a
6630 second contained a race. It did not block SIGALRM between setting the
6631 timer, and suspending (a couple of lines later). If the interval was short
6632 and the sigsuspend() was delayed until after it had expired, the suspension
6633 never ended. On busy systems this could lead to processes getting stuck for
6636 6. Some uncommon configurations may cause a lookup to happen in a queue runner
6637 process, before it forks any delivery processes. The open lookup caching
6638 mechanism meant that the open file or database connection was passed into
6639 the delivery process. The problem was that delivery processes always tidy
6640 up cached lookup data. This could cause a problem for the next delivery
6641 process started by the queue runner, because the external queue runner
6642 process does not know about the closure. So the next delivery process
6643 still has data in the lookup cache. In the case of a file lookup, there was
6644 no problem because closing a file descriptor in a subprocess doesn't affect
6645 the parent. However, if the lookup was caching a connection to a database,
6646 the connection was closed, and the second delivery process was likely to
6647 see errors such as "PGSQL: query failed: server closed the connection
6648 unexpectedly". The problem has been fixed by closing all cached lookups
6649 in a queue runner before running a delivery process.
6651 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
6652 seem to have the "const" qualifier which it has on other OS. I've
6655 8. Change 4.21/2 was too strict. It is only if there are two authenticators
6656 *of the same type* (client or server) with the same public name that an
6657 error should be diagnosed.
6659 9. When Exim looked up a host name for an IP address, but failed to find the
6660 original IP address when looking up the host name (a safety check), it
6661 output the message "<ip address> does not match any IP for NULL", which was
6662 confusing, to say the least. The bug was that the host name should have
6663 appeared instead of "NULL".
6665 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
6666 user that is built into the binary, and the -C or -D options is used, root
6667 privilege is dropped before the configuration file is read. In addition,
6668 logging is switched to stderr instead of the normal log files. If the
6669 configuration then re-defines the Exim user, the unprivileged environment
6670 is probably not what is expected, so Exim logs a panic warning message (but
6673 However, if deliver_drop_privilege is set, the unprivileged state may well
6674 be exactly what is intended, so the warning has been cut out in that case,
6675 and Exim is allowed to try to write to its normal log files.
6681 1. smtp_return_error_details was not giving details for temporary sender
6682 or receiver verification errors.
6684 2. Diagnose a configuration error if two authenticators have the same public
6687 3. Exim used not to create the message log file for a message until the first
6688 delivery attempt. This could be confusing when incoming messages were held
6689 for policy or load reasons. The message log file is now created at the time
6690 the message is received, and an initial "Received" line is written to it.
6692 4. The automatically generated man page for command line options had a minor
6693 bug that caused no ill effects; however, a more serious problem was that
6694 the procedure for building the man page automatically didn't always
6695 operate. Consequently, release 4.20 contains an out-of-date version. This
6696 shouldn't happen again.
6698 5. When building Exim with embedded Perl support, the script that builds the
6699 Makefile was calling 'perl' to find its compile-time parameters, ignoring
6700 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
6702 6. The freeze_tell option was not being used for messages that were frozen on
6703 arrival, either by an ACL or by local_scan().
6705 7. Added the smtp_incomplete_transaction log selector.
6707 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
6708 was accepting AUTH without a new EHLO.
6710 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
6711 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
6714 10. Logging of TCP/IP connections (when configured) now happens in the main
6715 daemon process instead of the child process, so that the TCP/IP connection
6716 count is more accurate (but it can never be perfect).
6718 11. The use of "drop" in a nested ACL was not being handled correctly in the
6719 outer ACL. Now, if condition failure induced by the nested "drop" causes
6720 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
6721 or "require"), the connection is dropped.
6723 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
6724 that yield "discard" can now be used with an "accept" or a "discard" verb,
6725 but an error is generated for any others (because I can't see a useful way
6726 to define what should happen).
6728 13. When an ACL is read dynamically from a file (or anywhere else), the lines
6729 are now processed in the same way as lines in the Exim configuration file.
6730 In particular, continuation lines are supported.
6732 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
6734 15. Added -ti meaning -t -i.
6736 16. Check for letters, digits, hyphens, and dots in the names of dnslist
6737 domains, and warn by logging if others are found.
6739 17. At least on BSD, alignment is not guaranteed for the array of ifreq's
6740 returned from GIFCONF when Exim is trying to find the list of interfaces on
6741 a host. The code in os.c has been modified to copy each ifreq to an aligned
6742 structure in all cases.
6744 Also, in some cases, the returned ifreq's were being copied to a 'struct
6745 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
6746 means the last couple of bytes of an IPv6 address could be chopped if the
6747 ifreq contained only a normal sockaddr (14 bytes storage).
6749 18. Named domain lists were not supported in the hosts_treat_as_local option.
6750 An entry such as +xxxx was not recognized, and was treated as a literal
6753 19. Ensure that header lines added by a DATA ACL are included in the reject log
6754 if the ACL subsequently rejects the message.
6756 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
6757 MD5 (which is deprecated).
6759 21. When testing a filter file using -bf, Exim was writing a message when it
6760 took the sender from a "From " line in the message, but it was not doing so
6761 when it took $return_path from a Return-Path: header line. It now does.
6763 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
6764 with a valid header line field name (a series of printing characters
6765 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
6767 23. Changed "disc" in the source to "disk" to conform to the documentation and
6768 the book and for uniformity.
6770 24. Ignore Sendmail's -Ooption=value command line item.
6772 25. When execve() failed while trying to run a command in a pipe transport,
6773 Exim was returning EX_UNAVAILABLE (69) from the subprocess. However, this
6774 could be confused with a return value of 69 from the command itself. This
6775 has been changed to 127, the value the shell returns if it is asked to run
6776 a non-existent command. The wording for the related log line suggests a
6777 non-existent command as the problem.
6779 26. If received_header_text expands to an empty string, do not add a Received:
6780 header line to the message. (Well, it adds a token one on the spool, but
6781 marks it "old" so that it doesn't get used or transmitted.)
6783 27. Installed eximstats 1.28 (addition of -nt option).
6785 28. There was no check for failure on the call to getsockname() in the daemon
6786 code. This can fail if there is a shortage of resources on the system, with
6787 ENOMEM, for example. A temporary error is now given on failure.
6789 29. Contrary to the C standard, it seems that in some environments, the
6790 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
6791 program. Exim now does this explicitly; it affects the formatting of
6792 timestamps using strftime().
6794 30. If exiqsumm was given junk data, it threw up some uninitialized variable
6795 complaints. I've now initialized all the variables, to avoid this.
6797 32. Header lines added by a system filter were not being "seen" during
6798 transport-time rewrites.
6800 33. The info_callback() function passed to OpenSSL is set up with type void
6801 (*)(SSL *, int, int), as described somewhere. However, when calling the
6802 function (actually a macro) that sets it up, the type void(*)() is
6803 expected. I've put in a cast to prevent warnings from picky compilers.
6805 34. If a DNS black list lookup found a CNAME record, but there were no A
6806 records associated with the domain it pointed at, Exim crashed.
6808 35. If a DNS black list lookup returned more than one A record, Exim ignored
6809 all but the first. It now scans all returned addresses if a particular IP
6810 value is being sought. In this situation, the contents of the
6811 $dnslist_value variable are a list of all the addresses, separated by a
6814 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
6815 to accept a host name and all its aliases if the forward lookup for any of
6816 them yielded the IP address of the incoming connection. Now it accepts only
6817 those names whose forward lookup yields the correct IP address. Any other
6818 names are discarded. This closes a loophole whereby a rogue DNS
6819 administrator could create reverse DNS records to break through a
6820 wildcarded host restriction in an ACL.
6822 37. If a user filter or a system filter that ran in a subprocess used any of
6823 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
6824 the wrong values were passed to the pipe command ($thisaddress had the
6825 value of $0, $0 had the value of $1, etc). This bug was introduced by
6826 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
6828 38. Improved the line breaking for long SMTP error messages from ACLs.
6829 Previously, if there was no break point between 40 and 75 characters, Exim
6830 left the rest of the message alone. Two changes have been made: (a) I've
6831 reduced the minimum length to 35 characters; (b) if it can't find a break
6832 point between 35 and 75 characters, it looks ahead and uses the first one
6833 that it finds. This may give the occasional overlong line, but at least the
6834 remaining text gets split now.
6836 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
6837 file descriptors might be low, and that setting 1000 would always raise it.
6838 It turns out that in some environments, the limit is already over 1000 and
6839 that lowering it causes trouble. So now Exim takes care not to decrease it.
6841 40. When delivering a message, the value of $return_path is set to $sender_
6842 address at the start of routing (routers may change the value). By an
6843 oversight, this default was not being set up when an address was tested by
6844 -bt or -bv, which affected the outcome if any router or filter referred to
6847 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
6848 log only when "message" or "log_message" are set. However, if one of the
6849 conditions was an address verification, or a call to a nested ACL, the
6850 messages generated by the underlying test were being passed through. This
6851 no longer happens. The underlying message is available in $acl_verify_
6852 message for both "message" and "log_message" expansions, so it can be
6853 passed through if needed.
6855 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
6856 new expansion $bh_ to give the encoded byte string without charset
6857 translation. Translation happens only if iconv() is available; HAVE_ICONV
6858 indicates this at build time. HEADERS_CHARSET gives the charset to
6859 translate to; headers_charset can change it in the configuration, and
6860 "headers charset" can change it in an individual filter file.
6862 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
6863 that creates RFC 2047 encoded "words" labels them as that charset instead
6864 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
6865 expansion operator; (ii) when Exim creates a From: line for a local
6866 message; (iii) when a header line is rewritten to include a "phrase" part.
6868 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
6869 buggy, causing it to skip the first lines of messages whose message ID
6870 ended in 'D'. This would not have bitten before Exim release 4.14, because
6871 message IDs were unlikely to end in 'D' before then. The effect was to have
6872 incorrect size information for certain domains.
6874 45. #include "config.h" was missing at the start of the crypt16.c module. This
6875 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
6878 46. If there was a timeout during a "random" callout check, Exim treated it as
6879 a failure of the random address, and carried on sending RSET and the real
6880 address. If the delay was just some slowness somewhere, the response to the
6881 original RCPT would be taken as a response to RSET and so on, causing
6882 mayhem of various kinds.
6884 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
6885 when I implemented it. It didn't allow for the fact that some option values
6886 may legitimately be negative (e.g. size_addition), and it didn't even do
6887 the right test for positive values.
6889 48. Domain names in DNS records are case-independent. Exim always looks them up
6890 in lower case. Some resolvers return domain names in exactly the case they
6891 appear in the zone file, that is, they may contain uppercase letters. Not
6892 all resolvers do this - some return always lower case. Exim was treating a
6893 change of case by a resolver as a change of domain, similar to a widening
6894 of a domain abbreviation. This triggered its re-routing code and so it was
6895 trying to route what was effectively the same domain again. This normally
6896 caused routing to fail (because the router wouldn't handle the domain
6897 twice). Now Exim checks for this case specially, and just changes the
6898 casing of the domain that it ultimately uses when it transmits the message
6901 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
6904 50. If a filter generated a file delivery with a non-absolute name (possible if
6905 no home directory exists for the router), the forbid_file option was not
6908 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
6909 the existing equality matching.
6911 52. Exim was using ints instead of ino_t variables in some places where it was
6912 dealing with inode numbers.
6914 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
6915 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
6916 called TMPDIR, and if it finds it is different, it changes its value.
6918 54. The smtp_printf() function is now made available to local_scan() so
6919 additional output lines can be written before returning. There is also an
6920 smtp_fflush() function to enable the detection of a dropped connection.
6921 The variables smtp_input and smtp_batched_input are exported to
6924 55. Changed the default runtime configuration: the message "Unknown user"
6925 has been removed from the ACL, and instead placed on the localuser router,
6926 using the cannot_route_message feature. This means that any verification
6927 failures that generate their own messages won't get overridden. Similarly,
6928 the "Unrouteable address" message that was in the ACL for unverifiable
6929 relay addresses has also been removed.
6931 56. Added hosts_avoid_esmtp to the smtp transport.
6933 57. The exicyclog script was not checking for the esoteric option
6934 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
6935 will work only if exicyclog is run under the appropriate euid.
6937 58. Following a discussion on the list, the rules by which Exim recognises line
6938 endings on incoming messages have been changed. The -dropcr and drop_cr
6939 options are now no-ops, retained only for backwards compatibility. The
6940 following line terminators are recognized: LF CRLF CR. However, special
6941 processing applies to CR:
6943 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
6944 nor a local message in the state where . is a terminator.
6946 (ii) If a bare CR is encountered in a header line, an extra space is added
6947 after the line terminator so as not to end the header. The reasoning
6948 behind this is that bare CRs in header lines are most likely either
6949 to be mistakes, or people trying to play silly games.
6951 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
6952 was being incorrectly given as 18 bytes larger than it should have been.
6953 This is a VOB (very old bug).
6955 60. This may never have affected anything current, but just in case it has:
6956 When the local host is found other than at the start of a list of hosts,
6957 the local host, those with the same MX, and any that follow, are discarded.
6958 When the list in question was part of a longer list of hosts, the following
6959 hosts (not currently being processed) were also being discarded. This no
6960 longer happens. I'm not sure if this situation could ever has previously
6963 61. Added the "/MX" feature to lists of hosts in the manualroute and query
6966 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
6967 header. This is something that is recommended in a new Internet Draft, and
6968 is something that is documented as being done by Sendmail. There are two
6969 possible values. For messages generated by the autoreply transport, Exim
6972 Auto-Submitted: auto-replied
6974 whereas for all other generated messages (e.g. bounces) it adds
6976 Auto-Submitted: auto-generated
6978 63. The "personal" condition in filters now includes a test for the
6979 Auto-Submitted: header. If it contains the string "auto-" the message it
6980 not considered personal.
6982 64. Added rcpt_include_affixes as a generic transport option.
6984 65. Added queue_only_override (default true).
6986 66. Added the syslog_duplication option.
6988 67. If what should have been the first header line of a message consisted of
6989 a space followed by a colon, Exim was mis-interpreting it as a header line.
6990 It isn't of course - it is syntactically invalid and should therefore be
6991 treated as the start of the message body. The misbehaviour could have
6992 caused a number of strange effects, including loss of data in subsequent
6993 header lines, and spool format errors.
6995 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
6996 client host had authenticated. This control can now be exercised by an ACL
6997 for more flexibility.
6999 69. By default, callouts do not happen when testing with -bh. There is now a
7000 variant, -bhc, which does actually run the callout code, including
7001 consulting and updating the callout cache.
7003 70. Added support for saslauthd authentication, courtesy of Alexander
7006 71. If statvfs() failed on the spool or log directories while checking their
7007 size for availability, Exim confusingly gave the error "space shortage".
7008 Furthermore, in debugging mode it crashed with a floating point exception.
7009 These checks are done if check_{spool,log}_{space,inodes} are set, and when
7010 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
7011 serious problem, Exim now writes to the main and panic logs when this
7012 happens, with details of the failure. It then refuses to accept the
7013 incoming message, giving the message "spool directory problem" or "log
7014 directory problem" with a 421 code for SMTP messages.
7016 72. When Exim is about to re-exec itself, it ensures that the file descriptors
7017 0, 1, and 2 exist, because some OS complain for execs without them (see
7018 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
7019 descriptors. However, the code omitted to check that the open succeeded,
7020 causing mysterious errors if for some reason the permissions on /dev/null
7021 got screwed. Now Exim writes a message to the main and panic logs, and
7022 bombs out if it can't open /dev/null.
7024 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
7025 interact so that it is all more flexible. It is supposed to remain
7026 backwards compatible. Also added extra_local_interfaces.
7028 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
7029 to bomb out with an assertion failure - to the client this appears as a
7030 connection drop. This problem occurs in the part of the code that was taken
7031 from the Samba project. Fortunately, the assertion is in a very simple
7032 function, so I have fixed this by reproducing the function inline in the
7033 one place where it is called, and arranging for authentication to fail
7034 instead of killing the process with assert().
7036 75. The SPA client code was not working when the server requested OEM rather
7037 than Unicode encoding.
7039 76. Added code to make require_files with a specific uid setting more usable in
7040 the case where statting the file as root fails - usually a non-root-mounted
7041 NFS file system. When this happens and the failure is EACCES, Exim now
7042 forks a subprocess and does the per-uid checking as the relevant uid.
7044 77. Added process_log_path.
7046 78. If log_file_path was not explicitly set, a setting of check_log_space or
7047 check_log_inodes was ignored.
7049 79. If a space check for the spool or log partitions fails, the incident is now
7050 logged. Of course, in the latter case the data may get lost...
7052 80. Added the %p formatting code to string_format() so that it can be used to
7053 print addresses in debug_print(). Adjusted all the address printing in the
7054 debugging in store.c to use %p rather than %d.
7056 81. There was a concern that a line of code in smtp_in.c could overflow a
7057 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
7058 initially expressed, the concern was not well-founded, because trailing
7059 spaces are removed early. However, if the trailing spaces were followed by
7060 a NULL, they did not get removed, so the overflow was possible. Two fixes
7063 (a) I re-wrote the offending code in a cleaner fashion.
7064 (b) If an incoming SMTP command contains a NULL character, it is rejected
7067 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
7068 runs initgroups(), so that if the Exim user is in any additional groups,
7069 they will be used during message reception.
7075 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / exim.git / blob