1 Change log file for Exim from version 4.21
2 ------------------------------------------
3 This document describes *changes* to previous versions, that might
4 affect Exim's operation, with an unchanged configuration file. For new
5 options, and new features, see the NewStuff file next to this ChangeLog.
11 JH/01 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 JH/16 Fix bug in DKIM verify: a buffer overflow could corrupt the malloc
86 metadata, resulting in a crash in free().
88 PP/01 Fix broken refactor of Heimdal GSSAPI authenticator init.
89 Broken in f2ed27cf5, missing an equals sign for specified-initialisers.
95 JH/01 Rework error string handling in TLS interface so that the caller in
96 more cases is responsible for logging. This permits library-sourced
97 string to be attached to addresses during delivery, and collapses
98 pairs of long lines into single ones.
100 PP/01 Allow PKG_CONFIG_PATH to be set in Local/Makefile and use it correctly
101 during configuration. Wildcards are allowed and expanded.
103 JH/02 Rework error string handling in DKIM to pass more info back to callers.
104 This permits better logging.
106 JH/03 Rework the transport continued-connection mechanism: when TLS is active,
107 do not close it down and have the child transport start it up again on
108 the passed-on TCP connection. Instead, proxy the child (and any
109 subsequent ones) for TLS via a unix-domain socket channel. Logging is
110 affected: the continued delivery log lines do not have any DNSSEC, TLS
111 Certificate or OCSP information. TLS cipher information is still logged.
113 JH/04 Shorten the log line for daemon startup by collapsing adjacent sets of
114 identical IP addresses on different listening ports. Will also affect
117 PP/02 Bug 2070: uClibc defines __GLIBC__ without providing glibc headers;
118 add noisy ifdef guards to special-case this sillyness.
119 Patch from Bernd Kuhls.
121 JH/05 Tighten up the checking in isip4 (et al): dotted-quad components larger
122 than 255 are no longer allowed.
124 JH/06 Default openssl_options to include +no_ticket, to reduce load on peers.
125 Disable the session-cache too, which might reduce our load. Since we
126 currrectly use a new context for every connection, both as server and
127 client, there is no benefit for these.
128 GnuTLS appears to not support tickets server-side by default (we don't
129 call gnutls_session_ticket_enable_server()) but client side is enabled
130 by default on recent versions (3.1.3 +) unless the PFS priority string
133 PP/03 Add $SOURCE_DATE_EPOCH support for reproducible builds, per spec at
134 <https://reproducible-builds.org/specs/source-date-epoch/>.
136 JH/07 Fix smtp transport use of limited max_rcpt under mua_wrapper. Previously
137 the check for any unsuccessful recipients did not notice the limit, and
138 erroneously found still-pending ones.
140 JH/08 Pipeline CHUNKING command and data together, on kernels that support
141 MSG_MORE. Only in-clear (not on TLS connections).
143 JH/09 Avoid using a temporary file during transport using dkim. Unless a
144 transport-filter is involved we can buffer the headers in memory for
145 creating the signature, and read the spool data file once for the
146 signature and again for transmission.
148 JH/10 Enable use of sendfile in Linux builds as default. It was disabled in
149 4.77 as the kernel support then wasn't solid, having issues in 64bit
150 mode. Now, it's been long enough. Add support for FreeBSD also.
152 JH/11 Bug 2104: Fix continued use of a transport connection with TLS. In the
153 case where the routing stage had gathered several addresses to send to
154 a host before calling the transport for the first, we previously failed
155 to close down TLS in the old transport process before passing the TCP
156 connection to the new process. The new one sent a STARTTLS command
157 which naturally failed, giving a failed delivery and bloating the retry
158 database. Investigation and fix prototype from Wolfgang Breyha.
160 JH/12 Fix check on SMTP command input synchronisation. Previously there were
161 false-negatives in the check that the sender had not preempted a response
162 or prompt from Exim (running as a server), due to that code's lack of
163 awareness of the SMTP input buffering.
165 PP/04 Add commandline_checks_require_admin option.
166 Exim drops privileges sanely, various checks such as -be aren't a
167 security problem, as long as you trust local users with access to their
168 own account. When invoked by services which pass untrusted data to
169 Exim, this might be an issue. Set this option in main configuration
170 AND make fixes to the calling application, such as using `--` to stop
173 JH/13 Do pipelining under TLS. Previously, although safe, no advantage was
174 taken. Now take care to pack both (client) MAIL,RCPT,DATA, and (server)
175 responses to those, into a single TLS record each way (this usually means
176 a single packet). As a side issue, smtp_enforce_sync now works on TLS
179 PP/05 OpenSSL/1.1: use DH_bits() for more accurate DH param sizes. This
180 affects you only if you're dancing at the edge of the param size limits.
181 If you are, and this message makes sense to you, then: raise the
182 configured limit or use OpenSSL 1.1. Nothing we can do for older
185 JH/14 For the "sock" variant of the malware scanner interface, accept an empty
186 cmdline element to get the documented default one. Previously it was
189 JH/15 Fix a crash in the smtp transport caused when two hosts in succession
190 are unsuable for non-message-specific reasons - eg. connection timeout,
191 banner-time rejection.
193 JH/16 Fix logging of delivery remote port, when specified by router, under
196 PP/06 Repair manualroute's ability to take options in any order, even if one
197 is the name of a transport.
200 HS/01 Cleanup, prevent repeated use of -p/-oMr (CVE-2017-1000369)
202 JH/17 Change the list-building routines interface to use the expanding-string
203 triplet model, for better allocation and copying behaviour.
205 JH/18 Prebuild the data-structure for "builtin" macros, for faster startup.
206 Previously it was constructed the first time a possibly-matching string
207 was met in the configuration file input during startup; now it is done
210 JH/19 Bug 2141: Use the full-complex API for Berkeley DB rather than the legacy-
211 compatible one, to avoid the (poorly documented) possibility of a config
212 file in the working directory redirecting the DB files, possibly correpting
213 some existing file. CVE-2017-10140 assigned for BDB.
215 JH/20 Bug 2147: Do not defer for a verify-with-callout-and-random which is not
216 cache-hot. Previously, although the result was properly cached, the
217 initial verify call returned a defer.
219 JH/21 Bug 2151: Avoid using SIZE on the MAIL for a callout verify, on any but
220 the main verify for receipient in uncached-mode.
222 JH/22 Retire historical build files to an "unsupported" subdir. These are
223 defined as "ones for which we have no current evidence of testing".
225 JH/23 DKIM: enforce the DNS pubkey record "h" permitted-hashes optional field,
226 if present. Previously it was ignored.
228 JH/24 Start using specified-initialisers in C structure init coding. This is
229 a C99 feature (it's 2017, so now considered safe).
231 JH/25 Use one-bit bitfields for flags in the "addr" data structure. Previously
232 if was a fixed-sized field and bitmask ops via macros; it is now more
235 PP/07 GitHub PR 56: Apply MariaDB build fix.
236 Patch provided by Jaroslav Škarvada.
238 PP/08 Bug 2161: Fix regression in sieve quoted-printable handling introduced
239 during Coverity cleanups [4.87 JH/47]
240 Diagnosis and fix provided by Michael Fischer v. Mollard.
242 JH/26 Fix DKIM bug: when the pseudoheader generated for signing was exactly
243 the right size to place the terminating semicolon on its own folded
244 line, the header hash was calculated to an incorrect value thanks to
245 the (relaxed) space the fold became.
247 HS/02 Fix Bug 2130: large writes from the transport subprocess where chunked
248 and confused the parent.
250 JH/27 Fix SOCKS bug: an unitialized pointer was deref'd by the transport process
251 which could crash as a result. This could lead to undeliverable messages.
253 JH/28 Logging: "next input sent too soon" now shows where input was truncated
256 JH/29 Fix queue_run_in_order to ignore the PID portion of the message ID. This
257 matters on fast-turnover and PID-randomising systems, which were getting
258 out-of-order delivery.
260 JH/30 Fix a logging bug on aarch64: an unsafe routine was previously used for
261 a possibly-overlapping copy. The symptom was that "Remote host closed
262 connection in response to HELO" was logged instead of the actual 4xx
265 JH/31 Fix CHUNKING code to properly flush the unwanted chunk after an error.
266 Previously only that bufferd was discarded, resulting in SYMTP command
269 JH/32 DKIM: when a message has multiple signatures matching an identity given
270 in dkim_verify_signers, run the dkim acl once for each. Previously only
271 one run was done. Bug 2189.
273 JH/33 Downgrade an unfound-list name (usually a typo in the config file) from
274 "panic the current process" to "deliberately defer". The panic log is
275 still written with the problem list name; the mail and reject logs now
276 get a temp-reject line for the message that was being handled, saying
277 something like "domains check lookup or other defer". The SMTP 451
278 message is still "Temporary local problem".
280 JH/34 Bug 2199: Fix a use-after-free while reading smtp input for header lines.
281 A crafted sequence of BDAT commands could result in in-use memory beeing
282 freed. CVE-2017-16943.
284 HS/03 Bug 2201: Fix checking for leading-dot on a line during headers reading
285 from SMTP input. Previously it was always done; now only done for DATA
286 and not BDAT commands. CVE-2017-16944.
288 JH/35 Bug 2201: Flush received data in BDAT mode after detecting an error fatal
289 to the message (such as an overlong header line). Previously this was
290 not done and we did not exit BDAT mode. Followon from the previous item
291 though a different problem.
297 JH/01 Bug 1922: Support IDNA2008. This has slightly different conversion rules
298 than -2003 did; needs libidn2 in addition to libidn.
300 JH/02 The path option on a pipe transport is now expanded before use.
302 PP/01 GitHub PR 50: Do not call ldap_start_tls_s on ldapi:// connections.
303 Patch provided by "Björn", documentation fix added too.
305 JH/03 Bug 2003: fix Proxy Protocol v2 handling: the address size field was
306 missing a wire-to-host endian conversion.
308 JH/04 Bug 2004: fix CHUNKING in non-PIPELINEING mode. Chunk data following
309 close after a BDAT command line could be taken as a following command,
310 giving a synch failure. Fix by only checking for synch immediately
311 before acknowledging the chunk.
313 PP/02 GitHub PR 52: many spelling fixes, which include fixing parsing of
314 no_require_dnssec option and creation of _HAVE_TRANSPORT_APPEND_MAILDIR
315 macro. Patches provided by Josh Soref.
317 JH/05 Have the EHLO response advertise VRFY, if there is a vrfy ACL defined.
318 Previously we did not; the RFC seems ambiguous and VRFY is not listed
319 by IANA as a service extension. However, John Klensin suggests that we
322 JH/06 Bug 2017: Fix DKIM verification in -bh test mode. The data feed into
323 the dkim code may be unix-mode line endings rather than smtp wire-format
324 CRLF, so prepend a CR to any bare LF.
326 JH/07 Rationalise the coding for callout smtp conversations and transport ones.
327 As a side-benfit, callouts can now use PIPELINING hence fewer round-trips.
329 JH/08 Bug 2016: Fix DKIM verification vs. CHUNKING. Any BDAT commands after
330 the first were themselves being wrongly included in the feed into dkim
331 processing; with most chunk sizes in use this resulted in an incorrect
332 body hash calculated value.
334 JH/09 Bug 2014: permit inclusion of a DKIM-Signature header in a received
335 DKIM signature block, for verification. Although advised against by
336 standards it is specifically not ruled illegal.
338 JH/10 Bug 2025: Fix reception of (quoted) local-parts with embedded spaces.
340 JH/11 Bug 2029: Fix crash in DKIM verification when a message signature block is
341 missing a body hash (the bh= tag).
343 JH/12 Bug 2018: Re-order Proxy Protocol startup versus TLS-on-connect startup.
344 It seems that HAProxy sends the Proxy Protocol information in clear and
345 only then does a TLS startup, so do the same.
347 JH/13 Bug 2027: Avoid attempting to use TCP Fast Open for non-transport client
348 TCP connections (such as for Spamd) unless the daemon successfully set
349 Fast Open mode on its listening sockets. This fixes breakage seen on
350 too-old kernels or those not configured for Fast Open, at the cost of
351 requiring both directions being enabled for TFO, and TFO never being used
352 by non-daemon-related Exim processes.
354 JH/14 Bug 2000: Reject messages recieved with CHUNKING but with malformed line
355 endings, at least on the first header line. Try to canonify any that get
356 past that check, despite the cost.
358 JH/15 Angle-bracket nesting (an error inserted by broken sendmails) levels are
359 now limited to an arbitrary five deep, while parsing addresses with the
360 strip_excess_angle_brackets option enabled.
362 PP/03 Bug 2018: For Proxy Protocol and TLS-on-connect, do not over-read and
363 instead leave the unprompted TLS handshake in socket buffer for the
364 TLS library to consume.
366 PP/04 Bug 2018: Also handle Proxy Protocol v2 safely.
368 PP/05 FreeBSD compat: handle that Ports no longer create /usr/bin/perl
370 JH/16 Drop variables when they go out of scope. Memory management drops a whole
371 region in one operation, for speed, and this leaves assigned pointers
372 dangling. Add checks run only under the testsuite which checks all
373 variables at a store-reset and panics on a dangling pointer; add code
374 explicitly nulling out all the variables discovered. Fixes one known
375 bug: a transport crash, where a dangling pointer for $sending_ip_address
376 originally assigned in a verify callout, is re-used.
378 PP/06 Drop '.' from @INC in various Perl scripts.
380 PP/07 Switch FreeBSD iconv to always use the base-system libc functions.
382 PP/08 Reduce a number of compilation warnings under clang; building with
383 CC=clang CFLAGS+=-Wno-dangling-else -Wno-logical-op-parentheses
384 should be warning-free.
386 JH/17 Fix inbound CHUNKING when DKIM disabled at runtime.
388 HS/01 Fix portability problems introduced by PP/08 for platforms where
389 realloc(NULL) is not equivalent to malloc() [SunOS et al].
391 HS/02 Bug 1974: Fix missing line terminator on the last received BDAT
392 chunk. This allows us to accept broken chunked messages. We need a more
393 general solution here.
395 PP/09 Wrote util/chunking_fixqueue_finalnewlines.pl to help recover
396 already-broken messages in the queue.
398 JH/18 Bug 2061: Fix ${extract } corrupting an enclosing ${reduce } $value.
400 JH/19 Fix reference counting bug in routing-generated-address tracking.
406 JH/01 Use SIZE on MAIL FROM in a cutthrough connection, if the destination
407 supports it and a size is available (ie. the sending peer gave us one).
409 JH/02 The obsolete acl condition "demime" is removed (finally, after ten
410 years of being deprecated). The replacements are the ACLs
411 acl_smtp_mime and acl_not_smtp_mime.
413 JH/03 Upgrade security requirements imposed for hosts_try_dane: previously
414 a downgraded non-dane trust-anchor for the TLS connection (CA-style)
415 or even an in-clear connection were permitted. Now, if the host lookup
416 was dnssec and dane was requested then the host is only used if the
417 TLSA lookup succeeds and is dnssec. Further hosts (eg. lower priority
418 MXs) will be tried (for hosts_try_dane though not for hosts_require_dane)
419 if one fails this test.
420 This means that a poorly-configured remote DNS will make it incommunicado;
421 but it protects against a DNS-interception attack on it.
423 JH/04 Bug 1810: make continued-use of an open smtp transport connection
424 non-noisy when a race steals the message being considered.
426 JH/05 If main configuration option tls_certificate is unset, generate a
427 self-signed certificate for inbound TLS connections.
429 JH/06 Bug 165: hide more cases of password exposure - this time in expansions
430 in rewrites and routers.
432 JH/07 Retire gnutls_require_mac et.al. These were nonfunctional since 4.80
433 and logged a warning sing 4.83; now they are a configuration file error.
435 JH/08 Bug 1836: Fix crash in VRFY handling when handed an unqualified name
436 (lacking @domain). Apply the same qualification processing as RCPT.
438 JH/09 Bug 1804: Avoid writing msglog files when in -bh or -bhc mode.
440 JH/10 Support ${sha256:} applied to a string (as well as the previous
443 JH/11 Cutthrough: avoid using the callout hints db on a verify callout when
444 a cutthrough deliver is pending, as we always want to make a connection.
445 This also avoids re-routing the message when later placing the cutthrough
446 connection after a verify cache hit.
447 Do not update it with the verify result either.
449 JH/12 Cutthrough: disable when verify option success_on_redirect is used, and
450 when routing results in more than one destination address.
452 JH/13 Cutthrough: expand transport dkim_domain option when testing for dkim
453 signing (which inhibits the cutthrough capability). Previously only
454 the presence of an option was tested; now an expansion evaluating as
455 empty is permissible (obviously it should depend only on data available
456 when the cutthrough connection is made).
458 JH/14 Fix logging of errors under PIPELINING. Previously the log line giving
459 the relevant preceding SMTP command did not note the pipelining mode.
461 JH/15 Fix counting of empty lines in $body_linecount and $message_linecount.
462 Previously they were not counted.
464 JH/16 DANE: treat a TLSA lookup response having all non-TLSA RRs, the same
465 as one having no matching records. Previously we deferred the message
466 that needed the lookup.
468 JH/17 Fakereject: previously logged as a normal message arrival "<="; now
469 distinguished as "(=".
471 JH/18 Bug 1867: make the fail_defer_domains option on a dnslookup router work
472 for missing MX records. Previously it only worked for missing A records.
474 JH/19 Bug 1850: support Radius libraries that return REJECT_RC.
476 JH/20 Bug 1872: Ensure that acl_smtp_notquit is run when the connection drops
477 after the data-go-ahead and data-ack. Patch from Jason Betts.
479 JH/21 Bug 1846: Send DMARC forensic reports for reject and quarantine results,
480 even for a "none" policy. Patch from Tony Meyer.
482 JH/22 Fix continued use of a connection for further deliveries. If a port was
483 specified by a router, it must also match for the delivery to be
486 JH/23 Bug 1874: fix continued use of a connection for further deliveries.
487 When one of the recipients of a message was unsuitable for the connection
488 (has no matching addresses), we lost track of needing to mark it
489 deferred. As a result mail would be lost.
491 JH/24 Bug 1832: Log EHLO response on getting conn-close response for HELO.
493 JH/25 Decoding ACL controls is now done using a binary search; the source code
494 takes up less space and should be simpler to maintain. Merge the ACL
495 condition decode tables also, with similar effect.
497 JH/26 Fix problem with one_time used on a redirect router which returned the
498 parent address unchanged. A retry would see the parent address marked as
499 delivered, so not attempt the (identical) child. As a result mail would
502 JH/27 Fix a possible security hole, wherein a process operating with the Exim
503 UID can gain a root shell. Credit to http://www.halfdog.net/ for
504 discovery and writeup. Ubuntu bug 1580454; no bug raised against Exim
507 JH/28 Enable {spool,log} filesystem space and inode checks as default.
508 Main config options check_{log,spool}_{inodes,space} are now
509 100 inodes, 10MB unless set otherwise in the configuration.
511 JH/29 Fix the connection_reject log selector to apply to the connect ACL.
512 Previously it only applied to the main-section connection policy
515 JH/30 Bug 1897: fix callouts connection fallback from TLS to cleartext.
517 PP/01 Changed default Diffie-Hellman parameters to be Exim-specific, created
518 by me. Added RFC7919 DH primes as an alternative.
520 PP/02 Unbreak build via pkg-config with new hash support when crypto headers
521 are not in the system include path.
523 JH/31 Fix longstanding bug with aborted TLS server connection handling. Under
524 GnuTLS, when a session startup failed (eg because the client disconnected)
525 Exim did stdio operations after fclose. This was exposed by a recent
526 change which nulled out the file handle after the fclose.
528 JH/32 Bug 1909: Fix OCSP proof verification for cases where the proof is
529 signed directly by the cert-signing cert, rather than an intermediate
530 OCSP-signing cert. This is the model used by LetsEncrypt.
532 JH/33 Bug 1914: Ensure socket is nonblocking before draining after SMTP QUIT.
534 HS/01 Fix leak in verify callout under GnuTLS, about 3MB per recipient on
535 an incoming connection.
537 HS/02 Bug 1802: Do not half-close the connection after sending a request
540 HS/03 Use "auto" as the default EC curve parameter. For OpenSSL < 1.0.2
541 fallback to "prime256v1".
543 JH/34 SECURITY: Use proper copy of DATA command in error message.
544 Could leak key material. Remotely exploitable. CVE-2016-9963.
550 JH/01 Bug 1664: Disable OCSP for GnuTLS library versions at/before 3.3.16
551 and 3.4.4 - once the server is enabled to respond to an OCSP request
552 it does even when not requested, resulting in a stapling non-aware
553 client dropping the TLS connection.
555 TF/01 Code cleanup: Overhaul the debug_selector and log_selector machinery to
556 support variable-length bit vectors. No functional change.
558 TF/02 Improve the consistency of logging incoming and outgoing interfaces.
559 The I= interface field on outgoing lines is now after the H= remote
560 host field, same as incoming lines. There is a separate
561 outgoing_interface log selector which allows you to disable the
564 JH/02 Bug 728: Close logfiles after a daemon-process "exceptional" log write.
565 If not running log_selector +smtp_connection the mainlog would be held
566 open indefinitely after a "too many connections" event, including to a
567 deleted file after a log rotate. Leave the per net connection logging
568 leaving it open for efficiency as that will be quickly detected by the
569 check on the next write.
571 HS/01 Bug 1671: Fix post transport crash.
572 Processing the wait-<transport> messages could crash the delivery
573 process if the message IDs didn't exist for some reason. When
574 using 'split_spool_directory=yes' the construction of the spool
575 file name failed already, exposing the same netto behaviour.
577 JH/03 Bug 425: Capture substrings in $regex1, $regex2 etc from regex &
578 mime_regex ACL conditions.
580 JH/04 Bug 1686: When compiled with EXPERIMENTAL_DSN_INFO: Add extra information
581 to DSN fail messages (bounces): remote IP, remote greeting, remote response
582 to HELO, local diagnostic string.
584 JH/05 Downgrade message for a TLS-certificate-based authentication fail from
585 log line to debug. Even when configured with a tls authenticator many
586 client connections are expected to not authenticate in this way, so
587 an authenticate fail is not an error.
589 HS/02 Add the Exim version string to the process info. This way exiwhat
590 gives some more detail about the running daemon.
592 JH/06 Bug 1395: time-limit caching of DNS lookups, to the TTL value. This may
593 matter for fast-change records such as DNSBLs.
595 JH/07 Bug 1678: Always record an interface option value, if set, as part of a
596 retry record, even if constant. There may be multiple transports with
597 different interface settings and the retry behaviour needs to be kept
600 JH/08 Bug 1586: exiqgrep now refuses to run if there are unexpected arguments.
602 JH/09 Bug 1700: ignore space & tab embedded in base64 during decode.
604 JH/10 Bug 840: fix log_defer_output option of pipe transport
606 JH/11 Bug 830: use same host for all RCPTS of a message, even under
607 hosts_randomize. This matters a lot when combined with mua_wrapper.
609 JH/12 Bug 1706: percent and underbar characters are no longer escaped by the
610 ${quote_pgsql:<string>} operator.
612 JH/13 Bug 1708: avoid misaligned access in cached lookup.
614 JH/14 Change header file name for freeradius-client. Relevant if compiling
615 with Radius support; from the Gentoo tree and checked under Fedora.
617 JH/15 Bug 1712: Introduce $prdr_requested flag variable
619 JH/16 Bug 1714: Permit an empty string as expansion result for transport
620 option transport_filter, meaning no filtering.
622 JH/17 Bug 1713: Fix non-PDKIM_DEBUG build. Patch from Jasen Betts.
624 JH/18 Bug 1709: When built with TLS support, the tls_advertise_hosts option now
625 defaults to "*" (all hosts). The variable is now available when not built
626 with TLS, default unset, mainly to enable keeping the testsuite sane.
627 If a server certificate is not supplied (via tls_certificate) an error is
628 logged, and clients will find TLS connections fail on startup. Presumably
629 they will retry in-clear.
630 Packagers of Exim are strongly encouraged to create a server certificate
631 at installation time.
633 HS/03 Add -bP config_file as a synonym for -bP configure_file, for consistency
634 with the $config_file variable.
636 JH/19 Two additional event types: msg:rcpt:defer and msg:rcpt:host:defer. Both
637 in transport context, after the attempt, and per-recipient. The latter type
638 is per host attempted. The event data is the error message, and the errno
639 information encodes the lookup type (A vs. MX) used for the (first) host,
640 and the trailing two digits of the smtp 4xx response.
642 GF/01 Bug 1715: Fix for race condition in exicyclog, where exim could attempt
643 to write to mainlog (or rejectlog, paniclog) in the window between file
644 creation and permissions/ownership being changed. Particularly affects
645 installations where exicyclog is run as root, rather than exim user;
646 result is that the running daemon panics and dies.
648 JH/20 Bug 1701: For MySQL lookups, support MySQL config file option group names.
650 JH/21 Bug 1720: Add support for priority groups and weighted-random proxy
651 selection for the EXPERIMENTAL_SOCKS feature, via new per-proxy options
652 "pri" and "weight". Note that the previous implicit priority given by the
653 list order is no longer honoured.
655 JH/22 Bugs 963, 1721: Fix some corner cases in message body canonicalization
658 JH/23 Move SOCKS5 support from Experimental to mainline, enabled for a build
659 by defining SUPPORT_SOCKS.
661 JH/26 Move PROXY support from Experimental to mainline, enabled for a build
662 by defining SUPPORT_PROXY. Note that the proxy_required_hosts option
663 is renamed to hosts_proxy, and the proxy_{host,target}_{address,port}.
664 variables are renamed to proxy_{local,external}_{address,port}.
666 JH/27 Move Internationalisation support from Experimental to mainline, enabled
667 for a build by defining SUPPORT_I18N
669 JH/28 Bug 1745: Fix redis lookups to handle (quoted) spaces embedded in parts
670 of the query string, and make ${quote_redis:} do that quoting.
672 JH/29 Move Events support from Experimental to mainline, enabled by default
673 and removable for a build by defining DISABLE_EVENT.
675 JH/30 Updated DANE implementation code to current from Viktor Dukhovni.
677 JH/31 Fix bug with hosts_connection_nolog and named-lists which were wrongly
678 cached by the daemon.
680 JH/32 Move Redis support from Experimental to mainline, enabled for a build
681 by defining LOOKUP_REDIS. The libhiredis library is required.
683 JH/33 Bug 1748: Permit ACL dnslists= condition in non-smtp ACLs if explicit
684 keys are given for lookup.
686 JH/34 Bug 1192: replace the embedded copy of PolarSSL RSA routines in the DKIM
687 support, by using OpenSSL or GnuTLS library ones. This means DKIM is
688 only supported when built with TLS support. The PolarSSL SHA routines
689 are still used when the TLS library is too old for convenient support.
691 JH/35 Require SINGLE_DH_USE by default in OpenSSL (main config option
692 openssl_options), for security. OpenSSL forces this from version 1.1.0
693 server-side so match that on older versions.
695 JH/36 Bug 1778: longstanding bug in memory use by the ${run } expansion: A fresh
696 allocation for $value could be released as the expansion processing
697 concluded, but leaving the global pointer active for it.
699 JH/37 Bug 1769: Permit a VRFY ACL to override the default 252 response,
700 and to use the domains and local_parts ACL conditions.
702 JH/38 Fix cutthrough bug with body lines having a single dot. The dot was
703 incorrectly not doubled on cutthrough transmission, hence seen as a
704 body-termination at the receiving system - resulting in truncated mails.
705 Commonly the sender saw a TCP-level error, and retransmitted the message
706 via the normal store-and-forward channel. This could result in duplicates
707 received - but deduplicating mailstores were liable to retain only the
708 initial truncated version.
710 JH/39 Bug 1781: Fix use of DKIM private-keys having trailing '=' in the base-64.
712 JH/40 Fix crash in queryprogram router when compiled with EXPERIMENTAL_SRS.
714 JH/41 Bug 1792: Fix selection of headers to sign for DKIM: bottom-up. While
715 we're in there, support oversigning also; bug 1309.
717 JH/42 Bug 1796: Fix error logged on a malware scanner connection failure.
719 HS/04 Add support for keep_environment and add_environment options.
721 JH/43 Tidy coding issues detected by gcc --fsanitize=undefined. Some remain;
722 either intentional arithmetic overflow during PRNG, or testing config-
725 JH/44 Bug 1800: The combination of a -bhc commandline option and cutthrough
726 delivery resulted in actual delivery. Cancel cutthrough before DATA
729 JH/45 Fix cutthrough, when connection not opened by verify and target hard-
730 rejects a recipient: pass the reject to the originator.
732 JH/46 Multiple issues raised by Coverity. Some were obvious or plausible bugs.
733 Many were false-positives and ignorable, but it's worth fixing the
736 JH/47 Fix build on HP-UX and older Solaris, which need (un)setenv now also
737 for the new environment-manipulation done at startup. Move the routines
738 from being local to tls.c to being global via the os.c file.
740 JH/48 Bug 1807: Fix ${extract } for the numeric/3-string case. While preparsing
741 an extract embedded as result-arg for a map, the first arg for extract
742 is unavailable so we cannot tell if this is a numbered or keyed
743 extraction. Accept either.
749 JH/01 Bug 1545: The smtp transport option "retry_include_ip_address" is now
752 JH/02 The smtp transport option "multi_domain" is now expanded.
754 JH/03 The smtp transport now requests PRDR by default, if the server offers
757 JH/04 Certificate name checking on server certificates, when exim is a client,
758 is now done by default. The transport option tls_verify_cert_hostnames
759 can be used to disable this per-host. The build option
760 EXPERIMENTAL_CERTNAMES is withdrawn.
762 JH/05 The value of the tls_verify_certificates smtp transport and main options
763 default to the word "system" to access the system default CA bundle.
764 For GnuTLS, only version 3.0.20 or later.
766 JH/06 Verification of the server certificate for a TLS connection is now tried
767 (but not required) by default. The verification status is now logged by
768 default, for both outbound TLS and client-certificate supplying inbound
771 JH/07 Changed the default rfc1413 lookup settings to disable calls. Few
774 JH/08 The EXPERIMENTAL_DSN compile option is no longer needed; all Delivery
775 Status Notification (bounce) messages are now MIME format per RFC 3464.
776 Support for RFC 3461 DSN options NOTIFY,ENVID,RET,ORCPT can be advertised
777 under the control of the dsn_advertise_hosts option, and routers may
778 have a dsn_lasthop option.
780 JH/09 A timeout of 2 minutes is now applied to all malware scanner types by
781 default, modifiable by a malware= option. The list separator for
782 the options can now be changed in the usual way. Bug 68.
784 JH/10 The smtp_receive_timeout main option is now expanded before use.
786 JH/11 The incoming_interface log option now also enables logging of the
787 local interface on delivery outgoing connections.
789 JH/12 The cutthrough-routing facility now supports multi-recipient mails,
790 if the interface and destination host and port all match.
792 JH/13 Bug 344: The verify = reverse_host_lookup ACL condition now accepts a
795 JH/14 Bug 1573: The spam= ACL condition now additionally supports Rspamd.
796 Patch from Andrew Lewis.
798 JH/15 Bug 670: The spamd_address main option (for the spam= ACL condition)
799 now supports optional time-restrictions, weighting, and priority
800 modifiers per server. Patch originally by <rommer@active.by>.
802 JH/16 The spamd_address main option now supports a mixed list of local
803 and remote servers. Remote servers can be IPv6 addresses, and
804 specify a port-range.
806 JH/17 Bug 68: The spamd_address main option now supports an optional
807 timeout value per server.
809 JH/18 Bug 1581: Router and transport options headers_add/remove can
810 now have the list separator specified.
812 JH/19 Bug 392: spamd_address, and clamd av_scanner, now support retry
815 JH/20 Bug 1571: Ensure that $tls_in_peerdn is set, when verification fails
818 JH/21 Support for the A6 type of dns record is withdrawn.
820 JH/22 Bug 608: The result of a QUIT or not-QUIT toplevel ACL now matters
821 rather than the verbs used.
823 JH/23 Bug 1572: Increase limit on SMTP confirmation message copy size
824 from 255 to 1024 chars.
826 JH/24 Verification callouts now attempt to use TLS by default.
828 HS/01 DNSSEC options (dnssec_require_domains, dnssec_request_domains)
829 are generic router options now. The defaults didn't change.
831 JH/25 Bug 466: Add RFC2322 support for MIME attachment filenames.
832 Original patch from Alexander Shikoff, worked over by JH.
834 HS/02 Bug 1575: exigrep falls back to autodetection of compressed
835 files if ZCAT_COMMAND is not executable.
837 JH/26 Bug 1539: Add timeout/retry options on dnsdb lookups.
839 JH/27 Bug 286: Support SOA lookup in dnsdb lookups.
841 JH/28 Bug 1588: Do not use the A lookup following an AAAA for setting the FQDN.
842 Normally benign, it bites when the pair was led to by a CNAME;
843 modern usage is to not canonicalize the domain to a CNAME target
844 (and we were inconsistent anyway for A-only vs AAAA+A).
846 JH/29 Bug 1632: Removed the word "rejected" from line logged for ACL discards.
848 JH/30 Check the forward DNS lookup for DNSSEC, in addition to the reverse,
849 when evaluating $sender_host_dnssec.
851 JH/31 Check the HELO verification lookup for DNSSEC, adding new
852 $sender_helo_dnssec variable.
854 JH/32 Bug 1397: Enable ECDHE on OpenSSL, just the NIST P-256 curve.
856 JH/33 Bug 1346: Note MAIL cmd seen in -bS batch, to avoid smtp_no_mail log.
858 JH/34 Bug 1648: Fix a memory leak seen with "mailq" and large queues.
860 JH/35 Bug 1642: Fix support of $spam_ variables at delivery time. Was
861 documented as working, but never had. Support all but $spam_report.
863 JH/36 Bug 1659: Guard checking of input smtp commands again pseudo-command
864 added for tls authenticator.
866 HS/03 Add perl_taintmode main config option
872 TL/01 When running the test suite, the README says that variables such as
873 no_msglog_check are global and can be placed anywhere in a specific
874 test's script, however it was observed that placement needed to be near
875 the beginning for it to behave that way. Changed the runtest perl
876 script to read through the entire script once to detect and set these
877 variables, reset to the beginning of the script, and then run through
878 the script parsing/test process like normal.
880 TL/02 The BSD's have an arc4random API. One of the functions to induce
881 adding randomness was arc4random_stir(), but it has been removed in
882 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
883 function when detected.
885 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
886 cause callback expansion.
888 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
889 syntax errors in an expansion can be treated as a string instead of
890 logging or causing an error, due to the internal use of bool_lax
891 instead of bool when processing it.
893 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
894 server certificates when making smtp deliveries.
896 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
898 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
900 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
902 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
903 Merged patch from Sebastian Wiedenroth.
905 JH/05 Fix results-pipe from transport process. Several recipients, combined
906 with certificate use, exposed issues where response data items split
907 over buffer boundaries were not parsed properly. This eventually
908 resulted in duplicates being sent. This issue only became common enough
909 to notice due to the introduction of connection certificate information,
910 the item size being so much larger. Found and fixed by Wolfgang Breyha.
912 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
913 size buffer was used, resulting in syntax errors when an expansion
916 JH/07 Add support for directories of certificates when compiled with a GnuTLS
917 version 3.3.6 or later.
919 JH/08 Rename the TPDA experimental facility to Event Actions. The #ifdef
920 is EXPERIMENTAL_EVENT, the main-configuration and transport options
921 both become "event_action", the variables become $event_name, $event_data
922 and $event_defer_errno. There is a new variable $verify_mode, usable in
923 routers, transports and related events. The tls:cert event is now also
924 raised for inbound connections, if the main configuration event_action
927 TL/06 In test suite, disable OCSP for old versions of openssl which contained
928 early OCSP support, but no stapling (appears to be less than 1.0.0).
930 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
931 server certificate names available under the smtp transport option
932 "tls_verify_cert_hostname" now do not permit multi-component wildcard
935 JH/10 Time-related extraction expansions from certificates now use the main
936 option "timezone" setting for output formatting, and are consistent
937 between OpenSSL and GnuTLS compilations. Bug 1541.
939 JH/11 Fix a crash in mime ACL when meeting a zero-length, quoted or RFC2047-
940 encoded parameter in the incoming message. Bug 1558.
942 JH/12 Bug 1527: Autogrow buffer used in reading spool files. Since they now
943 include certificate info, eximon was claiming there were spoolfile
946 JH/13 Bug 1521: Fix ldap lookup for single-attr request, multiple-attr return.
948 JH/14 Log delivery-related information more consistently, using the sequence
949 "H=<name> [<ip>]" wherever possible.
951 TL/07 Bug 1547: Omit RFCs from release. Draft and RFCs have licenses which
952 are problematic for Debian distribution, omit them from the release
955 JH/15 Updates and fixes to the EXPERIMENTAL_DSN feature.
957 JH/16 Fix string representation of time values on 64bit time_t architectures.
960 JH/17 Fix a null-indirection in certextract expansions when a nondefault
961 output list separator was used.
966 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
967 checkers that were complaining about end of non-void function with no
970 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
971 This was a regression introduced in 4.83 by another bugfix.
973 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
975 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
976 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
982 TF/01 Correctly close the server side of TLS when forking for delivery.
984 When a message was received over SMTP with TLS, Exim failed to clear up
985 the incoming connection properly after forking off the child process to
986 deliver the message. In some situations the subsequent outgoing
987 delivery connection happened to have the same fd number as the incoming
988 connection previously had. Exim would try to use TLS and fail, logging
989 a "Bad file descriptor" error.
991 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
992 utilities have not been installed.
994 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
995 temporary space as the ACL may create new global variables.
997 TL/01 LDAP support uses per connection or global context settings, depending
998 upon the detected version of the libraries at build time.
1000 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
1001 to extract and use the src ip:port in logging and expansions as if it
1002 were a direct connection from the outside internet. PPv2 support was
1003 updated based on HAProxy spec change in May 2014.
1005 JH/02 Add ${listextract {number}{list}{success}{fail}}.
1007 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
1008 Properly escape header and check for NULL return.
1010 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
1013 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
1015 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
1016 characters in header names, implemented as a verify condition.
1017 Contributed by Michael Fischer v. Mollard.
1019 TL/05 Rename SPF condition results err_perm and err_temp to standardized
1020 results permerror and temperror. Previous values are deprecated but
1021 still accepted. In a future release, err_perm and err_temp will be
1022 completely removed, which will be a backward incompatibility if the
1023 ACL tests for either of these two old results. Patch contributed by
1024 user bes-internal on the mailing list.
1026 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
1028 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
1029 selectors, in both main and reject logs.
1031 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
1034 JH/07 Add malware type "sock" for talking to simple daemon.
1036 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
1038 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
1039 routers/transports under cutthrough routing.
1041 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
1042 numbers. Touch up "bool" conditional to keep the same definition.
1044 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
1046 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
1048 JH/12 Expand items in router/transport headers_add or headers_remove lists
1049 individually rather than the list as a whole. Bug 1452.
1051 Required for reasonable handling of multiple headers_ options when
1052 they may be empty; requires that headers_remove items with embedded
1053 colons must have them doubled (or the list-separator changed).
1055 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
1056 view the policy declared in the DMARC record. Currently, $dmarc_status
1057 is a combined value of both the record presence and the result of the
1060 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
1062 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
1063 dnslookup router and the smtp transport (applying to the forward
1066 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
1067 of ldap servers used for a specific lookup. Patch provided by Heiko
1070 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
1071 New variable $lookup_dnssec_authenticated for observability.
1073 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
1074 Patch submitted by Lars Timman.
1076 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
1078 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
1079 Requires trusted mode and valid format message id, aborts otherwise.
1080 Patch contributed by Heiko Schlichting.
1082 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
1083 certextract with support for various fields. Bug 1358.
1085 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
1086 is requested by default, modifiable by smtp transport option
1089 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
1090 operate on certificate variables to give certificate fingerprints
1091 Also new ${sha256:cert_variable}.
1093 JH/23 The PRDR feature is moved from being Experimental into the mainline.
1095 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
1096 Christian Aistleitner.
1098 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
1100 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
1101 file. Patch from Wolfgang Breyha.
1103 JH/25 Expand the coverage of the delivery $host and $host_address to
1104 client authenticators run in verify callout. Bug 1476.
1106 JH/26 Port service names are now accepted for tls_on_connect_ports, to
1107 align with daemon_smtp_ports. Bug 72.
1109 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
1110 support and error reporting did not work properly.
1112 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
1113 and is readable. Patch from Andrew Colin Kissa.
1115 TL/14 Enhance documentation of ${run expansion and how it parses the
1116 commandline after expansion, particularly in the case when an
1117 unquoted variable expansion results in an empty value.
1119 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
1121 PP/02 Fix internal collision of T_APL on systems which support RFC3123
1122 by renaming away from it. Addresses GH issue 15, reported by
1125 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
1127 TL/15 SECURITY: prevent double expansion in math comparison functions
1128 (can expand unsanitized data). Not remotely exploitable.
1135 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
1137 PP/02 Make -n do something, by making it not do something.
1138 When combined with -bP, the name of an option is not output.
1140 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
1143 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
1144 $sender_host_name and config options to manage this, and basic check
1147 PP/05 DSCP support for outbound connections and control modifier for inbound.
1149 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
1150 (Only plugin which currently uses this is kerberos4, which nobody should
1151 be using, but we should make it available and other future plugins might
1152 conceivably use it, even though it would break NAT; stuff *should* be
1153 using channel bindings instead).
1155 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
1156 name; added for Sendmail compatibility; requires admin caller.
1157 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
1158 just ignore it); requires trusted caller.
1159 Also parse but ignore: -Ac -Am -X<logfile>
1162 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
1164 TL/02 Add +smtp_confirmation as a default logging option.
1166 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
1167 Patch by Magnus Holmgren from 2007-02-20.
1169 TL/04 Bugzilla 1281 - Spec typo.
1170 Bugzilla 1283 - Spec typo.
1171 Bugzilla 1290 - Spec grammar fixes.
1173 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
1175 TL/06 Add Experimental DMARC support using libopendmarc libraries.
1177 TL/07 Fix an out of order global option causing a segfault. Reported to dev
1178 mailing list by by Dmitry Isaikin.
1180 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
1182 JH/02 Support "G" suffix to numbers in ${if comparisons.
1184 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
1186 NM/01 Bugzilla 1197 - Spec typo
1187 Bugzilla 1196 - Spec examples corrections
1189 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
1191 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
1192 gnutls_enable_pkcs11, but renamed to more accurately indicate its
1195 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
1196 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
1198 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
1199 "acl {{name}{arg}...}", and optional args on acl condition
1202 JH/05 Permit multiple router/transport headers_add/remove lines.
1204 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
1206 JH/07 Avoid using a waiting database for a single-message-only transport.
1207 Performance patch from Paul Fisher. Bugzilla 1262.
1209 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
1212 JH/09 Add $headers_added variable, with content from use of ACL modifier
1213 add_header (but not yet added to the message). Bugzilla 199.
1215 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
1216 Pulled from Bugzilla 817 by Wolfgang Breyha.
1218 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
1220 (nb: this is the same fix as in Exim 4.80.1)
1222 JH/11 Add A= logging on delivery lines, and a client_set_id option on
1225 JH/12 Add optional authenticated_sender logging to A= and a log_selector
1228 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
1230 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
1231 advertise SMTP AUTH mechanism to us, instead of a generic
1232 protocol violation error. Also, make Exim more robust to bad
1233 data from the Dovecot auth socket.
1235 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
1237 When a queue runner is handling a message, Exim first routes the
1238 recipient addresses, during which it prunes them based on the retry
1239 hints database. After that it attempts to deliver the message to
1240 any remaining recipients. It then updates the hints database using
1243 So if a recipient address works intermittently, it can get repeatedly
1244 deferred at routing time. The retry hints record remains fresh so the
1245 address never reaches the final cutoff time.
1247 This is a fairly common occurrence when a user is bumping up against
1248 their storage quota. Exim had some logic in its local delivery code
1249 to deal with this. However it did not apply to per-recipient defers
1250 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
1252 This change adds a proper retry rule check during routing so that the
1253 final cutoff time is checked against the message's age. We only do
1254 this check if there is an address retry record and there is not a
1255 domain retry record; this implies that previous attempts to handle
1256 the address had the retry_use_local_parts option turned on. We use
1257 this as an approximation for the destination being like a local
1258 delivery, as in LMTP.
1260 I suspect this new check makes the old local delivery cutoff check
1261 redundant, but I have not verified this so I left the code in place.
1263 TF/02 Correct gecos expansion when From: is a prefix of the username.
1265 Test 0254 submits a message to Exim with the header
1269 When I ran the test suite under the user fanf2, Exim expanded
1270 the header to contain my full name, whereas it should have added
1271 a Resent-Sender: header. It erroneously treats any prefix of the
1272 username as equal to the username.
1274 This change corrects that bug.
1276 GF/01 DCC debug and logging tidyup
1277 Error conditions log to paniclog rather than rejectlog.
1278 Debug lines prefixed by "DCC: " to remove any ambiguity.
1280 TF/03 Avoid unnecessary rebuilds of lookup-related code.
1282 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
1283 Bug spotted by Jeremy Harris; was flawed since initial commit.
1284 Would have resulted in OCSP responses post-SNI triggering an Exim
1285 NULL dereference and crash.
1287 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
1289 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
1290 Bug detection, analysis and fix by Samuel Thibault.
1291 Bugzilla 1331, Debian bug #698092.
1293 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
1295 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
1296 Server implementation by Todd Lyons, client by JH.
1297 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
1298 config variable "prdr_enable" controls whether the server
1299 advertises the facility. If the client requests PRDR a new
1300 acl_data_smtp_prdr ACL is called once for each recipient, after
1301 the body content is received and before the acl_smtp_data ACL.
1302 The client is controlled by both of: a hosts_try_prdr option
1303 on the smtp transport, and the server advertisement.
1304 Default client logging of deliveries and rejections involving
1305 PRDR are flagged with the string "PRDR".
1307 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
1308 fclose(). Diagnosis by Todd Lyons.
1310 PP/17 Update configure.default to handle IPv6 localhost better.
1311 Patch by Alain Williams (plus minor tweaks).
1314 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
1315 This is now consistent with GnuTLS, and is now documented: the
1316 previous undocumented portable approach to treating the option as
1317 unset was to force an expansion failure. That still works, and
1318 an empty string is now equivalent.
1320 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
1321 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
1322 not performing validation itself.
1324 PP/20 Added force_command boolean option to pipe transport.
1325 Patch from Nick Koston, of cPanel Inc.
1327 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
1330 TF/04 Added udpsend ACL modifier and hexquote expansion operator
1332 PP/21 Fix eximon continuous updating with timestamped log-files.
1333 Broken in a format-string cleanup in 4.80, missed when I repaired the
1334 other false fix of the same issue.
1335 Report and fix from Heiko Schlichting.
1338 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
1339 Report from Prashanth Katuri.
1341 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
1342 It's SecureTransport, so affects any MacOS clients which use the
1343 system-integrated TLS libraries, including email clients.
1345 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
1346 using a MIME ACL for non-SMTP local injection.
1347 Report and assistance in diagnosis by Warren Baker.
1349 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
1351 JH/16 Fix comparisons for 64b. Bugzilla 1385.
1353 TL/09 Add expansion variable $authenticated_fail_id to keep track of
1354 last id that failed so it may be referenced in subsequent ACL's.
1356 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
1359 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
1360 ldap library initialization, allowing self-signed CA's to be
1361 used. Also properly sets require_cert option later in code by
1362 using NULL (global ldap config) instead of ldap handle (per
1363 session). Bug diagnosis and testing by alxgomz.
1365 TL/12 Enhanced documentation in the ratelimit.pl script provided in
1366 the src/util/ subdirectory.
1368 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
1369 renamed to Transport Post Delivery Action by Jeremy Harris, as
1372 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
1373 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
1374 redis_servers = needs to be configured which will be used by the redis
1375 lookup. Patch from Warren Baker, of The Packet Hub.
1377 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
1379 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
1380 hostname or reverse DNS when processing a host list. Used suggestions
1381 from multiple comments on this bug.
1383 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
1385 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
1386 Missed a few lines, added it to make the runtest require no keyboard
1389 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
1390 contains upper case chars. Make router use caseful_local_part.
1392 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
1393 support when GnuTLS has been built with p11-kit.
1399 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
1401 This, or similar/improved, will also be change PP/11 of 4.82.
1407 PP/01 Handle short writes when writing local log-files.
1408 In practice, only affects FreeBSD (8 onwards).
1409 Bugzilla 1053, with thanks to Dmitry Isaikin.
1411 NM/01 Bugzilla 949 - Documentation tweak
1413 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
1416 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
1418 PP/02 Implemented gsasl authenticator.
1420 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
1422 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
1423 `pkg-config foo` for cflags/libs.
1425 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
1426 with rest of GSASL and with heimdal_gssapi.
1428 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
1429 `pkg-config foo` for cflags/libs for the TLS implementation.
1431 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
1432 properties get this fed in as external SSF. A number of robustness
1433 and debugging improvements to the cyrus_sasl authenticator.
1435 PP/08 cyrus_sasl server now expands the server_realm option.
1437 PP/09 Bugzilla 1214 - Log authentication information in reject log.
1438 Patch by Jeremy Harris.
1440 PP/10 Added dbmjz lookup type.
1442 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
1444 PP/12 MAIL args handles TAB as well as SP, for better interop with
1445 non-compliant senders.
1446 Analysis and variant patch by Todd Lyons.
1448 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
1449 Bug report from Lars Müller <lars@samba.org> (via SUSE),
1450 Patch from Dirk Mueller <dmueller@suse.com>
1452 PP/13 tls_peerdn now print-escaped for spool files.
1453 Observed some $tls_peerdn in wild which contained \n, which resulted
1454 in spool file corruption.
1456 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
1457 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
1458 or write after TLS renegotiation, which otherwise led to messages
1461 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
1462 as a tracking header (ie: a signed header comes before the signature).
1463 Patch from Wolfgang Breyha.
1465 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
1466 comma-sep list; embedded commas doubled.
1468 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
1470 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
1472 Report and patch from Dmitry Banschikov.
1474 PP/16 Removed "dont_insert_empty_fragments" from "openssl_options".
1475 Removed SSL_clear() after SSL_new() which led to protocol negotiation
1476 failures. We appear to now support TLS1.1+ with Exim.
1478 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
1479 lets Exim select keys and certificates based upon TLS SNI from client.
1480 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
1481 before an outbound SMTP session. New log_selector, +tls_sni.
1483 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
1484 NULL dereference. Report and patch from Alun Jones.
1486 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
1487 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
1488 Not seeing resolver debug output on NetBSD, but suspect this is a
1489 resolver implementation change.
1491 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
1492 Left warnings. Added "eximon gdb" invocation mode.
1494 PP/21 Defaulting "accept_8bitmime" to true, not false.
1496 PP/22 Added -bw for inetd wait mode support.
1498 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
1499 locate the relevant includes and libraries. Made this the default.
1501 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
1502 Bugzilla 1246, report and most of solution from Tomasz Kusy.
1504 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
1505 This may cause build issues on older platforms.
1507 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
1508 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
1509 gnutls_require_mac & gnutls_require_protocols (no longer supported).
1510 Added SNI support via GnuTLS too.
1511 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
1513 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
1515 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
1516 Applied second patch from Janne, implementing suggestion to default
1517 multiple-strings-in-record handling to match SPF spec.
1519 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
1521 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
1522 read-only, out of scope).
1523 Patch from Wolfgang Breyha, report from Stuart Northfield.
1525 PP/29 Fix three issues highlighted by clang analyser static analysis.
1526 Only crash-plausible issue would require the Cambridge-specific
1527 iplookup router and a misconfiguration.
1528 Report from Marcin Mirosław.
1530 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
1532 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
1533 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
1534 As part of this, removing so much warning spew let me fix some minor
1535 real issues in debug logging.
1537 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
1538 assignment on my part. Fixed.
1540 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
1541 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
1542 Janne Snabb (who went above and beyond: thank you).
1544 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
1545 string otherwise requires a connection and a bunch more work and it's
1546 relatively easy to get wrong. Should also expose TLS library linkage
1549 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
1550 64-bit ${eval} (JH/03).
1552 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
1553 GNU libc to support some of the 64-bit stuff, should not lead to
1554 conflicts. Defined before os.h is pulled in, so if a given platform
1555 needs to override this, it can.
1557 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
1558 protection layer was required, which is not implemented.
1559 Bugzilla 1254, patch from Wolfgang Breyha.
1561 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
1562 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
1563 tls_dhparam take prime identifiers. Also unbreak combination of
1564 OpenSSL+DH_params+TLSSNI.
1566 PP/39 Disable SSLv2 by default in OpenSSL support.
1572 PP/01 Solaris build fix for Oracle's LDAP libraries.
1573 Bugzilla 1109, patch from Stephen Usher.
1575 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
1577 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
1580 TF/02 Fix a couple more cases where we did not log the error message
1581 when unlink() failed. See also change 4.74-TF/03.
1583 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
1584 lock up or crash if it happened to be inside a call to libc when it
1585 got a SIGUSR1 from exiwhat.
1587 The SIGUSR1 handler appends the current process status to the process
1588 log which is later printed by exiwhat. It used to use the general
1589 purpose logging code to do this, but several functions it calls are
1590 not safe for signals.
1592 The new output code in the SIGUSR1 handler is specific to the process
1593 log, and simple enough that it's easy to inspect for signal safety.
1594 Removing some special cases also simplifies the general logging code.
1595 Removing the spurious timestamps from the process log simplifies
1598 TF/04 Improved ratelimit ACL condition.
1600 The /noupdate option has been deprecated in favour of /readonly which
1601 has clearer semantics. The /leaky, /strict, and /readonly update modes
1602 are mutually exclusive. The update mode is no longer included in the
1603 database key; it just determines when the database is updated. (This
1604 means that when you upgrade Exim will forget old rate measurements.)
1606 Exim now checks that the per_* options are used with an update mode that
1607 makes sense for the current ACL. For example, when Exim is processing a
1608 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
1609 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
1610 must specify per_mail/readonly. If you omit the update mode it defaults to
1611 /leaky where that makes sense (as before) or /readonly where required.
1613 The /noupdate option is now undocumented but still supported for
1614 backwards compatibility. It is equivalent to /readonly except that in
1615 ACLs where /readonly is required you may specify /leaky/noupdate or
1616 /strict/noupdate which are treated the same as /readonly.
1618 A useful new feature is the /count= option. This is a generalization
1619 of the per_byte option, so that you can measure the throughput of other
1620 aggregate values. For example, the per_byte option is now equivalent
1621 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
1623 The per_rcpt option has been generalized using the /count= mechanism
1624 (though it's more complicated than the per_byte equivalence). When it is
1625 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
1626 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
1627 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
1628 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
1629 Note that using per_rcpt with a non-readonly update mode in more than
1630 one ACL will cause the recipients to be double-counted. (The per_mail
1631 and per_byte options don't have this problem.)
1633 The handling of very low rates has changed slightly. If the computed rate
1634 is less than the event's count (usually one) then this event is the first
1635 after a long gap. In this case the rate is set to the same as this event's
1636 count, so that the first message of a spam run is counted properly.
1638 The major new feature is a mechanism for counting the rate of unique
1639 events. The new per_addr option counts the number of different
1640 recipients that someone has sent messages to in the last time period. It
1641 behaves like per_rcpt if all the recipient addresses are different, but
1642 duplicate recipient addresses do not increase the measured rate. Like
1643 the /count= option this is a general mechanism, so the per_addr option
1644 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
1645 example, measure the rate that a client uses different sender addresses
1646 with the options per_mail/unique=$sender_address. There are further
1647 details in the main documentation.
1649 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
1651 TF/06 Removed a few PCRE remnants.
1653 TF/07 Automatically extract Exim's version number from tags in the git
1654 repository when doing development or release builds.
1656 PP/02 Raise smtp_cmd_buffer_size to 16kB.
1657 Bugzilla 879. Patch from Paul Fisher.
1659 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
1660 Heavily based on revision 40f9a89a from Simon Arlott's tree.
1663 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
1665 PP/05 Variable $av_failed, true if the AV scanner deferred.
1666 Bugzilla 1078. Patch from John Horne.
1668 PP/06 Stop make process more reliably on build failure.
1669 Bugzilla 1087. Patch from Heiko Schlittermann.
1671 PP/07 Make maildir_use_size_file an _expandable_ boolean.
1672 Bugzilla 1089. Patch from Heiko Schlittermann.
1674 PP/08 Handle ${run} returning more data than OS pipe buffer size.
1675 Bugzilla 1131. Patch from Holger Weiß.
1677 PP/09 Handle IPv6 addresses with SPF.
1678 Bugzilla 860. Patch from Wolfgang Breyha.
1680 PP/10 GnuTLS: support TLS 1.2 & 1.1.
1682 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
1685 PP/11 match_* no longer expand right-hand-side by default.
1686 New compile-time build option, EXPAND_LISTMATCH_RHS.
1687 New expansion conditions, "inlist", "inlisti".
1689 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
1691 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
1693 PP/14 fix log_write() format string regression from TF/03.
1694 Bugzilla 1152. Patch from Dmitry Isaikin.
1700 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
1702 PP/02 Harmonised TLS library version reporting; only show if debugging.
1703 Layout now matches that introduced for other libraries in 4.74 PP/03.
1705 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
1707 PP/04 New "dns_use_edns0" global option.
1709 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
1712 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
1713 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
1715 TK/01 Updated PolarSSL code to 0.14.2.
1716 Bugzilla 1097. Patch from Andreas Metzler.
1718 PP/07 Catch divide-by-zero in ${eval:...}.
1719 Fixes bugzilla 1102.
1721 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
1724 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
1725 format-string attack -- SECURITY: remote arbitrary code execution.
1727 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
1728 time unintentionally subject to list matching rules, letting the header
1729 cause arbitrary Exim lookups (of items which can occur in lists, *not*
1730 arbitrary string expansion). This allowed for information disclosure.
1732 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
1733 INT_MIN/-1 -- value coerced to INT_MAX.
1739 NM/01 Workaround for PCRE version dependency in version reporting
1742 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
1743 This fixes portability to compilers other than gcc, notably
1744 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
1746 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
1747 makefiles for portability to HP-UX and POSIX correctness.
1749 PP/01 Permit LOOKUP_foo enabling on the make command-line.
1750 Also via indented variable definition in the Makefile.
1751 (Debugging by Oliver Heesakkers).
1753 PP/02 Restore caching of spamd results with expanded spamd_address.
1754 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
1756 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
1757 Improves build reliability. Fix from: Frank Elsner
1759 NM/02 Fix wide character breakage in the rfc2047 coding
1760 Fixes bug 1064. Patch from Andrey N. Oktyabrski
1762 NM/03 Allow underscore in dnslist lookups
1763 Fixes bug 1026. Patch from Graeme Fowler
1765 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
1766 Code patches from Adam Ciarcinski of NetBSD.
1768 NM/04 Fixed exiqgrep to cope with mailq missing size issue
1771 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
1772 is logged, to avoid truncation. Patch from John Horne.
1774 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
1775 Patch from Jakob Hirsch.
1777 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
1778 SQL string expansion failure details.
1779 Patch from Andrey Oktyabrski.
1781 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
1782 Patch from Simon Arlott.
1784 PP/09 New lookups functionality failed to compile on old gcc which rejects
1785 extern declarations in function scope.
1786 Patch from Oliver Fleischmann
1788 PP/10 Use sig_atomic_t for flags set from signal handlers.
1789 Check getgroups() return and improve debugging.
1790 Fixed developed for diagnosis in bug 927 (which turned out to be
1793 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
1794 Patch from Mark Zealey.
1796 PP/12 Bugzilla 1056: Improved spamd server selection.
1797 Patch from Mark Zealey.
1799 PP/13 Bugzilla 1086: Deal with maildir quota file races.
1800 Based on patch from Heiko Schlittermann.
1802 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
1803 Patch from Uwe Doering, sign-off by Michael Haardt.
1805 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
1806 variable declaration deep within a block. Bug and patch from
1809 PP/15 lookups-Makefile IRIX compatibility coercion.
1811 PP/16 Make DISABLE_DKIM build knob functional.
1813 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
1814 Patch by Simon Arlott
1816 TF/03 Fix valgrind.h portability to C89 compilers that do not support
1817 variable argument macros. Our copy now differs from upstream.
1823 TF/01 Failure to get a lock on a hints database can have serious
1824 consequences so log it to the panic log.
1826 TF/02 Log LMTP confirmation messages in the same way as SMTP,
1827 controlled using the smtp_confirmation log selector.
1829 TF/03 Include the error message when we fail to unlink a spool file.
1831 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
1832 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
1833 for maintaining out-of-tree patches for some time.
1835 PP/01 Bugzilla 139: Documentation and portability issues.
1836 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
1837 Handle per-OS dynamic-module compilation flags.
1839 PP/02 Let /dev/null have normal permissions.
1840 The 4.73 fixes were a little too stringent and complained about the
1841 permissions on /dev/null. Exempt it from some checks.
1842 Reported by Andreas M. Kirchwitz.
1844 PP/03 Report version information for many libraries, including
1845 Exim version information for dynamically loaded libraries. Created
1846 version.h, now support a version extension string for distributors
1847 who patch heavily. Dynamic module ABI change.
1849 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
1850 privilege escalation vulnerability whereby the Exim run-time user
1851 can cause root to append content of the attacker's choosing to
1854 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
1857 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
1858 If dropping privileges for untrusted macros, we disabled normal logging
1859 on the basis that it would fail; for the Exim run-time user, this is not
1860 the case, and it resulted in successful deliveries going unlogged.
1861 Fixed. Reported by Andreas Metzler.
1867 PP/01 Date: & Message-Id: revert to normally being appended to a message,
1868 only prepend for the Resent-* case. Fixes regression introduced in
1869 Exim 4.70 by NM/22 for Bugzilla 607.
1871 PP/02 Include check_rfc2047_length in configure.default because we're seeing
1872 increasing numbers of administrators be bitten by this.
1874 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
1876 PP/03 Bugzilla 994: added openssl_options main configuration option.
1878 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
1880 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
1882 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
1884 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
1885 without a peer certificate, leading to a segfault because of an
1886 assumption that peers always have certificates. Be a little more
1887 paranoid. Problem reported by Martin Tscholak.
1889 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
1890 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
1891 NB: ClamAV planning to remove STREAM in "middle of 2010".
1892 CL also introduces -bmalware, various -d+acl logging additions and
1893 more caution in buffer sizes.
1895 PP/09 Implemented reverse_ip expansion operator.
1897 PP/10 Bugzilla 937: provide a "debug" ACL control.
1899 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
1901 PP/12 Bugzilla 973: Implement --version.
1903 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
1905 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
1907 PP/15 Bugzilla 816: support multiple condition rules on Routers.
1909 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
1910 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
1911 ignore trailing whitespace.
1913 JJ/02 prevent non-panic DKIM error from being sent to paniclog
1915 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
1918 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
1919 Notification from Dr Andrew Aitchison.
1921 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
1922 ExtendedDetectionInfo response format.
1923 Notification from John Horne.
1925 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
1928 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
1929 XSL and documented dependency on system catalogs, with examples of how
1932 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
1935 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
1936 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
1937 configuration file which is writeable by the Exim user or group.
1939 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
1940 of configuration files to cover files specified with the -C option if
1941 they are going to be used with root privileges, not just the default
1944 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
1945 option (effectively making it always true).
1947 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
1948 files to be used while preserving root privileges.
1950 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
1951 that rogue child processes cannot use them.
1953 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
1954 run-time user, instead of root.
1956 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
1957 Exim run-time user without dropping privileges.
1959 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
1960 result string, instead of calling string_vformat() twice with the same
1963 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
1964 for other users. Others should always drop root privileges if they use
1965 -C on the command line, even for a whitelisted configure file.
1967 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
1969 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
1975 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
1976 $data_path, and $header_path variables; fixed documentation bugs and
1979 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
1980 exipick to access non-standard spools, including the "frozen" queue
1983 NM/01 Bugzilla 965: Support mysql stored procedures.
1984 Patch from Alain Williams
1986 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
1988 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
1989 Patch from Andreas Metzler
1991 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
1992 Patch from Kirill Miazine
1994 NM/05 Bugzilla 671: Added umask to procmail example.
1996 JJ/03 installed exipick 20100323.0, fixing doc bug
1998 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
1999 directory. Notification and patch from Dan Rosenberg.
2001 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
2003 TK/02 Improve log output when DKIM signing operation fails.
2005 MH/01 Treat the transport option dkim_domain as a colon separated
2006 list, not as a single string, and sign the message with each element,
2007 omitting multiple occurences of the same signer.
2009 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
2010 Bugzilla 985, 986. Patch by Simon Arlott
2012 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
2013 Patch by Simon Arlott
2015 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
2016 MBX locking. Notification from Dan Rosenberg.
2022 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
2024 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
2026 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
2028 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
2030 NM/04 Bugzilla 915: Flag broken perl installation during build.
2036 TK/01 Added patch by Johannes Berg that expands the main option
2037 "spamd_address" if it starts with a dollar sign.
2039 TK/02 Write list of recipients to X-Envelope-Sender header when building
2040 the mbox-format spool file for content scanning (suggested by Jakob
2043 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
2044 (http://www.dcc-servers.net/) support via dccifd. Activated by
2045 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
2047 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
2048 by Mark Daniel Reidel <mr@df.eu>.
2050 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
2051 When building exim an external PCRE library is now needed -
2052 PCRE is a system library on the majority of modern systems.
2053 See entry on PCRE_LIBS in EDITME file.
2055 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
2056 conversation. Added nologin parameter to request.
2057 Patch contributed by Kirill Miazine.
2059 TF/01 Do not log submission mode rewrites if they do not change the address.
2061 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
2063 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
2064 log files in place. Contributed by Roberto Lima.
2066 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
2068 TF/03 Bugzilla 615: When checking the local_parts router precondition
2069 after a local_part_suffix or local_part_prefix option, Exim now
2070 does not use the address's named list lookup cache, since this
2071 contains cached lookups for the whole local part.
2073 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
2074 Robert Millan. Documentation is in experimental-spec.txt.
2076 TF/04 Bugzilla 668: Fix parallel build (make -j).
2078 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
2080 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
2081 Patch provided by Jan Srzednicki.
2083 TF/05 Leading white space used to be stripped from $spam_report which
2084 wrecked the formatting. Now it is preserved.
2086 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
2087 that they are available at delivery time.
2089 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
2091 TF/08 TLS error reporting now respects the incoming_interface and
2092 incoming_port log selectors.
2094 TF/09 Produce a more useful error message if an SMTP transport's hosts
2095 setting expands to an empty string.
2097 NM/06 Bugzilla 744: EXPN did not work under TLS.
2098 Patch provided by Phil Pennock.
2100 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
2101 Patch provided by Richard Godbee.
2103 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
2104 acl_smtp_notquit, added index entry.
2106 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
2107 Patch provided by Eugene Bujak.
2109 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
2110 accept(). Patch provided by Maxim Dounin.
2112 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
2113 Patch provided by Phil Pennock.
2115 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
2117 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
2118 Patch provided by Brad "anomie" Jorsch.
2120 NM/14 Bugzilla 622: Added timeout setting to transport filter.
2121 Patch provided by Dean Brooks.
2123 TK/05 Add native DKIM support (does not depend on external libraries).
2125 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
2126 Patch provided by Graeme Fowler.
2128 NM/16 Bugzilla 851: Documentation example syntax fix.
2130 NM/17 Changed NOTICE file to remove references to embedded PCRE.
2132 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
2135 NM/19 Bugzilla 745: TLS version reporting.
2136 Patch provided by Phil Pennock.
2138 NM/20 Bugzilla 167: bool: condition support.
2139 Patch provided by Phil Pennock.
2141 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
2142 clients. Patch provided by Phil Pennock.
2144 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
2145 Patch provided by Brad "anomie" Jorsch.
2147 NM/23 Bugzilla 687: Fix misparses in eximstats.
2148 Patch provided by Heiko Schlittermann.
2150 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
2151 Patch provided by Heiko Schlittermann.
2153 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
2154 plus update to original patch.
2156 NM/26 Bugzilla 799: Documentation correction for ratelimit.
2158 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
2159 Patch provided by David Brownlee.
2161 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
2163 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
2165 NM/30 Bugzilla 888: TLS documentation bugfixes.
2167 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
2169 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
2170 Unlike the original bugzilla I have changed all shell scripts in src tree.
2172 NM/33 Bugzilla 898: Transport filter timeout fix.
2173 Patch by Todd Rinaldo.
2175 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
2176 Patch by Serge Demonchaux.
2178 NM/35 Bugzilla 39: Base64 decode bug fixes.
2179 Patch by Jakob Hirsch.
2181 NM/36 Bugzilla 909: Correct connect() call in dcc code.
2183 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
2185 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
2187 NM/39 Bugzilla 911: Fixed MakeLinks build script.
2193 TK/01 Add preliminary DKIM support. Currently requires a forked version of
2194 ALT-N's libdkim that I have put here:
2195 http://duncanthrax.net/exim-experimental/
2197 Note to Michael Haardt: I had to rename some vars in sieve.c. They
2198 were called 'true' and it seems that C99 defines that as a reserved
2199 keyword to be used with 'bool' variable types. That means you could
2200 not include C99-style headers which use bools without triggering
2201 build errors in sieve.c.
2203 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
2204 as mailq or other aliases. Changed the --help handling significantly
2205 to do whats expected. exim_usage() emits usage/help information.
2207 SC/01 Added the -bylocaldomain option to eximstats.
2209 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
2211 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
2213 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
2219 PH/01 Another patch from the Sieve maintainer.
2221 PH/02 When an IPv6 address is converted to a string for single-key lookup
2222 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
2223 dots are used instead of colons so that keys in lsearch files need not
2224 contain colons. This was done some time before quoting was made available
2225 in lsearch files. However, iplsearch files do require colons in IPv6 keys
2226 (notated using the quote facility) so as to distinguish them from IPv4
2227 keys. This meant that lookups for IP addresses in host lists did not work
2228 for iplsearch lookups.
2230 This has been fixed by arranging for IPv6 addresses to be expressed with
2231 colons if the lookup type is iplsearch. This is not incompatible, because
2232 previously such lookups could never work.
2234 The situation is now rather anomalous, since one *can* have colons in
2235 ordinary lsearch keys. However, making the change in all cases is
2236 incompatible and would probably break a number of configurations.
2238 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
2241 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
2242 conversion specification without a maximum field width, thereby enabling
2243 a rogue spamd server to cause a buffer overflow. While nobody in their
2244 right mind would setup Exim to query an untrusted spamd server, an
2245 attacker that gains access to a server running spamd could potentially
2246 exploit this vulnerability to run arbitrary code as the Exim user.
2248 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
2249 $primary_hostname instead of what libspf2 thinks the hosts name is.
2251 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
2252 a directory entry by the name of the lookup key. Previously, if a
2253 symlink pointed to a non-existing file or a file in a directory that
2254 Exim lacked permissions to read, a lookup for a key matching that
2255 symlink would fail. Now it is enough that a matching directory entry
2256 exists, symlink or not. (Bugzilla 503.)
2258 PH/03 The body_linecount and body_zerocount variables are now exported in the
2261 PH/04 Added the $dnslist_matched variable.
2263 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
2264 This means they are set thereafter only if the connection becomes
2267 PH/06 Added the client_condition to authenticators so that some can be skipped
2268 by clients under certain conditions.
2270 PH/07 The error message for a badly-placed control=no_multiline_responses left
2271 "_responses" off the end of the name.
2273 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
2275 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
2276 (without spaces) instead of just copying the configuration text.
2278 PH/10 Added the /noupdate option to the ratelimit ACL condition.
2280 PH/11 Added $max_received_linelength.
2282 PH/12 Added +ignore_defer and +include_defer to host lists.
2284 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
2285 way in which PCRE > 7.0 is built.
2287 PH/14 Implemented queue_only_load_latch.
2289 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
2290 MAIL command. The effect was to mangle the value on 64-bit systems.
2292 PH/16 Another patch from the Sieve maintainer.
2294 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
2296 PH/18 If a system quota error occurred while trying to create the file for
2297 a maildir delivery, the message "Mailbox is full" was not appended to the
2298 bounce if the delivery eventually timed out. Change 4.67/27 below applied
2299 only to a quota excession during the actual writing of the file.
2301 PH/19 It seems that peer DN values may contain newlines (and other non-printing
2302 characters?) which causes problems in log lines. The DN values are now
2303 passed through string_printing() before being added to log lines.
2305 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
2306 and InterBase are left for another time.)
2308 PH/21 Added message_body_newlines option.
2310 PH/22 Guard against possible overflow in moan_check_errorcopy().
2312 PH/23 POSIX allows open() to be a macro; guard against that.
2314 PH/24 If the recipient of an error message contained an @ in the local part
2315 (suitably quoted, of course), incorrect values were put in $domain and
2316 $local_part during the evaluation of errors_copy.
2322 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
2323 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
2326 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
2327 issue a MAIL command.
2329 PH/02 In an ACL statement such as
2331 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
2333 if a client was not listed at all, or was listed with a value other than
2334 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
2335 the condition was not true (as it should be), so access was not denied.
2336 The bug was that the ! inversion was incorrectly passed on to the second
2337 item. This has been fixed.
2339 PH/03 Added additional dnslists conditions == and =& which are different from
2340 = and & when the dns lookup returns more than one IP address.
2342 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
2343 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
2345 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
2346 FSYNC, which compiles an option called disable_fsync that allows for
2347 bypassing fsync(). The documentation is heavily laced with warnings.
2349 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
2351 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
2352 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
2353 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
2354 including adding "make clean"; (3) Added -fPIC when compiling the test
2355 dynamically loaded module, to get rid of a warning.
2357 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
2358 message fails, move_frozen_messages = true and ignore_bounce_errors_after
2359 = 0s. The bug is otherwise harmless.
2361 PH/07 There was a bug in the dovecot authenticator such that the value of
2362 $auth1 could be overwritten, and so not correctly preserved, after a
2363 successful authentication. This usually meant that the value preserved by
2364 the server_setid option was incorrect.
2366 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
2368 PH/09 Installed PCRE release 7.0.
2370 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
2371 run for batched SMTP input. It is now run at the start of every message
2372 in the batch. While fixing this I discovered that the process information
2373 (output by running exiwhat) was not always getting set for -bs and -bS
2374 input. This is fixed, and it now also says "batched" for BSMTP.
2376 PH/11 Added control=no_pipelining.
2378 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
2379 patch, slightly modified), and move the expansion of helo_data till after
2380 the connection is made in the smtp transport (so it can use these
2383 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
2385 PH/14 Added log_selector = +pid.
2387 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
2389 PH/16 Add ${if forany and ${if forall.
2391 PH/17 Added dsn_from option to vary the From: line in DSNs.
2393 PH/18 Flush SMTP output before performing a callout, unless control =
2394 no_callout_flush is set.
2396 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
2397 was true (the default) a successful delivery failed to delete the retry
2398 item, thus causing premature timeout of the address. The bug is now
2401 PH/20 Added hosts_avoid_pipelining to the smtp transport.
2403 PH/21 Long custom messages for fakedefer and fakereject are now split up
2404 into multiline responses in the same way that messages for "deny" and
2405 other ACL rejections are.
2407 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
2408 with slight modification.
2410 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
2411 draft, changing the syntax and factoring some duplicate code".
2413 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
2414 for deliveries of the second and subsequent messages over the same SMTP
2417 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
2418 ${reduce, with only minor "tidies".
2420 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
2422 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
2423 expansion side effects.
2425 PH/27 When a message times out after an over-quota error from an Exim-imposed
2426 quota, the bounce message says "mailbox is full". This message was not
2427 being given when it was a system quota that was exceeded. It now should
2430 MH/03 Made $recipients available in local_scan(). local_scan() already has
2431 better access to the recipient list through recipients_list[], but
2432 $recipients can be useful in postmaster-provided expansion strings.
2434 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
2435 in the case of a MAIL command with additional options following the
2436 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
2437 were accidentally chopped off.
2439 PH/29 SMTP synchronization checks are implemented when a command is read -
2440 there is a check that no more input is waiting when there shouldn't be
2441 any. However, for some commands, a delay in an ACL can mean that it is
2442 some time before the response is written. In this time, more input might
2443 arrive, invalidly. So now there are extra checks after an ACL has run for
2444 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
2445 pipelining has not been advertised.
2447 PH/30 MH's patch to allow iscntrl() characters to be list separators.
2449 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
2450 returned in the SMTP response when smtp_return_error_details was false.
2451 This has been fixed.
2453 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
2454 instead of the C I/O that was originally supplied, because problems were
2455 reported on Solaris.
2457 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
2458 Exim which did not show up earlier: it was assuming that a call to
2459 SSL_CTX_set_info_callback() might give an error value. In fact, there is
2460 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
2461 was a macro that became an assignment, so it seemed to work. This has
2462 changed to a proper function call with a void return, hence the compile
2463 error. Exim's code has been fixed.
2465 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
2468 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
2470 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
2472 PH/37 If a message is not accepted after it has had an id assigned (e.g.
2473 because it turns out to be too big or there is a timeout) there is no
2474 "Completed" line in the log. When some messages of this type were
2475 selected by exigrep, they were listed as "not completed". Others were
2476 picked up by some special patterns. I have improved the selection
2477 criteria to be more general.
2479 PH/38 The host_find_failed option in the manualroute router can now be set
2480 to "ignore", to completely ignore a host whose IP address cannot be
2481 found. If all hosts are ignored, the behaviour is controlled by the new
2482 host_all_ignored option.
2484 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
2485 homing or because of multiple MX records with /mx) generated more than
2486 one IP address, and the following item turned out to be the local host,
2487 all the secondary addresses of the first item were incorrectly removed
2488 from the list, along with the local host and any following hosts (which
2489 is what is supposed to happen).
2491 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
2492 whoever called Exim into the -H file. In the case of the daemon it was
2493 behaving confusingly. When first started, it used values for whoever
2494 started the daemon, but after a SIGHUP it used the Exim user (because it
2495 calls itself on a restart). I have changed the code so that it now always
2498 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
2499 message are rejected with the same error (e.g. no authentication or bad
2500 sender address), and a DATA command is nevertheless sent (as can happen
2501 with PIPELINING or a stupid MUA), the error message that was given to the
2502 RCPT commands is included in the rejection of the DATA command. This is
2503 intended to be helpful for MUAs that show only the final error to their
2506 PH/42 Another patch from the Sieve maintainer.
2508 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
2509 Eximstats - Fixed some broken HTML links and added missing column headers
2511 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
2512 columns for Rejects, Temp Rejects, Ham, and Spam rows.
2514 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
2516 PH/43 Yet another patch from the Sieve maintainer.
2518 PH/44 I found a way to check for a TCP/IP connection going away before sending
2519 the response to the final '.' that terminates a message, but only in the
2520 case where the client has not sent further data following the '.'
2521 (unfortunately, this is allowed). However, in many cases there won't be
2522 any further data because there won't be any more messages to send. A call
2523 to select() can be used: if it shows that the input is "ready", there is
2524 either input waiting, or the socket has been closed. An attempt to read
2525 the next input character can distinguish the two cases. Previously, Exim
2526 would have sent an OK response which the client would never have see.
2527 This could lead to message repetition. This fix should cure that, at
2528 least in a lot of common cases.
2530 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
2531 advertised in response to EHLO.
2537 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
2538 fixed by 4.65/MH/01 (is this a record?) are fixed:
2540 (i) An empty string was always treated as zero by the numeric comparison
2541 operators. This behaviour has been restored.
2543 (ii) It is documented that the numeric comparison operators always treat
2544 their arguments as decimal numbers. This was broken in that numbers
2545 starting with 0 were being interpreted as octal.
2547 While fixing these problems I realized that there was another issue that
2548 hadn't been noticed. Values of message_size_limit (both the global option
2549 and the transport option) were treated as octal if they started with 0.
2550 The documentation was vague. These values are now always treated as
2551 decimal, and I will make that clear in the documentation.
2557 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
2558 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
2561 MH/01 Don't check that the operands of numeric comparison operators are
2562 integers when their expansion is in "skipping" mode (fixes bug
2563 introduced by 4.64-PH/07).
2565 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
2566 child addresses, Exim now panics and dies. Previously, because the count
2567 is held in a short int, deliveries were likely to be lost. As such a
2568 large number of recipients for a single message is ridiculous
2569 (performance will be very, very poor), I have chosen to impose a limit
2570 rather than extend the field.
2576 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
2577 leftover -K file (the existence of which was triggered by #402).
2578 While we were at it, introduced process PID as part of the -K
2579 filename. This should rule out race conditions when creating
2582 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
2583 processing considerably. Previous code took too long for large mails,
2584 triggering a timeout which in turn triggers #401.
2586 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
2587 in the DK code in transports.c. sendfile() is not really portable,
2588 hence the _LINUX specificness.
2590 TF/01 In the add_headers option to the mail command in an Exim filter,
2591 there was a bug that Exim would claim a syntax error in any
2592 header after the first one which had an odd number of characters
2595 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
2596 callout verification, Exim cached a "reject" for the entire domain. This
2597 is correct for most verifications, but it is not correct for a recipient
2598 verification with use_sender or use_postmaster set, because in that case
2599 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
2600 case of MAIL FROM:<> rejection from other early rejections (e.g.
2601 rejection of HELO). When verifying a recipient using a non-null MAIL
2602 address, the cache is ignored if it shows MAIL FROM:<> rejection.
2603 Whatever the result of the callout, the value of the domain cache is
2604 left unchanged (for any other kind of callout, getting as far as trying
2605 RCPT means that the domain itself is ok).
2607 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
2610 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
2611 manifest itself as EPIPE rather than ECONNECT. When tidying away a
2612 session, the daemon ignores ECONNECT errors and logs others; it now
2613 ignores EPIPE as well.
2615 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
2616 (quoted-printable decoding).
2618 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
2619 later the small subsequent patch to fix an introduced bug.
2621 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
2623 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
2625 PH/08 An error is now given if message_size_limit is specified negative.
2627 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
2628 to be given (somewhat) arbitrary names.
2630 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
2633 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
2634 miscellaneous code fixes
2636 PH/10 Added the log_reject_target ACL modifier to specify where to log
2639 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
2640 hostname. This is wrong, because it relates to the incoming message (and
2641 probably the interface on which it is arriving) and not to the outgoing
2642 callout (which could be using a different interface). This has been
2643 changed to use the value of the helo_data option from the smtp transport
2644 instead - this is what is used when a message is actually being sent. If
2645 there is no remote transport (possible with a router that sets up host
2646 addresses), $smtp_active_hostname is used.
2648 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
2649 tweaks were necessary in order to get it to work (see also 21 below):
2650 (a) The code assumed that strncpy() returns a negative number on buffer
2651 overflow, which isn't the case. Replaced with Exim's string_format()
2653 (b) There were several signed/unsigned issues. I just did the minimum
2654 hacking in of casts. There is scope for a larger refactoring.
2655 (c) The code used strcasecmp() which is not a standard C function.
2656 Replaced with Exim's strcmpic() function.
2657 (d) The code set only $1; it now sets $auth1 as well.
2658 (e) A simple test gave the error "authentication client didn't specify
2659 service in request". It would seem that Dovecot has changed its
2660 interface. Fortunately there's a specification; I followed it and
2661 changed what the client sends and it appears to be working now.
2663 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
2666 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
2667 address A is aliased to B and C, where B exists and C does not. Without
2668 -v the output is "A verified" because verification stops after a
2669 successful redirection if more than one address is generated. However,
2670 with -v the child addresses are also verified. Exim was outputting "A
2671 failed to verify" and then showing the successful verification for C,
2672 with its parentage. It now outputs "B failed to verify", showing B's
2673 parentage before showing the successful verification of C.
2675 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
2676 look up a TXT record in a specific list after matching in a combined
2679 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
2680 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
2681 they consult the DNS. I had assumed they would set it the way they
2682 wanted; and indeed my experiments on Linux seem to show that in some
2683 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
2684 To be on the safe side, however, I have now made the interface to
2685 host_find_byname() similar to host_find_bydns(), with an argument
2686 containing the DNS resolver options. The host_find_byname() function now
2687 sets these options at its start, just as host_find_bydns() does. The smtp
2688 transport options dns_qualify_single and dns_search_parents are passed to
2689 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
2690 of host_find_byname() use the default settings of RES_DEFNAMES
2691 (qualify_single) but not RES_DNSRCH (search_parents).
2693 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
2694 spool_read_header() do less string testing, by means of a preliminary
2695 switch on the second character of optional "-foo" lines. (This is
2696 overdue, caused by the large number of possibilities that now exist.
2697 Originally there were few.) While I was there, I also converted the
2698 str(n)cmp tests so they don't re-test the leading "-" and the first
2699 character, in the hope this might squeeze out yet more improvement.
2701 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
2702 flag allowing group syntax was set by the header_syntax check but not
2703 turned off, possible causing trouble later; (2) The flag was not being
2704 set at all for the header_verify test, causing "group"-style headers to
2705 be rejected. I have now set it in this case, and also caused header_
2706 verify to ignore an empty address taken from a group. While doing this, I
2707 came across some other cases where the code for allowing group syntax
2708 while scanning a header line wasn't quite right (mostly, not resetting
2709 the flag correctly in the right place). These bugs could have caused
2710 trouble for malformed header lines. I hope it is now all correct.
2712 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
2713 with the "reply" argument non-NULL. The code, however (which originally
2714 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
2715 but it didn't always do it. This confused somebody who was copying the
2716 code for some other use. I have removed all the tests.
2718 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
2719 feature that was used to support insecure browsers during the U.S. crypto
2720 embargo. It requires special client support, and Exim is probably the
2721 only MTA that supported it -- and would never use it because real RSA is
2722 always available. This code has been removed, because it had the bad
2723 effect of slowing Exim down by computing (never used) parameters for the
2724 RSA_EXPORT functionality.
2726 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
2727 authenticator to fail if there's a tab character in the incoming data
2728 (there should never be unless someone is messing about, as it's supposed
2729 to be base64-encoded). Also added, on Timo's advice, the "secured" option
2730 if the connection is using TLS or if the remote IP is the same as the
2731 local IP, and the "valid-client-cert option" if a client certificate has
2734 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
2735 authenticators. This can be used for authorization after authentication
2736 succeeds. (In the case of plaintext, it servers for both authentication
2739 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
2740 if any retry times were supplied.
2742 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
2743 connection, where there is no client IP address to check. In this
2744 situation, the verify now always succeeds.
2746 PH/25 Applied John Jetmore's -Mset patch.
2748 PH/26 Added -bem to be like -Mset, but loading a message from a file.
2750 PH/27 In a string expansion for a processed (not raw) header when multiple
2751 headers of the same name were present, leading whitespace was being
2752 removed from all of them, but trailing whitespace was being removed only
2753 from the last one. Now trailing whitespace is removed from each header
2754 before concatenation. Completely empty headers in a concatenation (as
2755 before) are ignored.
2757 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
2758 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
2760 PH/29 [Removed. This was a change that I later backed out, and forgot to
2761 correct the ChangeLog entry (that I had efficiently created) before
2762 committing the later change.]
2764 PH/30 Exim was sometimes attempting to deliver messages that had suffered
2765 address errors (4xx response to RCPT) over the same connection as other
2766 messages routed to the same hosts. Such deliveries are always "forced",
2767 so retry times are not inspected. This resulted in far too many retries
2768 for the affected addresses. The effect occurred only when there were more
2769 hosts than the hosts_max_try setting in the smtp transport when it had
2770 the 4xx errors. Those hosts that it had tried were not added to the list
2771 of hosts for which the message was waiting, so if all were tried, there
2772 was no problem. Two fixes have been applied:
2774 (i) If there are any address or message errors in an SMTP delivery, none
2775 of the hosts (tried or untried) are now added to the list of hosts
2776 for which the message is waiting, so the message should not be a
2777 candidate for sending over the same connection that was used for a
2778 successful delivery of some other message. This seems entirely
2779 reasonable: after all the message is NOT "waiting for some host".
2780 This is so "obvious" that I'm not sure why it wasn't done
2781 previously. Hope I haven't missed anything, but it can't do any
2782 harm, as the worst effect is to miss an optimization.
2784 (ii) If, despite (i), such a delivery is accidentally attempted, the
2785 routing retry time is respected, so at least it doesn't keep
2786 hammering the server.
2788 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
2789 in ${readsocket because some servers need this prod.
2791 PH/32 Added some extra debug output when updating a wait-xxx database.
2793 PH/33 The hint "could be header name not terminated by colon", which has been
2794 given for certain expansion errors for a long time, was not being given
2795 for the ${if def:h_colon_omitted{... case.
2797 PH/34 The spec says: "With one important exception, whenever a domain list is
2798 being scanned, $domain contains the subject domain." There was at least
2799 one case where this was not true.
2801 PH/35 The error "getsockname() failed: connection reset by peer" was being
2802 written to the panic log as well as the main log, but it isn't really
2803 panic-worthy as it just means the connection died rather early on. I have
2804 removed the panic log writing for the ECONNRESET error when getsockname()
2807 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
2808 runs only) independently of the message's sender address. This meant
2809 that, if the 4xx error was in fact related to the sender, a different
2810 message to the same recipient with a different sender could confuse
2811 things. In particular, this can happen when sending to a greylisting
2812 server, but other circumstances could also provoke similar problems.
2813 I have changed the default so that the retry time for these errors is now
2814 based a combination of the sender and recipient addresses. This change
2815 can be overridden by setting address_retry_include_sender=false in the
2818 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
2819 remote server are returned as part of bounce messages. This was not
2820 happening for LMTP over a pipe (the lmtp transport), but now it is the
2821 same for both kinds of LMTP.
2823 PH/38 Despite being documented as not happening, Exim was rewriting addresses
2824 in header lines that were in fact CNAMEs. This is no longer the case.
2826 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
2827 and queue runs started by the daemon processed all messages. This has
2828 been fixed so that -R and -S can now usefully be given with -q<time>.
2830 PH/40 Import PCRE release 6.7 (fixes some bugs).
2832 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
2834 PH/42 Give an error if -q is specified more than once.
2836 PH/43 Renamed the variables $interface_address and $interface_port as
2837 $received_ip_address and $received_port, to make it clear that these
2838 values apply to message reception, and not to the outgoing interface when
2839 a message is delivered. (The old names remain recognized, of course.)
2841 PH/44 There was no timeout on the connect() call when using a Unix domain
2842 socket in the ${readsocket expansion. There now is.
2844 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
2845 be meaningful with "accept".
2847 SC/01 Eximstats V1.43
2848 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
2850 SC/02 Eximstats V1.44
2851 Use a glob alias rather than an array ref in the generated
2852 parser. This improves both readability and performance.
2854 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
2855 Collect SpamAssassin and rejection statistics.
2856 Don't display local sender or destination tables unless
2857 there is data to show.
2858 Added average volumes into the top table text output.
2860 SC/04 Eximstats V1.46
2861 Collect data on the number of addresses (recipients)
2862 as well as the number of messages.
2864 SC/05 Eximstats V1.47
2865 Added 'Message too big' to the list of mail rejection
2866 reasons (thanks to Marco Gaiarin).
2868 SC/06 Eximstats V1.48
2869 Mainlog lines which have GMT offsets and are too short to
2870 have a flag are now skipped.
2872 SC/07 Eximstats V1.49 (Alain Williams)
2873 Added the -emptyok flag.
2875 SC/08 Eximstats V1.50
2876 Fixes for obtaining the IP address from reject messages.
2878 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
2879 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
2880 whitespace changes from 4.64-PH/27
2882 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
2885 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
2886 are found, allow negative numbers in numeric criteria)
2888 JJ/06 exipick.20061117.2, added new $message_body_missing variable
2890 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
2891 to match changes made in 4.64-PH/43
2893 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
2895 PH/47 Put in an explicit test for a DNS lookup of an address record where the
2896 "domain" is actually an IP address, and force a failure. This locks out
2897 those revolvers/nameservers that support "A-for-A" lookups, in
2898 contravention of the specifications.
2900 PH/48 When a host name was looked up from an IP address, and the subsequent
2901 forward lookup of the name timed out, the host name was left in
2902 $sender_host_name, contrary to the specification.
2904 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
2905 restricted to single-key lookups, Exim was not diagnosing an error if
2906 * or *@ was used with a query-style lookup.
2908 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
2910 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
2911 long ago, but noone interested enough thought of it. Let's just say that
2912 the "1.1" means that there are some new functions that weren't there at
2913 some point in the past.
2915 PH/51 Error processing for expansion failure of helo_data from an smtp
2916 transport during callout processing was broken.
2918 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
2919 tested/used via the -bh/-bhc/-bs options.
2921 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
2922 bug, fixed in subsequent PCRE releases).
2924 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
2925 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
2927 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
2933 SC/01 Use a glob alias rather than an array ref in eximstats generated
2934 parser. This improves both readability and performance.
2936 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
2937 Don't display local sender or destination tables in eximstats unless
2938 there is data to show.
2939 Added average volumes into the eximstats top table text output.
2941 SC/03 Collect data on the number of addresses (recipients) as well
2942 as the number of messages in eximstats.
2944 TF/01 Correct an error in the documentation for the redirect router. Exim
2945 does (usually) call initgroups() when daemonizing.
2947 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
2948 with consistent privilege compared to when running as a daemon.
2950 TF/03 Note in the spec that $authenticated_id is not set for local
2951 submissions from trusted users.
2953 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
2954 Thanks to Dean Brooks <dean@iglou.com> for the patch.
2956 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
2957 by adding some example configuration directives to the default
2958 configuration file. A little bit of work is required to uncomment the
2959 directives and define how usernames and passwords are checked, but
2960 there is now a framework to start from.
2962 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
2963 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
2964 without this. I don't know how relevant this is to other LDAP libraries.
2966 PH/02 Add the verb name to the "unknown ACL verb" error.
2968 PH/03 Magnus Holmgren's patch for filter_prepend_home.
2970 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
2972 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
2973 directory not expanded when it should be if an expanded home directory
2974 was set for the address (which is overridden by the transport).
2976 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
2979 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
2980 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
2981 because it is too late at that time, and has no effect.
2983 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
2984 security issue with \' (bugzilla #107). I could not use the
2985 PQescapeStringConn() function, because it needs a PGconn value as one of
2988 PH/08 When testing addresses using -bt, indicate those final addresses that
2989 are duplicates that would not cause an additional delivery. At least one
2990 person was confused, thinking that -bt output corresponded to deliveries.
2991 (Suppressing duplicates isn't a good idea as you lose the information
2992 about possibly different redirections that led to the duplicates.)
2994 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
2995 systems where poll() doesn't work, in particular OS X.
2997 PH/10 Added more information to debugging output for retry time not reached.
2999 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
3000 operations in malware.c.
3002 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
3005 PH/13 If write_rejectlog was set false when logging was sent to syslog with
3006 syslog_duplication set false, log lines that would normally be written
3007 both the the main log and to the reject log were not written to syslog at
3010 PH/14 In the default configuration, change the use of "message" in ACL warn
3011 statements to "add_header".
3013 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
3014 not followed by a command (e.g. "seen endif").
3016 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
3017 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
3020 PH/17 Added extra conditions to the default value of delay_warning_condition
3024 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
3025 { match{$h_precedence:}{(?i)bulk|list|junk} } \
3026 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
3029 The Auto-Submitted: and various List- headers are standardised, whereas I
3030 don't think Precedence: ever was.
3032 PH/18 Refactored debugging code in route_finduser() to show more information,
3033 in particular, the error code if getpwnam() issues one.
3035 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
3036 This is apparently needed in addition to the PH/07 change above to avoid
3037 any possible encoding problems.
3039 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
3040 but not after initializing Perl.
3042 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
3043 output them only if debugging. By default they are written stderr,
3044 apparently, which is not desirable.
3046 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
3049 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
3052 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
3054 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
3055 authenticated or an ident call has been made. Suppress the default
3056 values for $authenticated_id and $authenticated_sender (but permit -oMai
3057 and -oMas) when testing with -bh.
3059 PH/24 Re-jigged the order of the tests in the default configuration so that the
3060 tests for valid domains and recipients precede the DNS black list and CSA
3061 tests, on the grounds that those ones are more expensive.
3063 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
3064 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
3065 This bug exists in every version of Exim that I still have, right back to
3068 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
3069 However, an attempt to turn on case-sensitivity in a regex key by
3070 including (?-i) didn't work because the subject string was already
3071 lowercased, and the effects were non-intuitive. It turns out that a
3072 one-line patch can be used to allow (?-i) to work as expected.
3078 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
3079 other effects) broke the use of negated acl sub-conditions.
3081 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
3084 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
3085 "Deny" causes Exim to reject the incoming connection with a 554 error.
3086 Unfortunately, if there is a major crisis, such as a disk failure,
3087 tcp-wrappers gives "deny", whereas what one would like would be some
3088 kind of temporary error. A kludge has been added to help with this.
3089 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
3090 554 error is used if errno is still zero or contains ENOENT (which occurs
3091 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
3094 PH/03 Add -lutil to the default FreeBSD LIBS setting.
3096 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
3097 errors. Otherwise a message that provokes a temporary error (when other
3098 messages do not) can cause a whole host to time out.
3100 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
3101 addresses were routed directly to files or pipes from a redirect router.
3102 File deliveries just didn't batch; pipe deliveries might have suffered
3105 PH/06 A failure to get a lock for a hints database would erroneously always say
3106 "Failed to get write lock", even when it was really a read lock.
3108 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
3109 of 0600. This has been changed to use the value of the lockfile_mode
3110 option (which defaults to 0600).
3112 PH/08 Applied small patch from the Sieve maintainer.
3114 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
3115 folder from quota calculations, a direct delivery into this folder messed
3116 up the contents of the maildirsize file. This was because the regex was
3117 used only to exclude .Trash (or whatever) when the size of the mailbox
3118 was calculated. There was no check that a delivery was happening into an
3119 excluded directory. This bug has been fixed by ignoring all quota
3120 processing for deliveries into excluded directories.
3122 PH/10 Added the maildirfolder_create_regex option to appendfile.
3128 PH/01 The code for finding all the local interface addresses on a FreeBSD
3129 system running IPv6 was broken. This may well have applied to all BSD
3130 systems, as well as to others that have similar system calls. The broken
3131 code found IPv4 interfaces correctly, but gave incorrect values for the
3132 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
3133 that it would not match correctly against @[] and not recognize the IPv6
3136 PH/02 The ipliteral router was not recognizing addresses of the form user@
3137 [ipv6:....] because it didn't know about the "ipv6:" prefix.
3139 PH/03 Added disable_ipv6.
3141 PH/04 Changed $reply_address to use the raw form of the headers instead of the
3142 decoded form, because it is most often used to construct To: headers
3143 lines in autoreplies, and the decoded form may well be syntactically
3144 invalid. However, $reply_address has leading white space removed, and all
3145 newlines turned into spaces so that the autoreply transport does not
3148 PH/05 If group was specified without a user on a router, and no group or user
3149 was specified on a transport, the group from the router was ignored.
3151 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
3152 for visible compile-time settings that can be used to change these
3153 numbers, for those that want even more. Backwards compatibility with old
3154 spool files has been maintained. However, going back to a previous Exim
3155 release will lost any variables that are in spool files.
3157 PH/07 Two small changes when running in the test harness: increase delay when
3158 passing a TCP/IP connection to a new process, in case the original
3159 process has to generate a bounce, and remove special handling of
3160 127.0.0.2 (sic), which is no longer necessary.
3162 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
3163 be the same on different OS.
3165 PH/09 Moved a debug statement in filter processing to avoid a race problem when
3168 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
3169 whether --show-vars was specified or not
3171 JJ/02 exipick: Added support for new ACL variable spool format introduced
3174 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
3175 syntactically invalid From: or Reply-to: line, and a filter used this to
3176 generate an autoreply, and therefore failed to obtain an address for the
3177 autoreply, Exim could try to deliver to a non-existent relative file
3178 name, causing unrelated and misleading errors. What now happens is that
3179 it logs this as a hard delivery error, but does not attempt to create a
3182 PH/11 The exinext utility has a -C option for testing purposes, but although
3183 the given file was scanned by exinext itself; it wasn't being passed on
3184 when Exim was called.
3186 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
3187 an end-of-file indication when reading a command response.
3189 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
3190 compiled. In many other places in Exim, IPv6 addresses are always
3191 recognized, so I have changed this. It also means that IPv4 domain
3192 literals of the form [IPV4:n.n.n.n] are now always recognized.
3194 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
3195 used if the router is not running as root, for example, when verifying at
3196 ACL time, or when using -bh. The debugging output from this situation was
3197 non-existent - all you got was a failure to exec. I have made two
3200 (a) Failures to set uid/gid, the current directory, or a process leader
3201 in a subprocess such as that created by queryprogram now generate
3202 suitable debugging output when -d is set.
3204 (b) The queryprogram router detects when it is not running as root,
3205 outputs suitable debugging information if -d is set, and then runs
3206 the subprocess without attempting to change uid/gid.
3208 PH/15 Minor change to Makefile for building test_host (undocumented testing
3211 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
3212 additional section of a DNS packet that returns MX or SRV records.
3213 Instead, it always explicitly searches for A/AAAA records. This avoids
3214 major problems that occur when a DNS server includes only records of one
3215 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
3216 fixed another bug: if SRV records were looked up and the corresponding
3217 address records were *not* found in the additional section, the port
3218 values from the SRV records were lost.
3220 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
3221 using the correct key (the original address) when searching the retry
3222 rules in order to find which one to use for generating the retry hint.
3224 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
3225 adding the default one. Similarly, if it contains a Reply-To: header, the
3226 errors_reply_to option, if set, is not used.
3228 PH/19 When calculating a retry time, Exim used to measure the "time since
3229 failure" by looking at the "first failed" field in the retry record. Now
3230 it does not use this if it is later than than the arrival time of the
3231 message. Instead it uses the arrival time. This makes for better
3232 behaviour in cases where some deliveries succeed, thus re-setting the
3233 "first failed" field. An example is a quota failure for a huge message
3234 when small messages continue to be delivered. Without this change, the
3235 "time since failure" will always be short, possible causing more frequent
3236 delivery attempts for the huge message than are intended.
3237 [Note: This change was subsequently modified - see PH/04 for 4.62.]
3239 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
3240 $1, $2, $3) because the numerical variables can be reset during some
3241 expansion items (e.g. "match"), thereby losing the authentication data.
3243 PH/21 Make -bV show the size of off_t variables so that the test suite can
3244 decide whether to run tests for quotas > 2G.
3246 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
3247 mailbox_size, and mailbox_filecount in the appendfile transport. If a
3248 filecount value is greater than 2G or if a quota value is greater than 2G
3249 on a system where the size of off_t is not greater than 4, a panic error
3252 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
3253 never match. The debug and -bh output now contains an explicit error
3254 message indicating a malformed IPv4 address or mask.
3256 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
3257 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
3258 PH/23 above applies.
3260 PH/25 Do not write to syslog when running in the test harness. The only
3261 occasion when this arises is a failure to open the main or panic logs
3262 (for which there is an explicit test).
3264 PH/26 Added the /no_tell option to "control=freeze".
3266 PH/27 If a host name lookup failed very early in a connection, for example, if
3267 the IP address matched host_lookup and the reverse lookup yielded a name
3268 that did not have a forward lookup, an error message of the form "no IP
3269 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
3270 could be logged. Now it outputs the IP address instead of "NULL".
3272 PH/28 An enabling patch from MH: add new function child_open_exim2() which
3273 allows the sender and the authenticated sender to be set when
3274 submitting a message from within Exim. Since child_open_exim() is
3275 documented for local_scan(), the new function should be too.
3277 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
3278 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
3279 results in an empty string is now treated as unset.
3281 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
3283 PH/31 Added sender_verify_fail logging option.
3285 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
3286 needed by RFC 822 but not by RFC 2822 was commented out. I have now
3287 tidied the source and removed it altogether.
3289 PH/33 When a queue run was abandoned because the load average was too high, a
3290 log line was always written; now it is written only if the queue_run log
3291 selector is set. In addition, the log line for abandonment now contains
3292 information about the queue run such as the pid. This is always present
3293 in "start" and "stop" lines but was omitted from the "abandon" line.
3295 PH/34 Omit spaces between a header name and the colon in the error message that
3296 is given when verify = headers_syntax fails (if there are lots of them,
3297 the message gets confusing).
3299 PH/35 Change the default for dns_check_names_pattern to allow slashes within
3300 names, as there are now some PTR records that contain slashes. This check
3301 is only to protect against broken name servers that fall over on strange
3302 characters, so the fact that it applies to all lookups doesn't matter.
3304 PH/36 Now that the new test suite is complete, we can remove some of the
3305 special code in Exim that was needed for the old test suite. For example,
3306 sorting DNS records because real resolvers return them in an arbitrary
3307 order. The new test suite's fake resolver always returns records in the
3310 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
3311 bounces) except when queue_only is set, to avoid logging races between
3312 the different processes.
3314 PH/38 Panic-die if .include specifies a non-absolute path.
3316 PH/39 A tweak to the "H" retry rule from its user.
3318 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
3319 a label. They prevented compilation on older perls.
3321 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
3322 a warning to be raised on newish perls.
3324 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
3325 on queue. Changes to match documented behaviour of showing count of
3326 messages matching specified criteria.
3328 PH/40 Changed the default ident timeout from 30s to 5s.
3330 PH/41 Added support for the use of login_cap features, on those BSD systems
3331 that have them, for controlling the resources used by pipe deliveries.
3333 PH/42 The content-scanning code uses fopen() to create files in which to put
3334 message data. Previously it was not paying any attention to the mode of
3335 the files. Exim runs with umask(0) because the rest of the code creates
3336 files with open(), and sets the required mode explicitly. Thus, these
3337 files were ending up world-writeable. This was not a big issue, because,
3338 being within the spool directory, they were not world-accessible. I have
3339 created a function called modefopen, which takes an additional mode
3340 argument. It sets umask(777), creates the file, chmods it to the required
3341 mode, then resets the umask. All the relevant calls to fopen() in the
3342 content scanning code have been changed to use this function.
3344 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
3345 to 24 hours. This avoids potential overflow problems when processing G
3346 and H retry rules. I suspect nobody ever tinkers with this value.
3348 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
3350 PH/45 When the plaintext authenticator is running as a client, the server's
3351 challenges are checked to ensure they are valid base64 strings. By
3352 default, the authentication attempt is cancelled if an invalid string is
3353 received. Setting client_ignore_invalid_base64 true ignores these errors.
3354 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
3355 they are received. Thus, the responses can be made to depend on the
3356 challenges. If an invalid string is ignored, an empty string is placed in
3359 PH/46 Messages that are created by the autoreply transport now contains a
3360 References: header, in accordance with RFCs 2822 and 3834.
3362 PH/47 Added authenticated_sender_force to the smtp transport.
3364 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
3366 PH/49 Installed latest patch from the Sieve maintainer.
3368 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
3369 was also set, the appendfile transport was unnecessarily scanning a
3370 directory of message files (e.g. for maildir delivery) to find the count
3371 of files (along with the size), even though it did not need this
3372 information. It now does the scan only if it needs to find either the
3373 size of the count of files.
3375 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
3377 PH/52 Two bugs concerned with error handling when the smtp transport is
3380 (i) Exim was not creating retry information for temporary errors given
3381 for individual recipients after the DATA command when the smtp transport
3382 was used in LMTP mode. This meant that they could be retried too
3383 frequently, and not timed out correctly.
3385 (ii) Exim was setting the flag that allows error details to be returned
3386 for LMTP errors on RCPT commands, but not for LMTP errors for individual
3387 recipients that were returned after the DATA command.
3389 PH/53 This is related to PH/52, but is more general: for any failing address,
3390 when detailed error information was permitted to be returned to the
3391 sender, but the error was temporary, then after the final timeout, only
3392 "retry timeout exceeded" was returned. Now it returns the full error as
3393 well as "retry timeout exceeded".
3395 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
3396 do this, and (what is worse) MTAs that accept it.
3398 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
3399 will now be deprecated.
3401 PH/56 New os.c-cygwin from the Cygwin maintainer.
3403 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
3404 formats (previously only available in exim formats via -bpr, -bpru,
3405 and -bpra. Now also available in native and exiqgrep formats)
3407 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
3408 with very large, slow to parse queues
3410 JJ/08 exipick: added ! as generic prefix to negate any criteria format
3412 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
3414 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
3415 responses to authentication challenges, though it was showing the
3416 challenges; (ii) I've removed the CR characters from the debug output for
3419 PH/58 Allow for the insertion of a newline as well as a space when a string
3420 is turned into more than one encoded-word during RFC 2047 encoding. The
3421 Sieve code now uses this.
3423 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
3424 data_4xx, lost_connection, tls_required.
3426 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
3427 message was being sent as an SMTP response.
3429 PH/61 Add -l and -k options to exicyclog.
3431 PH/62 When verifying, if an address was redirected to one new address, so that
3432 verification continued, and the new address failed or deferred after
3433 having set something in $address_data, the value of $address_data was not
3434 passed back to the ACL. This was different to the case when no
3435 redirection occurred. The value is now passed back in both cases.
3437 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
3438 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
3439 use login_cap.h, so on its own it isn't the distinguishing feature. The
3440 new name refers directly to the setclassresources() function.
3442 PH/65 Added configuration files for NetBSD3.
3444 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
3446 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
3447 is preferred over IPv4.
3449 PH/68 The bounce_return_message and bounce_return_body options were not being
3450 honoured for bounces generated during the reception of non-SMTP messages.
3451 In particular, this applied to messages rejected by the ACL. This bug has
3452 been fixed. However, if bounce_return_message is true and bounce_return_
3453 body is false, the headers that are returned for a non-SMTP message
3454 include only those that have been read before the error was detected.
3455 (In the case of an ACL rejection, they have all been read.)
3457 PH/69 The HTML version of the specification is now built in a directory called
3458 spec_html instead of spec.html, because the latter looks like a path with
3459 a MIME-type, and this confuses some software.
3461 PH/70 Catch two compiler warnings in sieve.c.
3463 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
3464 function verify_get_ident() calls ip_connect() to connect a socket, but
3465 if the "connect()" function timed out, ip_connect() used to close the
3466 socket. However, verify_get_ident() also closes the socket later, and in
3467 between Exim writes to the log, which may get opened at this point. When
3468 the socket was closed in ip_connect(), the log could get the same file
3469 descriptor number as the socket. This naturally causes chaos. The fix is
3470 not to close the socket in ip_connect(); the socket should be closed by
3471 the function that creates it. There was only one place in the code where
3472 this was missing, in the iplookup router, which I don't think anybody now
3473 uses, but I've fixed it anyway.
3475 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
3476 well as to direct DNS lookups. Otherwise the handling of names in host
3477 lists is inconsistent and therefore confusing.
3483 PH/01 Two changes to the default runtime configuration:
3485 (1) Move the checks for relay_from_hosts and authenticated clients from
3486 after to before the (commented out) DNS black list checks.
3488 (2) Add control=submission to the relay_from_hosts and authenticated
3489 clients checks, on the grounds that messages accepted by these
3490 statements are most likely to be submissions.
3492 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
3494 (1) Generate an error if the third argument for the ${prvs expansion is
3497 (2) Treat a missing third argument of ${prvscheck as if it were an empty
3500 (3) Reset the variables that are obtained from the first argument of
3501 ${prvscheck and used in the second argument before leaving the code,
3502 because their memory is reclaimed, so using them afterwards may do
3505 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
3506 one (it's much easier than Tom thought :-).
3508 (5) Because of (4), we can now allow for the use of $prvscheck_result
3509 inside the third argument.
3511 PH/03 For some reason, the default setting of PATH when running a command from
3512 a pipe transport was just "/usr/bin". I have changed it to
3515 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
3516 anything to be listed in the output from -bV.
3518 PH/05 When a filter generated an autoreply, the entire To: header line was
3519 quoted in the delivery log line, like this:
3521 => >A.N.Other <ano@some.domain> <original@ddress> ...
3523 This has been changed so that it extracts the operative address. There
3524 may be more than one such address. If so, they are comma-separated, like
3527 => >ano@some.domain,ona@other.domain <original@ddress> ...
3529 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
3530 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
3531 not being looked up in the rDNS to get a host name, Exim was showing the
3532 IP address twice in Received: lines, even though the IP addresses were
3533 identical. For example:
3535 Received: from [1.2.3.4] (helo=[1.2.3.4])
3537 However, if the real host name was known, it was omitting the HELO data
3538 if it matched the actual IP address. This has been tidied up so that it
3539 doesn't show the same IP address twice.
3541 PH/07 When both +timestamp and +memory debugging was on, the value given by
3542 $tod_xxx expansions could be wrong, because the tod_stamp() function was
3543 called by the debug printing, thereby overwriting the timestamp buffer.
3544 Debugging no longer uses the tod_stamp() function when +timestamp is set.
3546 PH/08 When the original message was included in an autoreply transport, it
3547 always said "this is a copy of the message, including all the headers",
3548 even if body_only or headers_only was set. It now gives an appropriate
3551 PH/09 Applied a patch from the Sieve maintainer which:
3553 o fixes some comments
3554 o adds the (disabled) notify extension core
3555 o adds some debug output for the result of if/elsif tests
3556 o points to the current vacation draft in the documentation
3557 and documents the missing references header update
3561 o fixes a bug in processing the envelope test (when testing
3562 multiple envelope elements, the last element determined the
3565 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
3566 Electronic Mail") by including:
3568 Auto-submitted: auto-generated
3570 in the messages that it generates (bounce messages and others, such as
3571 warnings). In the case of bounce messages for non-SMTP messages, there was
3572 also a typo: it was using "Auto_submitted" (underscore instead of
3573 hyphen). Since every message generated by Exim is necessarily in response
3574 to another message, thes have all been changed to:
3576 Auto-Submitted: auto-replied
3578 in accordance with these statements in the RFC:
3580 The auto-replied keyword:
3582 - SHOULD be used on messages sent in direct response to another
3583 message by an automatic process,
3585 - MUST NOT be used on manually-generated messages,
3587 - MAY be used on Delivery Status Notifications (DSNs) and Message
3588 Disposition Notifications (MDNs),
3590 - MUST NOT be used on messages generated by automatic or periodic
3591 processes, except for messages which are automatic responses to
3594 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
3595 to the default Received: header definition.
3597 PH/12 Added log selector acl_warn_skipped (default on).
3599 PH/13 After a successful wildlsearch lookup, discard the values of numeric
3600 variables because (a) they are in the wrong storage pool and (b) even if
3601 they were copied, it wouldn't work properly because of the caching.
3603 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
3604 checking when decoding. Apparently there are clients that generate
3605 overlong encoded strings. Why am I not surprised?
3607 PH/15 If the first argument of "${if match_address" was not empty, but did not
3608 contain an "@" character, Exim crashed. Now it writes a panic log message
3609 and treats the condition as false.
3611 PH/16 In autoreply, treat an empty string for "once" the same as unset.
3613 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
3614 extension "envelope-auth". The code is finished and in agreement with
3615 other implementations, but there is no documentation so far and in fact,
3616 nobody wrote the draft yet. This extension is currently #undef'ed, thus
3617 not changing the active code.
3619 Print executed "if" and "elsif" statements when debugging is used. This
3620 helps a great deal to understand what a filter does.
3622 Document more things not specified clearly in RFC3028. I had all this
3623 sorted out, when out of a sudden new issues came to my mind. Oops."
3625 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
3628 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
3629 canonical form (as documented). However, after a host name lookup from
3630 the IP address, check_host() was doing a simple string comparison with
3631 addresses acquired from the DNS when checking that the found name did
3632 have the original IP as one of its addresses. Since any found IPv6
3633 addresses are likely to be in abbreviated form, the comparison could
3634 fail. Luckily, there already exists a function for doing the comparison
3635 by converting both addresses to binary, so now that is used instead of
3636 the text comparison.
3638 PH/20 There was another similar case to PH/19, when a complete host name was
3639 given in a host list; looking up its IP address could give an abbreviated
3640 form, whereas the current host's name might or might not be abbreviated.
3641 The same fix has been applied.
3647 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
3648 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
3651 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
3652 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
3654 PH/03 Typo: missing ".o" in src/pcre/Makefile.
3656 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
3657 header line, restrict the check to what is listed in RFCs 2369 and 2929.
3658 Also, for "Auto-Submitted", treat anything other than "no" as
3659 non-personal, in accordance with RFC 3834. (Previously it treated
3660 anything starting "auto-" as non-personal.)
3662 TF/01 The control=submission/name=... option had a problem with syntax
3663 errors if the name included a slash character. The /name= option
3664 now slurps the rest of the string, so it can include any characters
3665 but it must come last in the list of options (after /sender_retain
3668 PH/05 Some modifications to the interface to the fake nameserver for the new
3676 TK/01 Added the "success_on_redirect" address verification option. See
3677 NewStuff for rationale and an example.
3679 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
3681 PH/02 Patch to exigrep to allow it to work on syslog lines.
3683 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
3684 fread() to skip over the body file's header line, because in Cygwin the
3685 header line is locked and is inaccessible.
3687 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
3688 co-exist for some time) to make it clear that it is the Exim ID that is
3689 referenced, not the Message-ID: header line.
3691 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
3692 string_format() function, because snprintf() does not exist on all
3695 PH/06 The use of forbid_filter_existstest now also locks out the use of the
3696 ${stat: expansion item.
3698 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
3699 protocol synchronization error", to keep the pedants happy.
3701 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
3702 well as for IRIX systems, when gcc is being used. See the host.c source
3705 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
3707 PH/10 Named domain lists were not working if used in a queue_smtp_domains
3710 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
3711 transport and to the smtp transport in LMTP mode.
3713 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
3715 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
3716 run a filter in a subprocess. This could lead to confusion in subsequent
3717 lookups in the parent process. There should also be a search_tidyup() at
3718 the end of the subprocess.
3720 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
3721 only if the host matched helo_try_verify_hosts, which caused the
3722 verification to occur when the EHLO/HELO command was issued. The ACL just
3723 tested the remembered result. Now, if a previous verification attempt has
3724 not happened, "verify = helo" does it there and then.
3726 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
3728 TK/03 Fix log output including CR from clamd.
3730 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
3731 exist provoked a memory error which could cause a segfault.
3733 PH/15 Installed PCRE 6.2
3735 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
3737 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
3738 of the problem. Specifically, suggested +O2 rather than +O1 for the
3741 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
3743 PH/20 If a delivery was routed to a non-standard port by means of an SRV
3744 record, the port was not correctly logged when the outgoing_port log
3745 selector was set (it logged the transort's default port).
3747 PH/21 Added support for host-specific ports to manualroute, queryprogram,
3748 fallback_hosts, and "hosts" in the smtp transport.
3750 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
3751 host errors such as "Connection refused".
3753 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
3754 authentication with radiusclient 0.4.9:
3756 - Error returned from rc_read_config was caught wrongly
3757 - Username/password not passed on to radius server due to wrong length.
3759 The presumption is that some radiusclient API changes for 4.51/PH/17
3760 were not taken care of correctly. The code is still untested by me (my
3761 Linux distribution still has 0.3.2 of radiusclient), but it was
3762 contributed by a Radius user.
3764 PH/24 When doing a callout, the value of $domain wasn't set correctly when
3765 expanding the "port" option of the smtp transport.
3767 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
3768 while reading a MIME header. Thanks to Tom Hughes for a patch.
3770 PH/24 Include config.h inside local_scan.h so that configuration settings are
3773 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
3774 that in an ACL for RCPT (for example), you can examine exactly what was
3777 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
3778 commands, but it was not correctly comparing the address with the actual
3779 client host address. Thus, it would show the EHLO address in Received:
3780 header lines when this was not necessary.
3782 PH/27 Added the % operator to ${eval:}.
3784 PH/28 Exim tries to create and chdir to its spool directory when it starts;
3785 it should be ignoring failures (because with -C, for example, it has lost
3786 privilege). It wasn't ignoring creation failures other than "already
3789 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
3792 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
3793 because an input line was too long, either on its own, or by virtue of
3794 too many continuations, the temporary file was not being removed, and the
3795 return code was incorrect.
3797 PH/31 Missing "BOOL" in function definition in filtertest.c.
3799 PH/32 Applied Sieve patches from the maintainer.
3801 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
3803 PH/33 Added "verify = not_blind".
3805 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
3806 Local/Makefile (with some defaults set). These are used in built scripts
3807 such as exicyclog, but they have never been used in the exim_install
3808 script (though there are many overriding facilities there). I have
3809 arranged that the exim_install script now takes note of these two
3812 PH/35 Installed configuration files for Dragonfly.
3814 PH/36 When a locally submitted message by a trusted user did not contain a
3815 From: header, and the sender address was obtained from -f or from an SMTP
3816 MAIL command, and the trusted user did not use -F to supply a sender
3817 name, $originator_name was incorrectly used when constructing a From:
3818 header. Furthermore, $originator_name was used for submission mode
3819 messages from external hosts without From: headers in a similar way,
3820 which is clearly wrong.
3822 PH/37 Added control=suppress_local_fixups.
3824 PH/38 When log_selector = +received_sender was set, and the addition of the
3825 sender made the log line's construction buffer exactly full, or one byte
3826 less than full, an overflow happened when the terminating "\n" was
3829 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
3830 when the result of a list match is failure because a DNS lookup failed.
3832 PH/40 RM_COMMAND is now used in the building process.
3834 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
3835 the "build-* directories that it finds.
3837 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
3838 address in a domain literal was a prefix of an interface address.
3840 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
3841 when verifying a sender address, unless rewrite_headers is false.
3843 PH/44 Wrote a long comment about why errors_to addresses are verified as
3844 recipients, not senders.
3846 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
3847 the ratelimit ACL was added.
3849 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
3851 PH/46 Added extra information about PostgreSQL errors to the error string.
3853 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
3854 avoiding the need to install special zones in a real server. This is
3855 backwards compatible; if it can't find the fake resolver, it drops back.
3856 Thus, both old and new test suites can be run.
3858 TF/02 Added util/ratelimit.pl
3860 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
3861 clock is set back in time.
3863 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
3864 Candler <B.Candler@pobox.com>.
3866 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
3867 OK for addresses that are the result of redirections.
3869 PH/48 A number of further additions for the benefit of the new test suite,
3870 including a fake gethostbyname() that interfaces to the fake DNS resolver
3873 TF/06 The fix for widen_domains has also been applied to qualify_single and
3874 search_parents which are the other dnslookup options that can cause
3877 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
3880 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
3882 TF/07 Exim produced the error message "an SRV record indicated no SMTP
3883 service" if it encountered an MX record with an empty target hostname.
3884 The message is now "an MX or SRV record indicated no SMTP service".
3886 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
3887 if the DNS of the sending site is misconfigured. This is quite a
3888 common situation. This change restores the behaviour of treating a
3889 helo verification defer as a failure.
3891 PH/51 If self=fail was set on a router, the bounce message did not include the
3892 actual error message.
3898 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
3900 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
3901 command itself ended in error, the underlying message about the transport
3902 filter timeout was being overwritten with the pipe command error. Now the
3903 underlying error message should be appended to the second error message.
3905 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
3907 PH/02 Reduce the amount of output that "make" produces by default. Full output
3908 can still be requested.
3910 PH/03 The warning log line about a condition test deferring for a "warn" verb
3911 was being output only once per connection, rather than after each
3912 occurrence (because it was using the same function as for successful
3913 "warn" verbs). This seems wrong, so I have changed it.
3915 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
3916 it should not have, which might have caused a crash in the right
3917 circumstances, but probably never did.
3919 PH/04 Installed a modified version of Tony Finch's patch to make submission
3920 mode fix the return path as well as the Sender: header line, and to
3921 add a /name= option so that you can make the user's friendly name appear
3924 TF/03 Added the control = fakedefer ACL modifier.
3926 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
3927 Mark Lowes for thorough testing.
3929 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
3931 TK/03 Merged latest SRS patch from Miles Wilton.
3933 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
3934 with the definition in sysexits.h (which is #included earlier).
3935 Fortunately, Exim does not actually use EX_OK. The code used to try to
3936 preserve the sysexits.h value, by assuming that macro definitions were
3937 scanned for macro replacements. I have been disabused of this notion,
3938 so now the code just undefines EX_OK before #including unistd.h.
3940 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
3941 in the smtp transport. When a block could not be written in a single
3942 write() function, the timeout was being re-applied to each part-write.
3943 This seems wrong - if the receiver was accepting one byte at a time it
3944 would take for ever. The timeout is now adjusted when this happens. It
3945 doesn't have to be particularly precise.
3947 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
3948 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
3950 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
3951 without a domain if the check to <postmaster@domain> fails.
3953 SC/01 Eximstats: added -xls and the ability to specify output files
3954 (patch written by Frank Heydlauf).
3956 SC/02 Eximstats: use FileHandles for outputting results.
3958 SC/03 Eximstats: allow any combination of xls, txt, and html output.
3960 SC/04 Eximstats: fixed display of large numbers with -nvr option
3962 SC/05 Eximstats: fixed merging of reports with empty tables.
3964 SC/06 Eximstats: added the -include_original_destination flag
3966 SC/07 Eximstats: removed tabs and trailing whitespace.
3968 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
3970 TK/06 MBOX spool code: Add real "From " MBOX separator line
3971 so the .eml file is really in mbox format (even though
3972 most programs do not really care). Patch from Alex Miller.
3974 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
3975 The latter is generated from $received_to and is only set if the
3976 message has one envelope recipient. SA can use these headers,
3977 obviously out-of-the-box. Patch from Alex Miller.
3979 PH/08 The ${def test on a variable was returning false if the variable's
3980 value was "0", contrary to what the specification has always said!
3981 The result should be true unless the variable is empty.
3983 PH/09 The syntax error of a character other than { following "${if
3984 def:variable_name" (after optional whitespace) was not being diagnosed.
3985 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
3986 accidental colon was present, for example, could give incorrect results.
3988 PH/10 Tidied the code in a number of places where the st_size field of a stat()
3989 result is used (not including appendfile, where other changes are about
3992 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
3993 This involved changing a lot of size variables from int to off_t. It
3994 should work with maildirs and everything.
3996 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
3997 spamd dying while we are connected to it.
3999 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
4002 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
4003 function for BDB 4.3.
4005 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
4007 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
4008 the sources for PCRE have been re-arranged and more files are now
4011 PH/15 The code I had for printing potentially long long variables in PH/11
4012 above was not the best (it lost precision). The length of off_t variables
4013 is now inspected at build time, and an appropriate printing format (%ld
4014 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
4015 to be "long long int" or "long int". This is needed for the internal
4016 formatting function string_vformat().
4018 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
4019 the configuration file to be ":syslog", then the script "guesses" where
4020 the logs files are, rather than using the compiled in default. In our
4021 case the guess is not the same as the compiled default, so the script
4022 suddenly stopped working when I started to use syslog. The patch checks
4023 to see if log_file_path is "". If so, it attempts to read it from exim
4024 with no configuration file to get the compiled in version, before it
4025 falls back to the previous guessing code."
4027 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
4028 implementing BATV in an Exim configuration. See NewStuff for the gory
4031 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
4032 Makefile that are specific to HP-UX.
4034 PH/18 If the "use_postmaster" option was set for a recipient callout together
4035 with the "random" option, the postmaster address was used as the MAIL
4036 FROM address for the random test, but not for the subsequent recipient
4037 test. It is now used for both.
4039 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
4040 patch removes a few documentation additions to RFC 3028, because the
4041 latest draft now contains them. It adds the new en;ascii-case comparator
4042 and a new error check for 8bit text in MIME parts. Comparator and
4043 require names are now matched exactly. I enabled the subaddress
4044 extension, but it is not well tested yet (read: it works for me)."
4046 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
4047 rework some of the code of TK/09 above to avoid the hardwired use of
4048 "%lld" and "long long". Replaced the call to snprintf() with a call to
4051 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
4052 records point to non-existent hosts", "retry timeout exceeded", and
4053 "retry time not reached for any host after a long failure period".
4055 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
4056 experimental DomainKeys support:
4058 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
4059 (2) On an error such as an illegally used "control", the wrong name for
4060 the control was given.
4062 These problems did NOT occur unless DomainKeys support was compiled.
4064 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
4066 PH/24 Added ${if match_ip condition.
4068 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
4069 databases so that it will be absolutely obvious if a crash occurs in the
4070 DB library. This is a regular occurrence (often caused by mis-matched
4073 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
4074 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
4075 picked up on a user's system that detects such things. There doesn't seem
4076 to be a gcc warning option for this - only an attribute that has to be
4077 put on the function's prototype. It seems that in Fedora Core 4 they have
4078 set this on a number of new functions. No doubt there will be more in due
4081 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
4082 specify a transport. However, if an address that was verified by such a
4083 router was the subject of a callout, Exim crashed because it tried to
4084 read the rcpt_include_affixes from the non-existent transport. Now it
4085 just assumes that the setting of that option is false. This bug was
4086 introduced by 4.51/PH/31.
4088 PH/28 Changed -d+all to exclude +memory, because that information is very
4089 rarely of interest, but it makes the output a lot bigger. People tend to
4090 do -d+all out of habit.
4092 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
4093 code in os-type was giving problems when libc.so lives in lib64, like on
4096 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
4097 aren't the modern standard, and it seems that some systems' include files
4098 don't always have them. Exim was already checking for some of the newer
4099 ones like T_AAAA, and defining it itself. I've added checks for all the
4100 record types that Exim uses.
4102 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
4103 not automatically generating a new one, as it is supposed to. This
4104 prevented TLS from working. If the file did exist, but contained invalid
4105 data, a new version was generated, as expected. It was only the case of a
4106 non-existent file that was broken.
4108 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
4109 with a change in libdomainkeys > 0.64.
4111 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
4112 from DNS. If the selector record carries the flag, it now has
4113 precedence over the domain-wide flag.
4115 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
4117 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
4118 the use of a port name that isn't defined in /etc/services) occurred, the
4119 message was deferred as in a normal delivery, and thus remained on the
4120 spool, instead of being failed because of the mua_wrapper setting. This
4121 is now fixed, and I tidied up some of the mua_wrapper messages at the
4124 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
4125 the messages in a hash of arrays rather than using individual hashes.
4126 This is a bit cleaner and results in dramatic memory savings, albeit
4127 at a slight CPU cost.
4129 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
4130 as requested by Marc Sherman.
4132 SC/10 Eximstats: added histograms for user specified patterns as requested
4135 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
4137 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
4138 fopen() in the content-scanning modules that did not already have it.
4144 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
4145 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
4147 TK/02 Fix ACL "control" statement not being available in MIME ACL.
4149 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
4151 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
4152 to test Sieve filters that use "vacation".
4154 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
4155 that changes the way the GnuTLS parameters are stored in the cache file.
4156 The new format can be generated externally. For backward compatibility,
4157 if the data in the cache doesn't make sense, Exim assumes it has read an
4158 old-format file, and it generates new data and writes a new file. This
4159 means that you can't go back to an older release without removing the
4162 PH/03 A redirect router that has both "unseen" and "one_time" set does not
4163 work if there are any delivery delays because "one_time" forces the
4164 parent to be marked "delivered", so its unseen clone is never tried
4165 again. For this reason, Exim now forbids the simultaneous setting of
4168 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
4169 redirected to themselves ("homonym" addresses). Read the long ChangeLog
4170 entry if you want to know the details. The fix, however, neglected to
4171 consider the case when local delivery batching is involved. The test for
4172 "previously delivered" was not happening when checking to see if an
4173 address could be batched with a previous (undelivered) one; under
4174 certain circumstances this could lead to multiple deliveries to the same
4177 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
4178 in its include files, and this causes problems building Exim.
4180 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
4181 header_syntax) but Exim was just ignoring anything given after a slash.
4182 In particular, this caused confusion with an attempt to use "verify =
4183 reverse_host_lookup/defer_ok". An error is now given when options are
4184 supplied for verify items that do not have them. (Maybe reverse_host_
4185 lookup should have a defer_ok option, but that's a different point.)
4187 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
4188 defined by RFC 821) to 2048, because there were problems with some AUTH
4189 commands, and RFC 1869 says the size should be increased for extended
4190 SMTP commands that take arguments.
4192 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
4195 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
4196 "unknown" error; now it says that the functionality isn't in the binary.
4198 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
4199 an address' error message when a string expansion fails (syntax or
4200 whatever). Otherwise the password may appear in the log. Following change
4201 PH/42 below, there is no longer a chance of it appearing in a bounce
4204 PH/11 Installed exipick version 20050225.0 from John Jetmore.
4206 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
4207 of its addresses was ever tried. (Bugzilla bug #2.)
4209 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
4210 the result incorrectly in the debug output. (It correctly added a newline
4211 to what was transported.)
4213 TF/01 Added $received_time.
4215 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
4216 commented out examples of how to interface to a virus scanner and to
4217 SpamAssassin. Also added commented examples of av_scanner and
4218 spamd_address settings.
4220 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
4221 and controls are allowed in which ACLs. There were a couple of minor
4222 errors. Some of the entries in the conditions table (which is a table of
4223 where they are NOT allowed) were getting very unwieldy; rewrote them as a
4224 negation of where the condition IS allowed.
4226 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
4228 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
4229 header file does not have a version number, so I've had to invent a new
4230 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
4231 API. The code is untested by me (my Linux distribution still has 0.3.2 of
4232 radiusclient), but it was contributed by a Radius user.
4234 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
4235 files or directories, for OpenSSL.
4237 PH/19 When an Exim process that is running as root has to create an Exim log
4238 file, it does so in a subprocess that runs as exim:exim so as to get the
4239 ownership right at creation (otherwise, other Exim processes might see
4240 the file with the wrong ownership). There was no test for failure of this
4241 fork() call, which would lead to the process getting stuck as it waited
4242 for a non-existent subprocess. Forks do occasionally fail when resources
4243 run out. I reviewed all the other calls to fork(); they all seem to check
4246 PH/20 When checking for unexpected SMTP input at connect time (before writing
4247 the banner), Exim was not dealing correctly with a non-positive return
4248 from the read() function. If the client had disconnected by this time,
4249 the result was a log entry for a synchronization error with an empty
4250 string after "input=" when read() returned zero. If read() returned -1
4251 (an event I could not check), uninitialized data bytes were printed.
4252 There were reports of junk text (parts of files, etc) appearing after
4255 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
4257 PH/22 Added support for macro redefinition, and (re)definition in between
4258 driver and ACL definitions.
4260 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
4261 forgetting to use the resulting value; it was using the unexpanded value.
4263 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
4264 hadn't been configured. The fix is from Juergen Kreileder, who
4265 understands it better than I do:
4267 "Here's what I see happening with three configured cyrus_sasl
4268 authenticators configured (plain, login, cram-md5):
4270 On startup auth_cyrus_sasl_init() gets called for each of these.
4271 This means three calls to sasl_listmech() without a specified mech_list.
4272 => SASL tests which mechs of all available mechs actually work
4273 => three warnings about OTP not working
4274 => the returned list contains: plain, login, cram-md5, digest-md5, ...
4276 With the patch, sasl_listmech() also gets called three times. But now
4277 SASL's mech_list option is set to the server_mech specified in the the
4278 authenticator. Or in other words, the answer from sasl_listmech()
4279 gets limited to just the mech you're testing for (which is different
4281 => the return list contains just 'plain' or 'login', 'cram-md5' or
4282 nothing depending on the value of ob->server_mech.
4284 I've just tested the patch: Authentication still works fine,
4285 unavailable mechs specified in the exim configuration are still
4286 caught, and the auth.log warnings about OTP are gone."
4288 PH/25 When debugging is enabled, the contents of the command line are added
4289 to the debugging output, even when log_selector=+arguments is not
4292 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
4293 answer is "GNU", and only if the return is "GNU/something" is the answer
4296 PH/27 $acl_verify_message is now set immediately after the failure of a
4297 verification in an ACL, and so is available in subsequent modifiers. In
4298 particular, the message can be preserved by coding like this:
4300 warn !verify = sender
4301 set acl_m0 = $acl_verify_message
4303 Previously, $acl_verify_message was set only while expanding "message"
4304 and "log_message" when a very denied access.
4306 PH/28 Modified OS/os.c-Linux with
4308 -#ifndef OS_LOAD_AVERAGE
4309 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
4311 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
4312 nomenclature these days.)
4314 PH/29 Installed patch from the Sieve maintainer that adds the options
4315 sieve_useraddress and sieve_subaddress to the redirect router.
4317 PH/30 In these circumstances:
4318 . Two addresses routed to the same list of hosts;
4319 . First host does not offer TLS;
4320 . First host accepts first address;
4321 . First host gives temporary error to second address;
4322 . Second host offers TLS and a TLS session is established;
4323 . Second host accepts second address.
4324 Exim incorrectly logged both deliveries with the TLS parameters (cipher
4325 and peerdn, if requested) that were in fact used only for the second
4328 PH/31 When doing a callout as part of verifying an address, Exim was not paying
4329 attention to any local part prefix or suffix that was matched by the
4330 router that accepted the address. It now behaves in the same way as it
4331 does for delivery: the affixes are removed from the local part unless
4332 rcpt_include_affixes is set on the transport.
4334 PH/32 Add the sender address, as F=<...>, to the log line when logging a
4335 timeout during the DATA phase of an incoming message.
4337 PH/33 Sieve envelope tests were broken for match types other than :is. I have
4338 applied a patch sanctioned by the Sieve maintainer.
4340 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
4341 the uid or gid is negative. A case of a negative gid caused this to be
4342 noticed. The fix allows for either to be negative.
4344 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
4345 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
4348 PH/36 The change PH/12 above was broken. Fixed it.
4350 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
4351 the grounds that routing the same address twice would always produce the
4352 same answer. This might have been true once, but it is certainly no
4353 longer true now. Routing a child address may depend on the previous
4354 routing that produced that child. Some complicated redirection strategies
4355 went wrong when messages had multiple recipients, and made Exim's
4356 behaviour dependent on the order in which the addresses were given.
4358 I have moved the duplicate checking until after the routing is complete.
4359 Exim scans the addresses that are assigned to local and remote
4360 transports, and removes any duplicates. This means that more work will be
4361 done, as duplicates will always all be routed, but duplicates are
4362 presumably rare, so I don't expect this is of any significance.
4364 For deliveries to pipes, files, and autoreplies, the duplicate checking
4365 still happens during the routing process, since they are not going to be
4368 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
4369 It corrects a timeout issue with spamd. This is Ian's comment: "The
4370 background is that sometimes spamd either never reads data from a
4371 connection it has accepted, or it never writes response data. The exiscan
4372 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
4373 blindly assumes that writes won't block so it may never time out."
4375 PH/39 Allow G after quota size as well as K and M.
4377 PH/40 The value set for $authenticated_id in an authenticator may not contain
4378 binary zeroes or newlines because the value is written to log lines and
4379 to spool files. There was no check on this. Now the value is run through
4380 the string_printing() function so that such characters are converted to
4381 printable escape sequences.
4383 PH/41 $message_linecount is a new variable that contains the total number of
4384 lines in the message. Compare $body_linecount, which is the count for the
4387 PH/42 Exim no longer gives details of delivery errors for specific addresses in
4388 bounce and delay warning messages, except in certain special cases, which
4391 (a) An SMTP error message from a remote host;
4392 (b) A message specified in a :fail: redirection;
4393 (c) A message specified in a "fail" command in a system filter;
4394 (d) A message specified in a FAIL return from the queryprogram router;
4395 (e) A message specified by the cannot_route_message router option.
4397 In these cases only, Exim does include the error details in bounce and
4398 warning messages. There are also a few cases where bland messages such
4399 as "unrouteable address" or "local delivery error" are given.
4401 PH/43 $value is now also set for the "else" part of a ${run expansion.
4403 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
4404 being worked on, but at least Exim now implements the latest version to
4407 PH/45 In a pipe transport, although a timeout while waiting for the pipe
4408 process to complete was treated as a delivery failure, a timeout while
4409 writing the message to the pipe was logged, but erroneously treated as a
4410 successful delivery. Such timeouts include transport filter timeouts. For
4411 consistency with the overall process timeout, these timeouts are now
4412 treated as errors, giving rise to delivery failures by default. However,
4413 there is now a new Boolean option for the pipe transport called
4414 timeout_defer, which, if set TRUE, converts the failures into defers for
4415 both kinds of timeout. A transport filter timeout is now identified in
4418 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
4419 systems where "make" and "gmake" are different, calling "gmake" at top
4420 level broke things. I've arranged for the value of $(MAKE) to be passed
4421 from the Makefile to this script so that it can call the same version of
4425 A note about Exim versions 4.44 and 4.50
4426 ----------------------------------------
4428 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
4429 changes of various kinds. As a consequence, a big documentation update was
4430 needed. This delayed the release for rather longer than seemed good, especially
4431 in the light of a couple of (minor) security issues. Therefore, the changes
4432 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
4433 release. So 4.44 and 4.50 are in effect two different branches that both start
4436 I have left the 4.50 change log unchanged; it contains all the changes since
4437 4.43. The change log for 4.44 is below; many of its items are identical to
4438 those for 4.50. This seems to be the most sensible way to preserve the
4439 historical information.
4445 1. Minor wording change to the doc/README.SIEVE file.
4447 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
4448 computation of the current number of files was incorrect.
4450 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4451 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4454 4. Give more explanation in the error message when the command for a transport
4455 filter fails to execute.
4457 5. There are several places where Exim runs a non-Exim command in a
4458 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4459 was being done only for the command run by the queryprogram router. It is
4460 now done for all such subprocesses. The other cases are: ${run, transport
4461 filters, and the commands run by the lmtp and pipe transports.
4463 6. Added CONFIGURE_GROUP build-time option.
4465 7. Some older OS have a limit of 256 on the maximum number of file
4466 descriptors. Exim was using setrlimit() to set 1000 as a large value
4467 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4468 systems. I've change it so that if it can't get 1000, it tries for 256.
4470 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4471 was an oversight, and furthermore, ever since the addition of extra
4472 controls (e.g. 4.43/32), the checks on when to allow different forms of
4473 "control" were broken. There should now be diagnostics for all cases when a
4474 control that does not make sense is encountered.
4476 9. Added the /retain_sender option to "control=submission".
4478 10. $recipients is now available in the predata ACL (oversight).
4480 11. Tidy the search cache before the fork to do a delivery from a message
4481 received from the command line. Otherwise the child will trigger a lookup
4482 failure and thereby defer the delivery if it tries to use (for example) a
4483 cached ldap connection that the parent has called unbind on.
4485 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4486 of $address_data from the recipient verification was clobbered by the
4487 sender verification.
4489 13. The value of address_data from a sender verification is now available in
4490 $sender_address_data in subsequent conditions in the ACL statement.
4492 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
4494 15. Added a new option "connect=<time>" to callout options, to set a different
4497 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4498 was its contents. (It was OK if the option was not defined at all.)
4500 17. A "Completed" log line is now written for messages that are removed from
4501 the spool by the -Mrm option.
4503 18. New variables $sender_verify_failure and $recipient_verify_failure contain
4504 information about exactly what failed.
4506 19. Added -dd to debug only the daemon process.
4508 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
4509 handles timeouts, both on the server side and network timeouts. Renamed the
4510 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
4512 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
4513 It is now set to "smtps".
4515 22. $host_address is now set to the target address during the checking of
4516 ignore_target_hosts.
4518 23. When checking ignore_target_hosts for an ipliteral router, no host name was
4519 being passed; this would have caused $sender_host_name to have been used if
4520 matching the list had actually called for a host name (not very likely,
4521 since this list is usually IP addresses). A host name is now passed as
4524 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4525 code that specifies a non-restarting handler (typically sigaction() in
4526 modern systems) in an attempt to fix a rare and obscure crash bug.
4528 25. Narrowed the window for a race in the daemon that could cause it to ignore
4529 SIGCHLD signals. This is not a major problem, because they are used only to
4530 wake it up if nothing else does.
4532 26. A malformed maildirsize file could cause Exim to calculate negative values
4533 for the mailbox size or file count. Odd effects could occur as a result.
4534 The maildirsize information is now recalculated if the size or filecount
4537 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4538 support for a long time. Removed HAVE_SYS_VFS_H.
4540 28. Installed the latest version of exipick from John Jetmore.
4542 29. In an address list, if the pattern was not a regular expression, an empty
4543 subject address (from a bounce message) matched only if the pattern was an
4544 empty string. Non-empty patterns were not even tested. This was the wrong
4545 because it is perfectly reasonable to use an empty address as part of a
4546 database query. An empty address is now tested by patterns that are
4547 lookups. However, all the other forms of pattern expect the subject to
4548 contain a local part and a domain, and therefore, for them, an empty
4549 address still always fails if the pattern is not itself empty.
4551 30. Exim went into a mad DNS loop when attempting to do a callout where the
4552 host was specified on an smtp transport, and looking it up yielded more
4553 than one IP address.
4555 31. Re-factored the code for checking spool and log partition space into a
4556 function that finds that data and another that does the check. The former
4557 is then used to implement four new variables: $spool_space, $log_space,
4558 $spool_inodes, and $log_inodes.
4560 32. The RFC2047 encoding function was originally intended for short strings
4561 such as real names; it was not keeping to the 75-character limit for
4562 encoded words that the RFC imposes. It now respects the limit, and
4563 generates multiple encoded words if necessary. To be on the safe side, I
4564 have increased the buffer size for the ${rfc2047: expansion operator from
4567 33. It is now permitted to omit both strings after an "if" condition; if the
4568 condition is true, the result is "true". As before, when the second string
4569 is omitted, a false condition yields an empty string. This makes it less
4570 cumbersome to write custom ACL and router conditions.
4572 34. Failure to deliver a bounce message always caused it to be frozen, even if
4573 there was an errors_to setting on the router. The errors_to setting is now
4576 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4577 canonical form (fully expanded) before being placed in
4578 $sender_host_address.
4580 36. The table in the code that translates DNS record types into text (T_A to
4581 "A" for instance) was missing entries for NS and CNAME. It is just possible
4582 that this could have caused confusion if both these types were looked up
4583 for the same domain, because the text type is used as part of Exim's
4584 per-process caching. But the chance of anyone hitting this buglet seems
4587 37. The dnsdb lookup has been extended in a number of ways.
4589 (1) There is a new type, "zns", which walks up the domain tree until it
4590 finds some nameserver records. It should be used with care.
4592 (2) There is a new type, "mxh", which is like "mx" except that it returns
4593 just the host names, not the priorities.
4595 (3) It is now possible to give a list of domains (or IP addresses) to be
4596 looked up. The behaviour when one of the lookups defers can be
4597 controlled by a keyword.
4599 (4) It is now possible to specify the separator character for use when
4600 multiple records are returned.
4602 38. The dnslists ACL condition has been extended: it is now possible to supply
4603 a list of IP addresses and/or domains to be looked up in a particular DNS
4606 39. Added log_selector=+queue_time_overall.
4608 40. When running the queue in the test harness, wait just a tad after forking a
4609 delivery process, to get repeatability of debugging output.
4611 41. Include certificate and key file names in error message when GnuTLS fails
4612 to set them up, because the GnuTLS error message doesn't include the name
4613 of the failing file when there is a problem reading it.
4615 42. Allow both -bf and -bF in the same test run.
4617 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
4619 44. The "Exiscan patch" is now merged into the mainline Exim source.
4621 45. Sometimes the final signoff response after QUIT could fail to get
4622 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4623 < 0 before doing a fflush(). This bug looks as though it goes back to the
4624 introduction of TLS in release 3.20, but "sometimes" must have been rare
4625 because the tests only now provoked it.
4627 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
4628 (this can affect the format of dates).
4630 47. exim_tidydb, when checking for the continued existence of a message for
4631 which it has found a message-specific retry record, was not finding
4632 messages that were in split spool directories. Consequently, it was
4633 deleting retry records that should have stayed in existence.
4635 48. Steve fixed some bugs in eximstats.
4637 49. The SPA authentication driver was not abandoning authentication and moving
4638 on to the next authenticator when an expansion was forced to fail,
4639 contradicting the general specification for all authenticators. Instead it
4640 was generating a temporary error. It now behaves as specified.
4642 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
4643 (the order specifies the preference for clients). The order is now AES256,
4644 AES128, 3DES, ARCFOUR128.
4646 51. Small patch to Sieve code - explicitly set From: when generating an
4649 52. Exim crashed if a remote delivery caused a very long error message to be
4650 recorded - for instance if somebody sent an entire SpamAssassin report back
4651 as a large number of 550 error lines. This bug was coincidentally fixed by
4652 increasing the size of one of Exim's internal buffers (big_buffer) that
4653 happened as part of the Exiscan merge. However, to be on the safe side, I
4654 have made the code more robust (and fixed the comments that describe what
4657 53. Now that there can be additional text after "Completed" in log lines (if
4658 the queue_time_overall log selector is set), a one-byte patch to exigrep
4659 was needed to allow it to recognize "Completed" as not the last thing in
4662 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
4663 patch that reportedly fixes this has been added. I am not expert enough to
4664 create a test for it. This is what the patch creator wrote:
4666 "I found a little strange behaviour of ldap code when working with
4667 Windows 2003 AD Domain, where users was placed in more than one
4668 Organization Units. When I tried to give exim partial DN, the exit code
4669 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
4670 But simultaneously result of request was absolutely normal ldap result,
4671 so I produce this patch..."
4673 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
4674 REFERENCE, so I have modified the code to exclude the patch when that macro
4677 55. Some experimental protocols are using DNS PTR records for new purposes. The
4678 keys for these records are domain names, not reversed IP addresses. The
4679 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4680 leaves it alone. Component reversal etc. now happens only for IP addresses.
4683 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
4685 57. Double the size of the debug message buffer (to 2048) so that more of very
4686 long debug lines gets shown.
4688 58. The exicyclog utility now does better if the number of log files to keep
4689 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
4691 59. Two changes related to the smtp_active_hostname option:
4693 (1) $smtp_active_hostname is now available as a variable.
4694 (2) The default for smtp_banner uses $smtp_active_hostname instead
4695 of $primary_hostname.
4697 60. The host_aton() function is supposed to be passed a string that is known
4698 to be a valid IP address. However, in the case of IPv6 addresses, it was
4699 not checking this. This is a hostage to fortune. Exim now panics and dies
4700 if the condition is not met. A case was found where this could be provoked
4701 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4702 components; fortuitously, this particular loophole had already been fixed
4703 by change 4.50/55 above.
4705 If there are any other similar loopholes, the new check in host_aton()
4706 itself should stop them being exploited. The report I received stated that
4707 data on the command line could provoke the exploit when Exim was running as
4708 exim, but did not say which command line option was involved. All I could
4709 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4710 running as the user.
4713 61. There was a buffer overflow vulnerability in the SPA authentication code
4714 (which came originally from the Samba project). I have added a test to the
4715 spa_base64_to_bits() function which I hope fixes it.
4718 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
4719 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
4721 63. The daemon start-up calls getloadavg() while still root for those OS that
4722 need the first call to be done as root, but it missed one case: when
4723 deliver_queue_load_max is set with deliver_drop_privilege. This is
4724 necessary for the benefit of the queue runner, because there is no re-exec
4725 when deliver_drop_privilege is set.
4727 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
4728 This has been fixed.
4730 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4731 was in use, was not putting the data itself into the right store pool;
4732 consequently, it could be overwritten for a subsequent message in the same
4733 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4736 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
4738 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
4739 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
4740 the function were treating the return as a boolean value, which happened to
4741 work because 0=false and not-0=true, but is not correct code.
4743 68. The host_aton() function was not handling scoped IPv6 addresses (those
4744 with, for example, "%eth0" on the end) correctly.
4746 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
4747 negated items (that is, ~something) in unsigned ints. Some compilers
4748 apparently mutter when there is no cast.
4750 70. If an address verification called from an ACL failed, and did not produce a
4751 user-specific message (i.e. there was only a "system" message), nothing was
4752 put in $acl_verify_message. In this situation, it now puts the system
4755 71. Change 4.23/11 added synchronization checking at the start of an SMTP
4756 session; change 4.31/43 added the unwanted input to the log line - except
4757 that it did not do this in the start of session case. It now does.
4759 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
4760 This is wrong and can cause the other end to generate a synchronization
4761 error if it is another Exim or anything else that does the synchronization
4762 check. A QUIT command is no longer sent after a timeout.
4764 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
4765 during host lookups.
4767 74. The defer_ok option of callout verification was not working if it was used
4768 when verifying addresses in header lines, that is, for this case:
4770 verify = header_sender/callout=defer_ok
4772 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
4773 those file descriptors could be used for SMTP connections. If anything
4774 wrote to stderr (the example that came up was "warn" in embedded Perl), it
4775 could be sent to the SMTP client, causing chaos. The daemon now opens
4776 stdin, stdout, and stderr to /dev/null when it puts itself into the
4779 76. Arrange for output from Perl's "warn" command to be written to Exim's main
4780 log by default. The user can override this with suitable Perl magic.
4782 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
4783 the log message when discard triggers, was not working for the DATA ACL or
4784 for the non-SMTP ACL.
4786 78. Error message wording change in sieve.c.
4788 79. If smtp_accept_max_per_host was set, the number of connections could be
4789 restricted to fewer than expected, because the daemon was trying to set up
4790 a new connection before checking whether the processes handling previous
4791 connections had finished. The check for completed processes is now done
4792 earlier. On busy systems, this bug wouldn't be noticed because something
4793 else would have woken the daemon, and it would have reaped the completed
4796 80. If a message was submitted locally by a user whose login name contained one
4797 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
4798 It caused a spool format error. I have fixed the spool reading code. A
4799 related problem was that the "from" clause in the Received: line became
4800 illegal because of the space(s). It is now covered by ${quote_local_part.
4802 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
4805 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
4807 83. Patch from Sieve maintainer to fix unterminated string problem in
4808 "vacation" handling.
4810 84. Some minor changes to the Linux configuration files to help with other
4811 OS variants using glibc.
4813 85. One more patch for Sieve to update vacation handling to latest spec.
4816 ----------------------------------------------------
4817 See the note above about the 4.44 and 4.50 releases.
4818 ----------------------------------------------------
4824 1. Change 4.43/35 introduced a bug that caused file counts to be
4825 incorrectly computed when quota_filecount was set in an appendfile
4828 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4829 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4832 3. Give more explanation in the error message when the command for a transport
4833 filter fails to execute.
4835 4. There are several places where Exim runs a non-Exim command in a
4836 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4837 was being done only for the command run by the queryprogram router. It is
4838 now done for all such subprocesses. The other cases are: ${run, transport
4839 filters, and the commands run by the lmtp and pipe transports.
4841 5. Some older OS have a limit of 256 on the maximum number of file
4842 descriptors. Exim was using setrlimit() to set 1000 as a large value
4843 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4844 systems. I've change it so that if it can't get 1000, it tries for 256.
4846 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4847 was an oversight, and furthermore, ever since the addition of extra
4848 controls (e.g. 4.43/32), the checks on when to allow different forms of
4849 "control" were broken. There should now be diagnostics for all cases when a
4850 control that does not make sense is encountered.
4852 7. $recipients is now available in the predata ACL (oversight).
4854 8. Tidy the search cache before the fork to do a delivery from a message
4855 received from the command line. Otherwise the child will trigger a lookup
4856 failure and thereby defer the delivery if it tries to use (for example) a
4857 cached ldap connection that the parent has called unbind on.
4859 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4860 of $address_data from the recipient verification was clobbered by the
4861 sender verification.
4863 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4864 was its contents. (It was OK if the option was not defined at all.)
4866 11. A "Completed" log line is now written for messages that are removed from
4867 the spool by the -Mrm option.
4869 12. $host_address is now set to the target address during the checking of
4870 ignore_target_hosts.
4872 13. When checking ignore_target_hosts for an ipliteral router, no host name was
4873 being passed; this would have caused $sender_host_name to have been used if
4874 matching the list had actually called for a host name (not very likely,
4875 since this list is usually IP addresses). A host name is now passed as
4878 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4879 code that specifies a non-restarting handler (typically sigaction() in
4880 modern systems) in an attempt to fix a rare and obscure crash bug.
4882 15. Narrowed the window for a race in the daemon that could cause it to ignore
4883 SIGCHLD signals. This is not a major problem, because they are used only to
4884 wake it up if nothing else does.
4886 16. A malformed maildirsize file could cause Exim to calculate negative values
4887 for the mailbox size or file count. Odd effects could occur as a result.
4888 The maildirsize information is now recalculated if the size or filecount
4891 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4892 support for a long time. Removed HAVE_SYS_VFS_H.
4894 18. Updated exipick to current release from John Jetmore.
4896 19. Allow an empty sender to be matched against a lookup in an address list.
4897 Previously the only cases considered were a regular expression, or an
4900 20. Exim went into a mad DNS lookup loop when doing a callout where the
4901 host was specified on the transport, if the DNS lookup yielded more than
4904 21. The RFC2047 encoding function was originally intended for short strings
4905 such as real names; it was not keeping to the 75-character limit for
4906 encoded words that the RFC imposes. It now respects the limit, and
4907 generates multiple encoded words if necessary. To be on the safe side, I
4908 have increased the buffer size for the ${rfc2047: expansion operator from
4911 22. Failure to deliver a bounce message always caused it to be frozen, even if
4912 there was an errors_to setting on the router. The errors_to setting is now
4915 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4916 canonical form (fully expanded) before being placed in
4917 $sender_host_address.
4919 24. Updated eximstats to version 1.33
4921 25. Include certificate and key file names in error message when GnuTLS fails
4922 to set them up, because the GnuTLS error message doesn't include the name
4923 of the failing file when there is a problem reading it.
4925 26. Expand error message when OpenSSL has problems setting up cert/key files.
4928 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
4929 (this can affect the format of dates).
4931 28. exim_tidydb, when checking for the continued existence of a message for
4932 which it has found a message-specific retry record, was not finding
4933 messages that were in split spool directories. Consequently, it was
4934 deleting retry records that should have stayed in existence.
4936 29. eximstats updated to version 1.35
4937 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
4938 1.35 - bugfix such that pie charts by volume are generated correctly
4940 30. The SPA authentication driver was not abandoning authentication and moving
4941 on to the next authenticator when an expansion was forced to fail,
4942 contradicting the general specification for all authenticators. Instead it
4943 was generating a temporary error. It now behaves as specified.
4945 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
4946 (the order specifies the preference for clients). The order is now AES256,
4947 AES128, 3DES, ARCFOUR128.
4949 31. Small patch to Sieve code - explicitly set From: when generating an
4952 32. Exim crashed if a remote delivery caused a very long error message to be
4953 recorded - for instance if somebody sent an entire SpamAssassin report back
4954 as a large number of 550 error lines. This bug was coincidentally fixed by
4955 increasing the size of one of Exim's internal buffers (big_buffer) that
4956 happened as part of the Exiscan merge. However, to be on the safe side, I
4957 have made the code more robust (and fixed the comments that describe what
4960 33. Some experimental protocols are using DNS PTR records for new purposes. The
4961 keys for these records are domain names, not reversed IP addresses. The
4962 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4963 leaves it alone. Component reversal etc. now happens only for IP addresses.
4966 34. The host_aton() function is supposed to be passed a string that is known
4967 to be a valid IP address. However, in the case of IPv6 addresses, it was
4968 not checking this. This is a hostage to fortune. Exim now panics and dies
4969 if the condition is not met. A case was found where this could be provoked
4970 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4971 components; fortuitously, this particular loophole had already been fixed
4972 by change 4.50/55 or 4.44/33 above.
4974 If there are any other similar loopholes, the new check in host_aton()
4975 itself should stop them being exploited. The report I received stated that
4976 data on the command line could provoke the exploit when Exim was running as
4977 exim, but did not say which command line option was involved. All I could
4978 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4979 running as the user.
4982 35. There was a buffer overflow vulnerability in the SPA authentication code
4983 (which came originally from the Samba project). I have added a test to the
4984 spa_base64_to_bits() function which I hope fixes it.
4987 36. The daemon start-up calls getloadavg() while still root for those OS that
4988 need the first call to be done as root, but it missed one case: when
4989 deliver_queue_load_max is set with deliver_drop_privilege. This is
4990 necessary for the benefit of the queue runner, because there is no re-exec
4991 when deliver_drop_privilege is set.
4993 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4994 was in use, was not putting the data itself into the right store pool;
4995 consequently, it could be overwritten for a subsequent message in the same
4996 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4999 38. Sometimes the final signoff response after QUIT could fail to get
5000 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
5001 < 0 before doing a fflush(). This bug looks as though it goes back to the
5002 introduction of TLS in release 3.20, but "sometimes" must have been rare
5003 because the tests only now provoked it.
5009 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
5010 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
5011 Now it returns an integer. A number of calls were still expecting a T/F
5012 return. Fortuitously, in all cases, the tests worked in OK situations,
5013 which is the norm. However, things would have gone wrong on any write
5014 failures on the smtp file descriptor. This function is used when sending
5015 messages over SMTP and also when doing verify callouts.
5017 2. When Exim is called to do synchronous delivery of a locally submitted
5018 message (the -odf or -odi options), it no longer closes stderr before doing
5021 3. Implemented the mua_wrapper option.
5023 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
5025 5. Implemented the functions header_remove(), header_testname(),
5026 header_add_at_position(), and receive_remove_recipient(), and exported them
5029 6. If an ACL "warn" statement specified the addition of headers, Exim already
5030 inserted X-ACL-Warn: at the start if there was no header name. However, it
5031 was not making this test for the second and subsequent header lines if
5032 there were newlines in the string. This meant that an invalid header could
5033 be inserted if Exim was badly configured.
5035 7. Allow an ACL "warn" statement to add header lines at the start or after all
5036 the Received: headers, as well as at the end.
5038 8. Added the rcpt_4xx retry error code.
5040 9. Added postmaster_mailfrom=xxx to callout verification option.
5042 10. Added mailfrom=xxxx to the callout verification option, for verify=
5045 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
5046 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
5048 12. Inserted some casts to stop certain compilers warning when using pointer
5049 differences as field lengths or precisions in printf-type calls (mostly
5050 affecting debugging statements).
5052 13. Added optional readline() support for -be (dynamically loaded).
5054 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
5055 same clock tick as a message's arrival, so that its received time was the
5056 same as the "first fail" time on the retry record, and that message
5057 remained on the queue past the ultimate address timeout, every queue runner
5058 would try a delivery (because it was past the ultimate address timeout) but
5059 after another failure, the ultimate address timeout, which should have then
5060 bounced the address, did not kick in. This was a "< instead of <=" error;
5061 in most cases the first failure would have been in the next clock tick
5062 after the received time, and all would be well.
5064 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
5065 being recognized when the domain list was tested by the match_domain
5066 condition in an expansion string.
5068 16. Added the ${str2b64: operator.
5070 17. Exim was always calling setrlimit() to set a large limit for the number of
5071 processes, without checking whether the existing limit was already
5072 adequate. (It did check for the limit on file descriptors.) Furthermore,
5073 errors from getrlimit() and setrlimit() were being ignored. Now they are
5074 logged to the main and panic logs, but Exim does carry on, to try to do its
5075 job under whatever limits there are.
5077 18. Imported PCRE 5.0.
5079 19. Trivial typo in log message " temporarily refused connection" (the leading
5082 20. If the log selector return_path_on_delivery was set and an address was
5083 redirected to /dev/null, the delivery process crashed because it assumed
5084 that a return path would always be set for a "successful" delivery. In this
5085 case, the whole delivery is bypassed as an optimization, and therefore no
5088 21. Internal re-arrangement: the function for sending a challenge and reading
5089 a response while authentication was assuming a zero-terminated challenge
5090 string. It's now changed to take a pointer and a length, to allow for
5091 binary data in such strings.
5093 22. Added the cyrus_sasl authenticator (code supplied by MBM).
5095 23. Exim was not respecting finduser_retries when seeking the login of the
5096 uid under which it was called; it was always trying 10 times. (The default
5097 setting of finduser_retries is zero.) Also, it was sleeping after the final
5098 failure, which is pointless.
5100 24. Implemented tls_on_connect_ports.
5102 25. Implemented acl_smtp_predata.
5104 26. If the domain in control=submission is set empty, Exim assumes that the
5105 authenticated id is a complete email address when it generates From: or
5106 Sender: header lines.
5108 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
5109 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
5110 chown and chgrp in /bin and hostname in /usr/bin.
5112 28. Exim was keeping the "process log" file open after each use, just as it
5113 does for the main log. This opens the possibility of it remaining open for
5114 long periods when the USR1 signal hits a daemon. Occasional processlog
5115 errors were reported, that could have been caused by this. Anyway, it seems
5116 much more sensible not to leave this file open at all, so that is what now
5119 29. The long-running daemon process does not normally write to the log once it
5120 has entered its main loop, and it closes the log before doing so. This is
5121 so that log files can straightforwardly be renamed and moved. However,
5122 there are a couple of unusual error situations where the daemon does write
5123 log entries, and I had neglected to close the log afterwards.
5125 30. The text of an SMTP error response that was received during a remote
5126 delivery was being truncated at 512 bytes. This is too short for some of
5127 the long messages that one sometimes sees. I've increased the limit to
5130 31. It is now possible to make retry rules that apply only when a message has a
5131 specific sender, in particular, an empty sender.
5133 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
5134 it possible to be selective about when SMTP synchronization is enforced.
5136 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
5138 32. Implemented hosts_connection_nolog.
5140 33. Added an ACL for QUIT.
5142 34. Setting "delay_warning=" to disable warnings was not working; it gave a
5145 35. Added mailbox_size and mailbox_filecount to appendfile.
5147 36. Added control = no_multiline_responses to ACLs.
5149 37. There was a bug in the logic of the code that waits for the clock to tick
5150 in the case where the clock went backwards by a substantial amount such
5151 that the microsecond fraction of "now" was more than the microsecond
5152 fraction of "then" (but the whole seconds number was less).
5154 38. Added support for the libradius Radius client library this is found on
5155 FreeBSD (previously only the radiusclient library was supported).
5161 1. When certain lookups returned multiple values in the form name=value, the
5162 quoting of the values was not always being done properly. Specifically:
5163 (a) If the value started with a double quote, but contained no whitespace,
5165 (b) If the value contained whitespace other than a space character (i.e.
5166 tabs or newlines or carriage returns) it was not quoted.
5167 This fix has been applied to the mysql and pgsql lookups by writing a
5168 separate quoting function and calling it from the lookup code. The fix
5169 should probably also be applied to nisplus, ibase and oracle lookups, but
5170 since I cannot test any of those, I have not disturbed their existing code.
5172 2. A hit in the callout cache for a specific address caused a log line with no
5173 reason for rejecting RCPT. Now it says "Previous (cached) callout
5174 verification failure".
5176 3. There was an off-by-one bug in the queryprogram router. An over-long
5177 return line was truncated at 256 instead of 255 characters, thereby
5178 overflowing its buffer with the terminating zero. As well as fixing this, I
5179 have increased the buffer size to 1024 (and made a note to document this).
5181 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
5182 when Exim is waiting for an SMTP response from a remote server, Exim
5183 restarts its select() call on the socket, thereby resetting its timeout.
5184 This is not a problem when such interrupts are rare. Somebody set up a cron
5185 job to run exiwhat every 2 minutes, which is less than the normal select()
5186 timeout (5 or 10 minutes). This meant that the select() timeout never
5187 kicked in because it was always reset. I have fixed this by comparing the
5188 time when an interrupt arrives with the time at the start of the first call
5189 to select(). If more time than the timeout has elapsed, the interrupt is
5190 treated as a timeout.
5192 5. Some internal re-factoring in preparation for the addition of Sieve
5193 extensions (by MH). In particular, the "personal" test is moved to a
5194 separate function, and given an option for scanning Cc: and Bcc: (which is
5195 not set for Exim filters).
5197 6. When Exim created an email address using the login of the caller as the
5198 local part (e.g. when creating a From: or Sender: header line), it was not
5199 quoting the local part when it contained special characters such as @.
5201 7. Installed new OpenBSD configuration files.
5203 8. Reworded some messages for syntax errors in "and" and "or" conditions to
5204 try to make them clearer.
5206 9. Callout options, other than the timeout value, were being ignored when
5207 verifying sender addresses in header lines. For example, when using
5209 verify = header_sender/callout=no_cache
5211 the cache was (incorrectly) being used.
5213 10. Added a missing instance of ${EXE} to the exim_install script; this affects
5214 only the Cygwin environment.
5216 11. When return_path_on_delivery was set as a log selector, if different remote
5217 addresses in the same message used different return paths and parallel
5218 remote delivery occurred, the wrong values would sometimes be logged.
5219 (Whenever a remote delivery process finished, the return path value from
5220 the most recently started remote delivery process was logged.)
5222 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
5223 lines when AUTH and/or TLS are in use. This is the "received protocol"
5224 field. Exim used to use "asmtp" for authenticated SMTP, without any
5225 indication (in the protocol name) for TLS use. Now it follows the RFC and
5226 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
5227 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
5228 names appear in log lines as well as in Received: header lines.
5230 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
5231 extensions, and comparison tests, and to fix some bugs.
5233 14. Changes to the "personal" filter test:
5235 (1) The test was buggy in that it was just doing the equivalent of
5236 "contains" tests on header lines. For example, if a user's address was
5237 anne@some.where, the "personal" test would incorrectly be true for
5239 To: susanne@some.where
5241 This test is now done by extracting each address from the header in turn,
5242 and checking the entire address. Other tests that are part of "personal"
5243 are now done using regular expressions (for example, to check local parts
5244 of addresses in From: header lines).
5246 (2) The list of non-personal local parts in From: addresses has been
5247 extended to include "listserv", "majordomo", "*-request", and "owner-*",
5248 taken from the Sieve specification recommendations.
5250 (3) If the message contains any header line starting with "List-" it is
5251 treated as non-personal.
5253 (4) The test for "circular" in the Subject: header line has been removed
5254 because it now seems ill-conceived.
5256 15. Minor typos in src/EDITME comments corrected.
5258 16. Installed latest exipick from John Jetmore.
5260 17. If headers_add on a router specified a text string that was too long for
5261 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
5262 of string_sprintf() is now avoided.
5264 18. $message_body_size was not set (it was always zero) when running the DATA
5265 ACL and the local_scan() function.
5267 19. For the "mail" command in an Exim filter, no default was being set for
5268 the once_repeat time, causing a random time value to be used if "once" was
5269 specified. (If the value happened to be <= 0, no repeat happened.) The
5270 default is now 0s, meaning "never repeat". The "vacation" command was OK
5271 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
5272 (I found it when inspecting the code).
5274 20. There is now an overall timeout for performing a callout verification. It
5275 defaults to 4 times the callout timeout, which applies to individual SMTP
5276 commands during the callout. The overall timeout applies when there is more
5277 than one host that can be tried. The timeout is checked before trying the
5278 next host. This prevents very long delays if there are a large number of
5279 hosts and all are timing out (e.g. when the network connections are timing
5280 out). The value of the overall timeout can be changed by specifying an
5281 additional sub-option for "callout", called "maxwait". For example:
5283 verify = sender/callout=5s,maxwait=20s
5285 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
5286 to the end before writing, but this should make it even safer).
5288 22. Exim was forgetting that it had advertised PIPELINING for the second and
5289 subsequent messages on an SMTP connection. It was also not resetting its
5290 memory on STARTTLS and an internal HELO.
5292 23. When Exim logs an SMTP synchronization error within a session, it now
5293 records whether PIPELINING has been advertised or not.
5295 24. Added 3 instances of "(long int)" casts to time_t variables that were being
5296 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
5297 rather than long int.
5299 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
5301 26. Added the never_mail option to autoreply.
5307 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
5308 crash if the getsockname() call failed; this can happen if a connection is
5309 closed very soon after it is established. The problem was simply in the
5310 order in which certain operations were done, causing Exim to try to write
5311 to the SMTP stream before it had set up the file descriptor. The bug has
5312 been fixed by making things happen in the correct order.
5318 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
5319 before the connection was closed, thus losing the rejection response.
5321 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
5322 some early Solaris releases, but causes trouble in current releases where
5323 socklen_t is defined.
5325 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
5328 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
5331 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
5332 that releasing the top of it at the end releases what was used for sub-
5333 expansions (unless the block got too big). However, discard this block if
5334 the first thing is a variable or header, so that we can use its block when
5335 it is dynamic (useful for very large $message_headers, for example).
5337 6. Lookups now cache *every* query, not just the most recent. A new, separate
5338 store pool is used for this. It can be recovered when all lookup caches are
5339 flushed. Lookups now release memory at the end of their result strings.
5340 This has involved some general refactoring of the lookup sources.
5342 7. Some code has been added to the store_xxx() functions to reduce the amount
5343 of flapping under certain conditions.
5345 8. log_incoming_interface used to affect only the <= reception log lines. Now
5346 it causes the local interface and port to be added to several more SMTP log
5347 lines, for example "SMTP connection from", and rejection lines.
5349 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
5351 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
5353 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
5354 could be overwritten at the end of the current message (or the start of a
5355 new message if it was set in a HELO ACL). The value is now preserved for
5356 the duration of the SMTP connection.
5358 12. If a transport had a headers_rewrite setting, and a matching header line
5359 contained an unqualified address, that address was qualified, even if it
5360 did not match any rewriting rules. The underlying bug was that the values
5361 of the flags that permit the existence of unqualified sender and recipient
5362 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
5363 non-local messages, and by -bnq for local messages) were not being
5364 preserved with the message after it was received.
5366 13. When Exim was logging an SMTP synchronization error, it could sometimes log
5367 "next input=" as part of the text comprising the host identity instead of
5368 the correct text. The code was using the same buffer for two different
5369 strings. However, depending on which order the printing function evaluated
5370 its arguments, the bug did not always show up. Under Linux, for example, my
5371 test suite worked just fine.
5373 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
5374 doesn't work with some older versions of Perl. It has been changed to "my",
5375 which in any case is probably the better facility to use.
5377 15. A really picky compiler found some instances of statements for creating
5378 error messages that either had too many or two few arguments for the format
5381 16. The size of the buffer for calls to the DNS resolver has been increased
5382 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
5383 for addresses that have a lot of PTR records. This alleviates a problem; it
5384 does not fully solve it.
5386 17. A dnsdb lookup for PTR records that receives more data than will fit in the
5387 buffer now truncates the list and logs the incident, which is the same
5388 action as happens when Exim is looking up a host name and its aliases.
5389 Previously in this situation something unpredictable would happen;
5390 sometimes it was "internal error: store_reset failed".
5392 18. If a server dropped the connection unexpectedly when an Exim client was
5393 using GnuTLS and trying to read a response, the client delivery process
5394 crashed while trying to generate an error log message.
5396 19. If a "warn" verb in an ACL added multiple headers to a message in a single
5397 string, for example:
5399 warn message = H1: something\nH2: something
5401 the text was added as a single header line from Exim's point of view
5402 though it ended up OK in the delivered message. However, searching for the
5403 second and subsequent header lines using $h_h2: did not work. This has been
5404 fixed. Similarly, if a system filter added multiple headers in this way,
5405 the routers could not see them.
5407 20. Expanded the error message when iplsearch is called with an invalid key to
5408 suggest using net-iplsearch in a host list.
5410 21. When running tests using -bh, any delays imposed by "delay" modifiers in
5411 ACLs are no longer actually imposed (and a message to that effect is
5414 22. If a "gecos" field in a passwd entry contained escaped characters, in
5415 particular, if it contained a \" sequence, Exim got it wrong when building
5416 a From: or a Sender: header from that name. A second bug also caused
5417 incorrect handling when an unquoted " was present following a character
5418 that needed quoting.
5420 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
5421 was not being matched caselessly.
5423 24. Arranged for all hyphens in the exim.8 source to be escaped with
5426 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
5427 properly. Recipient callout cache records were still being keyed to include
5428 the sender, even when use_sender was set false. This led to far more
5429 callouts that were necessary. The sender is no longer included in the key
5430 when use_sender is false.
5432 26. Added "control = submission" modifier to ACLs.
5434 27. Added the ${base62d: operator to decode base 62 numbers.
5436 28. dnsdb lookups can now access SRV records.
5438 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
5439 the configuration file.
5441 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
5442 (-v) mode. This makes the output for a verbose queue run more intelligible.
5444 31. Added a use_postmaster feature to recipient callouts.
5446 32. Added the $body_zerocount variable, containing the number of binary zero
5447 bytes in the message body.
5449 33. The time of last modification of the "new" subdirectory is now used as the
5450 "mailbox time last read" when there is a quota error for a maildir
5453 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
5455 35. Added +ignore_unknown as a special item in host lists.
5457 36. Code for decoding IPv6 addresses in host lists is now included, even if
5458 IPv6 support is not being compiled. This fixes a bug in which an IPv6
5459 address was recognized as an IP address, but was then not correctly decoded
5460 into binary, causing unexpected and incorrect effects when compared with
5467 1. Very minor rewording of debugging text in manualroute to say "list of
5468 hosts" instead of "hostlist".
5470 2. If verify=header_syntax was set, and a header line with an unqualified
5471 address (no domain) and a large number of spaces between the end of the
5472 name and the colon was received, the reception process suffered a buffer
5473 overflow, and (when I tested it) crashed. This was caused by some obsolete
5474 code that should have been removed. The fix is to remove it!
5476 3. When running in the test harness, delay a bit after writing a bounce
5477 message to get a bit more predictability in the log output.
5479 4. Added a call to search_tidyup() just before forking a reception process. In
5480 theory, someone could use a lookup in the expansion of smtp_accept_max_
5481 per_host which, without the tidyup, could leave open a database connection.
5483 5. Added the variables $recipient_data and $sender_data which get set from a
5484 lookup success in an ACL "recipients" or "senders" condition, or a router
5485 "senders" option, similar to $domain_data and $local_part_data.
5487 6. Moved the writing of debug_print from before to after the "senders" test
5490 7. Change 4.31/66 (moving the time when the Received: is generated) caused
5491 problems for message scanning, either using a data ACL, or using
5492 local_scan() because the Received: header was not generated till after they
5493 were called (in order to set the time as the time of reception completion).
5494 I have revised the way this works. The header is now generated after the
5495 body is received, but before the ACL or local_scan() are called. After they
5496 are run, the timestamp in the header is updated.
5502 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
5503 before starting a queue runner without re-exec. This happened only when
5504 deliver_drop_privilege was set or when the Exim user was set to root. The
5505 effect of the bug was that timeouts during subsequent deliveries caused
5506 crashes instead of being properly handled. The handler is now left at its
5507 default (and expected) setting.
5509 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
5510 message, again when deliver_drop_privilege is set or Exim is run as root.
5511 The bug described in (1) was not present in this case, but the tidying up
5512 of the other signals was missing. I have made the two cases consistent.
5514 3. The ignore_target_hosts setting on a manualroute router was being ignored
5515 for hosts that were looked up using the /MX notation.
5517 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
5520 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
5521 operated on the sender address. After changing the $sender_address to <>
5522 for the sender address verify, Exim was re-instated it as the original
5523 (before rewriting) address, but remembering that it had rewritten it, so it
5524 wasn't rewriting it again. This bug also had the effect of breaking the
5525 sender address verification caching when the sender address was rewritten.
5527 6. The ignore_target_hosts option was being ignored by the ipliteral router.
5528 This has been changed so that if the ip literal address matches
5529 ignore_target_hosts, the router declines.
5531 7. Added expansion conditions match_domain, match_address, and match_local_
5532 part (NOT match_host).
5534 8. The placeholder for the Received: header didn't have a length field set.
5536 9. Added code to Exim itself and to exim_lock to test for a specific race
5537 condition that could lead to file corruption when using MBX delivery. The
5538 issue is with the lockfile that is created in /tmp. If this file is removed
5539 after a process has opened it but before that process has acquired a lock,
5540 there is the potential for a second process to recreate the file and also
5541 acquire a lock. This could lead to two Exim processes writing to the file
5542 at the same time. The added code performs the same test as UW imapd; it
5543 checks after acquiring the lock that its file descriptor still refers to
5544 the same named file.
5546 10. The buffer for building added header lines was of fixed size, 8192 bytes.
5547 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
5550 11. Added the smtp_active_hostname option. If used, this will typically be made
5551 to depend on the incoming interface address. Because $interface_address is
5552 not set up until the daemon has forked a reception process, error responses
5553 that can happen earlier (such as "too many connections") no longer contain
5556 12. If an expansion in a condition on a "warn" statement fails because a lookup
5557 defers, the "warn" statement is abandoned, and the next ACL statement is
5558 processed. Previously this caused the whole ACL to be aborted.
5560 13. Added the iplsearch lookup type.
5562 14. Added ident_timeout as a log selector.
5564 15. Added tls_certificate_verified as a log selector.
5566 16. Added a global option tls_require_ciphers (compare the smtp transport
5567 option of the same name). This controls incoming TLS connections.
5569 17. I finally figured out how to make tls_require_ciphers do a similar thing
5570 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
5571 before starting the TLS session.
5573 18. Tabs are now shown as \t in -bP output.
5575 19. If the log selector return_path_on_delivery was set, Exim crashed when
5576 bouncing a message because it had too many Received: header lines.
5578 20. If two routers both had headers_remove settings, and the first one included
5579 a superfluous trailing colon, the final name in the first list and the
5580 first name in the second list were incorrectly joined into one item (with a
5581 colon in the middle).
5587 1. Added -C and -D options to the exinext utility, mainly to make it easier
5588 to include in the automated testing, but these could be helpful when
5589 multiple configurations are in use.
5591 2. The exinext utility was not formatting the output nicely when there was
5592 an alternate port involved in the retry record key, nor when there was a
5593 message id as well (for retries that were specific to a specific message
5594 and a specific host). It was also confused by IPv6 addresses, because of
5595 the additional colons they contain. I have fixed the IPv4 problem, and
5596 patched it up to do a reasonable job for IPv6.
5598 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
5599 delivery, the log line now contains "pipelined" if PIPELINING was used.
5601 4. An SMTP transport process used to panic and die if the bind() call to set
5602 an explicit outgoing interface failed. This has been changed; it is now
5603 treated in the same way as a connect() failure.
5605 5. A reference to $sender_host_name in the part of a conditional expansion
5606 that was being skipped was still causing a DNS lookup. This no longer
5609 6. The def: expansion condition was not recognizing references to header lines
5610 that used bh_ and bheader_.
5612 7. Added the _cache feature to named lists.
5614 8. The code for checking quota_filecount in the appendfile transport was
5615 allowing one more file than it should have been.
5617 9. For compatibility with Sendmail, the command line option
5625 and sets the incoming protocol and host name (for trusted callers). The
5626 host name and its colon can be omitted when only the protocol is to be set.
5627 Note the Exim already has two private options, -pd and -ps, that refer to
5628 embedded Perl. It is therefore impossible to set a protocol value of "d" or
5629 "s", but I don't think that's a major issue.
5631 10. A number of refactoring changes to the code, none of which should affect
5634 (a) The number of logging options was getting close to filling up the
5635 32-bit word that was used as a bit map. I have split them into two classes:
5636 those that are passed in the argument to log_write(), and those that are
5637 only ever tested independently outside of that function. These are now in
5638 separate 32-bit words, so there is plenty of room for expansion again.
5639 There is no change in the user interface or the logging behaviour.
5641 (b) When building, for example, log lines, the code previously used a
5642 macro that called string_cat() twice, in order to add two strings. This is
5643 not really sufficiently general. Furthermore, there was one instance where
5644 it was actually wrong because one of the argument was used twice, and in
5645 one call a function was used. (As it happened, calling the function twice
5646 did not affect the overall behaviour.) The macro has been replaced by a
5647 function that can join an arbitrary number of extra strings onto a growing
5650 (c) The code for expansion conditions now uses a table and a binary chop
5651 instead of a serial search (which was left over from when there were very
5652 few conditions). Also, it now recognizes conditions like "pam" even when
5653 the relevant support is not compiled in: a suitably worded error message is
5654 given if an attempt is made to use such a condition.
5656 11. Added ${time_interval:xxxxx}.
5658 12. A bug was causing one of the ddress fields not to be passed back correctly
5659 from remote delivery subprocesses. The field in question was not being
5660 subsequently used, so this caused to problems in practice.
5662 13. Added new log selectors queue_time and deliver_time.
5664 14. Might have fixed a bug in maildirsizefile handling that threw up
5665 "unexpected character" debug warnings, and recalculated the data
5666 unnecessarily. In any case, I expanded the warning message to give more
5669 15. Added the message "Restricted characters in address" to the statements in
5670 the default ACL that block characters like @ and % in local parts.
5672 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
5673 Three changes have been made:
5675 (a) There was a serious bug; a negative response to MAIL caused the whole
5676 recipient domain to be cached as invalid, thereby blocking all messages
5677 to all local parts at the same domain, from all senders. This bug has
5678 been fixed. The domain is no longer cached after a negative response to
5679 MAIL if the sender used is not empty.
5681 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
5684 (c) A new callout option, "use_sender" has been added for people who want
5685 the modified behaviour.
5691 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
5694 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
5695 indeed breaks things for older releases.
5697 3. Added additional logging to the case where there is a problem reading data
5698 from a filter that is running in a subprocess using a pipe, in order to
5699 try to track down a specific problem.
5701 4. Testing facility fudge: when running in the test harness and attempting
5702 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
5703 getting "No route to host". Convert this to a timeout.
5705 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
5708 6. Some OS don't have socklen_t but use size_t instead. This affects the
5709 fifth argument of getsockopt() amongst other things. This is now
5710 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
5711 can be set for individual OS. I have set it for SunOS5, OSF1, and
5712 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
5713 some earlier ones do not.
5715 7. Change 4.30/15 was not doing the test caselessly.
5717 8. The standard form for an IPv6 address literal was being rejected by address
5718 parsing in, for example, MAIL and RCPT commands. An example of this kind of
5719 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5720 this, as well as the form without the "IPv6" on the front (but only when
5721 address literals are enabled, of course).
5723 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
5725 10. Exim crashed if a message with an empty sender address specified by -f
5726 encountered a router with an errors_to setting. This could be provoked only
5727 by a command such as
5731 where an empty string was supplied; "<>" did not hit this bug.
5733 11. Installed PCRE release 4.5.
5735 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
5736 remained set. It is now erased.
5738 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
5739 times from message ids (which are base 36 rather than the normal 62).
5741 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
5742 were being counted as actual protocol errors, and logged if the log
5743 selector +smtp_protocol_error was set. One cannot be perfect in this test,
5744 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
5745 and DATA following a set of rejected RCPTs do not count as protocol errors.
5746 In other words, Exim assumes they were pipelined, though this may not
5747 actually be the case. Of course, in all cases the client gets an
5748 appropriate error code.
5750 15. If a lookup fails in an ACL condition, a message about the failure may
5751 be available; it is used if testing the ACL cannot continue, because most
5752 such messages specify what the cause of the deferral is. However, some
5753 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
5754 that caused an old message to be retained and used if a later statement
5755 caused a defer, replacing the real cause of the deferral.
5757 16. If an IP address had so many PTR records that the DNS lookup buffer
5758 was not large enough to hold them, Exim could crash while trying to process
5759 the truncated data. It now detects and logs this case.
5761 17. Further to 4.21/58, another change has been made: if (and only if) the
5762 first line of a message (the first header line) ends with CRLF, a bare LF
5763 in a subsequent header line has a space inserted after it, so as not to
5764 terminate the header.
5766 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
5767 unnecessarily, used atoi() instead of strtol(), and didn't check the
5768 termination when getting file sizes from file names by regex.
5770 19. Completely re-implemented the support for maildirsize files, in the light
5771 of a number of problems with the previous contributed implementation
5772 (4.30/29). In particular:
5774 . If the quota is zero, the maildirsize file is maintained, but no quota is
5777 . If the maildir directory does not exist, it is created before any attempt
5778 to write a maildirsize file.
5780 . The quota value in the file is just a cache; if the quota is changed in
5781 the transport, the new value overrides.
5783 . A regular expression is available for excluding directories from the
5786 20. The autoreply transport checks the characters in options that define the
5787 message's headers; it allows continued headers, but it was checking with
5788 isspace() after an embedded newline instead of explicitly looking for a
5791 21. If all the "regular" hosts to which an address was routed had passed their
5792 expiry times, and had not reached their retry times, the address was
5793 bounced, even if fallback hosts were defined. Now Exim should go on to try
5796 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
5797 equivalent code in the SMTP transport. Some hosts send humungous responses
5798 to HELO/EHLO, more than 1024 it seems.
5800 23. Refactoring: code in filter.c used (void *) for "any old type" but this
5801 gives compiler warnings in some environments. I've now done it "properly",
5804 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
5805 (because of problems with the built-in one) was declared to return uschar *
5806 instead of char *, causing compiler failure.
5808 25. Fixed a file descriptor leak when processing alias/forward files.
5810 26. Fixed a minor format string issue in dbfn.c.
5812 27. Typo in exim.c: ("dmbnz" for "dbmnz").
5814 28. If a filter file refered to $h_xxx or $message_headers, and the headers
5815 contained RFC 2047 "words", Exim's memory could, under certain conditions,
5818 29. When a sender address is verified, it is cached, to save repeating the test
5819 when there is more than one recipient in a message. However, when the
5820 verification involves a callout, it is possible for different callout
5821 options to be set for different recipients. It is too complicated to keep
5822 track of this in the cache, so now Exim always runs a verification when a
5823 callout is required, relying on the callout cache for the optimization.
5824 The overhead is duplication of the address routing, but this should not be
5827 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
5828 to be verified with callout=postmaster, and the main callout worked but the
5829 postmaster check failed, the verification correctly failed. However, if a
5830 subsequent RCPT command asked for sender verification *without* the
5831 postmaster check, incorrect caching caused this verification also to fail,
5834 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
5835 it was not caching the DNS options (qualify_single, search_parents) that
5836 were used when the lookup failed. A subsequent lookup with different
5837 options therefore always gave the same answer, though there were cases
5838 where it should not have. (Example: a "domains = !$mx_any" option on a
5839 dnslookup router: the "domains" option is always processed without any
5840 widening, but the router might have qualify_single set.) Now Exim uses the
5841 cached value only when the same options are set.
5843 32. Added John Jetmore's "exipick" utility to the distribution.
5845 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
5846 than a timeout (e.g. a certificate is required, and is not provided), an
5847 Exim server now closes the connection immediately. Previously it waited for
5848 the client to close - but if the client is SSL, it seems that they each
5849 wait for each other, leading to a delay before one of them times out.
5851 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
5852 maintained 0.8.x compatibility because I don't think many are using it, and
5853 it is clearly obsolete.
5855 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
5858 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
5859 client certificate was expired. A simple patch fixes this, though I don't
5860 understand the full logic of why the verify callback is called multiple
5863 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
5864 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
5865 which causes problems with some clients (such as the Certicom SSL Plus
5866 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
5867 disables the coutermeasure allowing Eudora to connect."
5869 38. Exim was not checking that a write() to a log file succeeded. This could
5870 lead to Bad Things if a log got too big, in particular if it hit a file
5871 size limit. Exim now panics and dies if it cannot write to a log file, just
5872 as it does if it cannot open a log file.
5874 39. Modified OS/Makefile-Linux so that it now contains
5876 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
5878 The two -D definitions ensure that Exim is compiled with large file
5879 support, which makes it possible to handle log files that are bigger than
5882 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
5883 instance) a domain was checked against a named list that involved a lookup,
5884 causing $domain_data to be set, then another domain was checked against the
5885 same list, then the first domain was re-checked, the value of $domain_data
5886 after the final check could be wrong. In particular, if the second check
5887 failed, it could be set empty. This bug probably also applied to
5890 41. The strip_trailing_dot option was not being applied to the address given
5891 with the -f command-line option.
5893 42. The code for reading a message's header from the spool was incrementing
5894 $received_count, but never initializing it. This meant that the value was
5895 incorrect (doubled) while delivering a message in the same process in which
5896 it was received. In the most common configuration of Exim, this never
5897 happens - a fresh exec is done - but it can happen when
5898 deliver_drop_privilege is set.
5900 43. When Exim logs an SMTP synchronization error - client data sent too soon -
5901 it now includes up to 150 characters of the unexpected data in the log
5904 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
5905 and building data strings. The size of both of these buffers was 10 000
5906 bytes - far larger than anybody would *ever* want, thought I. Needless to
5907 say, somebody hit the limit. I have increased the maximum line length to
5908 20 000 and the maximum data length of concatenated lines to 100 000. I have
5909 also fixed two bugs, because there was no checking on these buffers. Tsk,
5910 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
5911 buffer is too small.
5913 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
5914 lsearch lookups. Now it does.
5916 46. When parsing a route_list item in a manualroute router, a fixed-length
5917 buffer was used for the list of hosts. I made this 1024 bytes long,
5918 thinking that nobody would ever have a list of hosts that long. Wrong.
5919 Somebody had a whole pile of complicated expansion conditions, and the
5920 string was silently truncated, leading to an expansion error. It turns out
5921 that it is easier to change to an unlimited length (owing to other changes
5922 that have happened since this code was originally written) than to build
5923 structure for giving a limitation error. The length of the item that
5924 expands into the list of hosts is now unlimited.
5926 47. The lsearch lookup could not handle data where the length of text line was
5927 more than 4095 characters. Such lines were truncated, leading to shortened
5928 data being returned. It should now handle lines of any length.
5930 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
5931 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
5934 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
5935 replacing strings such as BIN_DIRECTORY in a source file: the replacement
5936 no longer happens in comment lines. A list of replacements is now placed
5937 at the head of all of the source files, except those whose only change is
5938 to replace PERL_COMMAND in the very first #! line.
5940 50. Replaced the slow insertion sort in queue.c, for sorting the list of
5941 messages on the queue, with a bottom-up merge sort, using code contributed
5942 by Michael Haardt. This should make operations like -bp somewhat faster on
5943 large queues. It won't affect queue runners, except when queue_run_in_order
5946 51. Installed eximstats 1.31 in the distribution.
5948 52. Added support for SRV lookups to the dnslookup router.
5950 53. If an ACL referred to $message_body or $message_body_end, the value was not
5951 reset for any messages that followed in the same SMTP session.
5953 54. The store-handling optimization for building very long strings was not
5954 differentiating between the different store pools. I don't think this
5955 actually made any difference in practice, but I've tidied it.
5957 55. While running the routers to verify a sender address, $sender_address
5958 was still set to the sender address. This is wrong, because when routing to
5959 send a bounce to the sender, it would be empty. Therefore, I have changed
5960 it so that, while verifying a sender address, $sender_address is set to <>.
5961 (There is no change to what happens when verifying a recipient address.)
5963 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
5964 target A or AAAA records (if not already returned) without resetting the
5965 qualify_single or search_parents options of the DNS resolver. These are
5966 inappropriate in this case because the targets of MX and SRV records must
5967 be FQDNs. A broken DNS record could cause trouble if it happened to have a
5968 target that, when qualified, matched something in the local domain. These
5969 two options are now turned off when doing these lookups.
5971 57. It seems that at least some releases of Reiserfs (which does not have the
5972 concept of a fixed number of inodes) returns zero and not -1 for the
5973 number of available inodes. This interacted badly with check_spool_inodes,
5974 which assumed that -1 was the "no such thing" setting. What I have done is
5975 to check that the total number of inodes is greater than zero before doing
5976 the test of how many are available.
5978 58. When a "warn" ACL statement has a log_message modifier, the message is
5979 remembered, and not repeated. This is to avoid a lot of repetition when a
5980 message has many recipients that cause the same warning to be written.
5981 However, Exim was preserving the list of already written lines for an
5982 entire SMTP session, which doesn't seem right. The memory is now reset if a
5983 new message is started.
5985 59. The "rewrite" debugging flag was not showing the result of rewriting in the
5986 debugging output unless log_rewrite was also set.
5988 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
5989 of (int)(handle) when we know that handle contains (void *)(-1).
5991 61. The Exim daemon panic-logs an error return when it closes the incoming
5992 connection. However "connection reset by peer" seems to be common, and
5993 isn't really an error worthy of noting specially, so that particular error
5996 62. When Exim is trying to find all the local interfaces, it used to panic and
5997 die if the ioctl to get the interface flags failed. However, it seems that
5998 on at least one OS (Solaris 9) it is possible to have an interface that is
5999 included in the list of interfaces, but for which you get a failure error
6000 for this call. This happens when the interface is not "plumbed" into a
6001 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
6002 failure of the "get flags" call assumes that the interface is down.
6004 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
6005 makes life easier for people who are doing arithmetic on fields extracted
6006 from dates, where you often get leading zeros that should not be
6007 interpreted as octal.
6009 64. Added qualify_domain to the redirect router, to override the global
6012 65. If a pathologically long header line contained very many addresses (the
6013 report of this problem mentioned 10 000) and each of them was rewritten,
6014 Exim could use up a very large amount of memory. (It kept on making new
6015 copies of the header line as it rewrote, and never released the old ones.)
6016 At the expense of a bit more processing, the header rewriting function has
6017 been changed so that it no longer eats memory in this way.
6019 66. The generation of the Received: header has been moved from the time that a
6020 message starts to be received, to the time that it finishes. The timestamp
6021 in the Received: header should now be very close to that of the <= log
6022 line. There are two side-effects of this change:
6024 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
6025 logged header lines no longer include the local Received: line, because
6026 it has not yet been created. The same applies to a copy of the message
6027 that is returned to a non-SMTP sender when a message is rejected.
6029 (b) When a filter file is tested using -bf, no additional Received: header
6030 is added to the test message. After some thought, I decided that this
6033 This change does not affect the value of $received_for. It is still set
6034 after address rewriting, but before local_scan() is called.
6036 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
6038 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
6039 gave an unhelpful panic error message, and a defer error. I have managed to
6040 change this behaviour so that it now rejects any supplied certificate,
6041 which seems right, as the list of acceptable certificates is empty.
6043 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
6044 gave an unhelpful defer error. I have not managed to make this reject any
6045 supplied certificates, but the error message it gives is "no certificate
6046 supplied", which is not helpful.
6048 70. exigrep's output now also includes lines that are not associated with any
6049 message, but which match the given pattern. Implemented by a patch from
6050 Martin Sluka, which also tidied up the Perl a bit.
6052 71. Recipient callout verification, like sender verification, was using <> in
6053 the MAIL FROM command. This isn't really the right thing, since the actual
6054 sender may affect whether the remote host accepts the recipient or not. I
6055 have changed it to use the actual sender in the callout; this means that
6056 the cache record is now keyed on a recipient/sender pair, not just the
6057 recipient address. There doesn't seem to be a real danger of callout loops,
6058 since a callout by the remote host to check the sender would use <>.
6059 [SEE ABOVE: changed after hitting problems.]
6061 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
6062 temporary errors. However, in the case of such a code being given after
6063 the end of a data transmission (i.e. after ".") Exim was failing to write
6064 a retry record for the message. (Yes, there was some broken host that was
6065 actually sending 8xx at this point.)
6067 73. An unknown lookup type in a host list could cause Exim to panic-die when
6068 the list was checked. (An example that provoked this was putting <; in the
6069 middle of a list instead of at the start.) If this happened during a DATA
6070 ACL check, a -D file could be left lying around. This kind of configuration
6071 error no longer causes Exim to die; instead it causes a defer error. The
6072 incident is still logged to the main and panic logs.
6074 74. Buglet left over from Exim 3 conversion. The message "too many messages
6075 in one connection" was written to the rejectlog but not the mainlog, except
6076 when address rewriting (yes!) was being logged.
6078 75. Added write_rejectlog option.
6080 76. When a system filter was run not as root (that is, when system_filter_user
6081 was set), the values of the $n variables were not being returned to the
6082 main process; thus, they were not subsequently available in the $sn
6085 77. Added +return_path_on_delivery log selector.
6087 78. A connection timeout was being treated differently from recipients deferred
6088 when testing hosts_max_try with a message that was older than the host's
6089 retry timeout. (The host should not be counted, thus allowing all hosts to
6090 be tried at least once before bouncing.) This may have been the cause of an
6091 occasionally reported bug whereby a message would remain on the queue
6092 longer than the retry timeout, but would be bounced if a delivery was
6093 forced. I say "may" because I never totally pinned down the problem;
6094 setting up timeout/retry tests is difficult. See also the next item.
6096 79. The ultimate address timeout was not being applied to errors that involved
6097 a combination of host plus message (for example, a timeout on a MAIL
6098 command). When an address resolved to a number of possible hosts, and they
6099 were not all tried for each delivery (e.g. because of hosts_max_try), a
6100 message could remain on the queue longer than the retry timeout.
6102 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
6105 81. Fixed an obscure SMTP outgoing bug which required at least the following
6106 conditions: (a) there was another message waiting for the same server;
6107 (b) the server returned 5xx to all RCPT commands in the first message so
6108 that the message was not completed; (c) the server dropped the connection
6109 or gave a negative response to the RSET that Exim sends to abort the
6110 transaction. The observed case was a dropped connection after DATA that had
6111 been sent in pipelining mode. That is, the server had advertised PIPELINING
6112 but was not implementing it correctly. The effect of the bug was incorrect
6113 behaviour, such as trying another host, and this could lead to a crash.
6119 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
6120 and daemon.c were passed as pointers to ints; they should have been
6121 pointers to socklen_t variables (which are typically unsigned ints).
6123 2. Some signed/unsigned type warnings in the os.c file for Linux have been
6126 3. Fixed a really odd bug that affected only the testing scheme; patching a
6127 certain fixed string in the binary changed the value of another string that
6128 happened to be identical to the end of the original first string.
6130 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
6131 name", it returns that address as the IP address. On some operating
6132 systems (e.g. Solaris), it also passes back the IP address string as the
6133 "host name". However, on others (e.g. Linux), it passes back an empty
6134 string. Exim wasn't checking for this, and was changing the host name to an
6135 empty string, assuming it had been canonicalized.
6137 5. Although rare, it is permitted to have more than one PTR record for a given
6138 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
6139 all the names associated with an address, because they do in Solaris.
6140 However, it seems that they do not in Linux for data that comes from the
6141 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
6142 I found this out when I moved to a new Linux workstation and tried to run
6143 the Exim test suite.
6145 To get round this problem I have changed the code so that it now does its
6146 own call to the DNS to look up PTR records when searching for a host name.
6147 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
6148 addresses that are only in /etc/hosts are still found.
6150 This behaviour is, however, controlled by an option called host_lookup_
6151 order, which defaults to "bydns:byaddr". If people want to use the other
6152 order, or indeed, just use one or the other means of lookup, they can
6153 specify it in this variable.
6155 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
6156 some operating systems, this comes back from gethostbyaddr() as an empty
6157 string, and this is what Exim used to test for. However, it seems that in
6158 other systems, "." is yielded. Exim now tests for this case too.
6160 7. The values of check_spool_space and check_log_space are now held internally
6161 as a number of kilobytes instead of an absolute number of bytes. If a
6162 numbers is specified without 'K' or 'M', it is rounded up to the nearest
6163 kilobyte. This means that much larger values can be stored.
6165 8. Exim monitor: an attempt to get the action menu when not actually pointing
6166 at a message produces an empty menu entitled "No message selected". This
6167 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
6168 no entries in it ("Shell widget menu has zero width and/or height"). So I
6169 have added a single, blank menu entry in this case.
6171 9. Added ${quote_local_part.
6173 10. MIME decoding is now applied to the contents of Subject: header lines when
6176 11. Now that a reference to $sender_host_address automatically causes a reverse
6177 lookup to occur if necessary (4.13/18), there is no need to arrange for a
6178 host lookup before query-style lookups in lists that might use this
6179 variable. This has therefore been abolished, and the "net-" prefix is no
6180 longer necessary for query-style lookups.
6182 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
6183 have been a typo for LFLAGS, so it has been changed.
6185 13. The install script calls Exim with "-C /dev/null" in order to find the
6186 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
6187 to be output. However, since Exim outputs its version number before the
6188 error, it didn't break the script. It just looked ugly. I fixed this by
6189 always allowing "-C /dev/null" if the caller is root.
6191 14. Ignore overlarge ACL variable number when reading spool file - insurance
6192 against a later release with more variables having written the file.
6194 15. The standard form for an IPv6 address literal was being rejected by EHLO.
6195 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
6196 this, as well as the form without the "IPv6" on the front.
6198 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
6199 OS/Makefile-Darwin file.
6201 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
6202 only to LDAP libraries that do not have LDAP_OPT_DEREF.
6204 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
6205 variable. However, some OS use ints for inodes. I've added cast to long int
6206 to get rid of the compiler warning.
6208 19. I had forgotten to lock out "/../" in configuration file names when
6209 ALT_CONFIG_PREFIX was set.
6211 20. Routers used for verification do not need to specify transports. However,
6212 if such a router generated a host list, and callout was configured, Exim
6213 crashed, because it could not find a port number from the (non-existent)
6214 transport. It now assumes port 25 in this circumstance.
6216 21. Added the -t option to exigrep.
6218 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
6219 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
6220 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
6221 them is set, LOOKUP_LSEARCH is forced.
6223 23. "exim -bV" now outputs a list of lookups that are included in the binary.
6225 24. Added sender and host information to the "rejected by local_scan()" log
6226 line; previously there was no indication of these.
6228 25. Added .include_if_exists.
6230 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
6231 Linux. It turns out that not all file systems support this. Apparently some
6232 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
6233 do it.) To cope with this, the error EINVAL, which means that sync-ing is
6234 not supported on the file descriptor, is now ignored when Exim is trying to
6235 sync a directory. This applies only to Linux.
6237 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
6239 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
6240 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
6243 29. Added support for maildirsize files from supplied patch (modified a bit).
6245 30. The use of :fail: followed by an empty string could lead Exim to respond to
6246 sender verification failures with (e.g.):
6248 550 Verification failed for <xxx>
6249 550 Sender verify failed
6251 where the first response line was missing the '-' that indicates it is not
6252 the final line of the response.
6254 31. The loop for finding the name of the user that called Exim had a hardwired
6255 limit of 10; it now uses the value of finduser_retries, which is used for
6256 all other user lookups.
6258 32. Added $received_count variable, available in data and not_smtp ACLs, and at
6261 33. Exim was neglecting to zero errno before one call of strtol() when
6262 expanding a string and expecting an integer value. On some systems this
6263 resulted in spurious "integer overflow" errors. Also, it was casting the
6264 result into an int without checking.
6266 34. Testing for a connection timeout using "timeout_connect" in the retry rules
6267 did not work. The code looks as if it has *never* worked, though it appears
6268 to have been documented since at least release 1.62. I have made it work.
6270 35. The "timeout_DNS" error in retry rules, also documented since at least
6271 1.62, also never worked. As it isn't clear exactly what this means, and
6272 clearly it isn't a major issue, I have abolished the feature by treating it
6273 as "timeout", and writing a warning to the main and panic logs.
6275 36. The display of retry rules for -brt wasn't always showing the error code
6278 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
6279 timeout_connect_A, timeout_connect_MX.
6281 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
6282 to the empty sender.
6284 39. The daemon was not analysing the content of -oX till after it had closed
6285 stderr and disconnected from the controlling terminal. This meant that any
6286 syntax errors were only noted on the panic log, and the return code from
6287 the command was 0. By re-arranging the code a little, I've made the
6288 decoding happen first, so such errors now appear on stderr, and the return
6289 code is 1. However, the actual setting up of the sockets still happens in
6290 the disconnected process, so errors there are still only recorded on the
6293 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
6294 connections (as happens on some IP stacks) was logged at start up time as
6295 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
6296 it from "IPv6 and IPv4", which means that two separate sockets are being
6299 41. The debug output for gethostbyname2() or getipnodebyname() failures now
6300 says whether AF_INET or AF_INET6 was passed as an argument.
6302 42. Exiwhat output was messed up when time zones were included in log
6305 43. Exiwhat now gives more information about the daemon's listening ports,
6306 and whether -tls-on-connect was used.
6308 44. The "port" option of the smtp transport is now expanded.
6310 45. A "message" modifier in a "warn" statement in a non-message ACL was being
6311 silently ignored. Now an error message is written to the main and panic
6314 46. There's a new ACL modifier called "logwrite" which writes to a log file
6315 as soon as it is encountered.
6317 47. Added $local_user_uid and $local_user_gid at routing time.
6319 48. Exim crashed when trying to verify a sender address that was being
6322 49. Exim was recognizing only a space character after ".include". It now also
6323 recognizes a tab character.
6325 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
6326 extracting the relevant information from the specification. The man page no
6327 longer contains scrambled data for the -d option, and I've added a section
6328 at the front about calling Exim under different names.
6330 51. Added "extra_headers" argument to the "mail" command in filter files.
6332 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
6335 53. Installed eximstats 1.29.
6337 54. Added transport_filter_timeout as a generic transport option.
6339 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
6340 Cc: header lines. This was required by RFC 822, but it not required by RFC
6343 56. Exim used to add From:, Date:, and Message-Id: header lines to any
6344 incoming messages that did not have them. Now it does so only if the
6345 message originates locally, that is, if there is no associated remote host
6346 address. When Resent- header lines are present, this applies to the Resent-
6347 lines rather than the non-Resent- lines.
6349 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
6350 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
6352 58. Messages for configuration errors now include the name of the main
6353 configuration file - useful now that there may be more than one file in a
6354 list (.included file names were always shown).
6356 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
6357 for those rare installations that do not start the daemon as root or run it
6358 setuid root. I've cut out the call to initgroups() if the daemon is not
6361 60. The Exim user and group can now be bound into the binary as text strings
6362 that are looked up at the start of Exim's processing.
6364 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
6366 62. Added $mailstore_basename variable.
6368 63. Installed patch to sieve.c from Michael Haardt.
6370 64. When Exim failed to open the panic log after failing to open the main log,
6371 the original message it was trying to log was written to stderr and debug
6372 output, but if they were not available (the usual case in production), it
6373 was lost. Now it is written to syslog before the two lines that record the
6374 failures to open the logs.
6376 65. Users' Exim filters run in subprocesses under the user's uid. It is
6377 possible for a "deliver" command or an alias in a "personal" command to
6378 provoke an address rewrite. If logging of address rewriting is configured,
6379 this fails because the process is not running as root or exim. There may be
6380 a better way of dealing with this, but for the moment (because 4.30 needs
6381 to be released), I have disabled address rewrite logging when running a
6382 filter in a non-root, non-exim process.
6388 1. The buildconfig auxiliary program wasn't quoting the value set for
6389 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
6390 not defined. This bug was masked in 4.22 by the effect that was fixed in
6393 2. Some messages that were rejected after a message id was allocated were
6394 shown as "incomplete" by exigrep. It no longer does this for messages that
6395 are rejected by local_scan() or the DATA or non-SMTP ACLs.
6397 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
6398 have allow_domain_literals set, the ID did not get logged in the <= line.
6399 Domain literals are now always recognized in Message-ID: header lines.
6401 4. The first argument for a ${extract expansion item is the key name or field
6402 number. Leading and trailing spaces in this item were not being ignored,
6403 causing some misleading effects.
6405 5. When deliver_drop_privilege was set, single queue runner processes started
6406 manually (i.e. by the command "exim -q") or by the daemon (which uses the
6407 same command in the process it spins off) were not dropping privilege.
6409 6. When the daemon running as "exim" started a queue runner, it always
6410 re-executed Exim in the spun-off process. This is a waste of effort when
6411 deliver_drop_privilege is set. The new process now just calls the
6412 queue-runner function directly.
6418 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
6421 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
6422 ignored. Though the use of -oP was forcing the writing of a pid file, it
6423 was always written to the default place.
6425 3. If the message "no IP address found for host xxxx" is generated during
6426 incoming verification, it is now followed by identification of the incoming
6427 connection (so you can more easily find what provoked it).
6429 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
6431 5. Added some features to "harden" Exim a bit more against certain attacks:
6433 (a) There is now a build-time option called FIXED_NEVER_USERS that can
6434 be put in Local/Makefile. This is like the never_users runtime option,
6435 but it cannot be overridden. The default setting is "root".
6437 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
6438 prefix string with which any file named in a -C command line option
6441 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
6442 is retained for -C and -D only if the caller of Exim is root. Without
6443 it, the exim user may also use -C and -D and retain privilege.
6445 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
6446 command line option is disabled.
6448 6. Macro names set by the -D option must start with an upper case letter, just
6449 like macro names defined in the configuration file.
6451 7. Added "dereference=" facility to LDAP.
6453 8. Two instances of the typo "uknown" in the source files are fixed.
6455 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
6456 the Configure-Makefile script screwed up while processing it.
6458 10. Incorporated PCRE 4.4.
6460 11. The SMTP synchronization check was not operating right at the start of an
6461 SMTP session. For example, it could not catch a HELO sent before the client
6462 waited for the greeting. There is now a check for outstanding input at the
6463 point when the greeting is written. Because of the duplex, asynchronous
6464 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
6465 way, but not yet received, when the check is performed.
6467 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
6468 on TCP/IP sockets, because this apparently causes some broken clients to
6471 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
6472 unchanged) from the Cygwin maintainer.
6474 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
6475 was supported instead of testing for mbx. Effectively a typo.
6477 15. The spa authenticator server code was not checking that the input it
6478 received was valid base64.
6480 16. The debug output line for the "set" modifier in ACLs was not showing the
6481 name of the variable that was being set.
6483 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
6485 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
6486 lookup on its own. Something else was needed to trigger the lookup. For
6487 example, a match in host_lookup or the need for a host name in a host list.
6488 Now, if $sender_host_name is referenced and the host name has not yet been
6489 looked up, a lookup is performed. If the lookup fails, the variable remains
6490 empty, and $host_lookup_failed is set to "1".
6492 19. Added "eqi" as a case-independent comparison operator.
6494 20. The saslauthd authentication condition could segfault if neither service
6495 nor realm was specified.
6497 21. If an overflowing value such as "2048M" was set for message_size_limit, the
6498 error message that was logged was misleading, and incoming SMTP
6499 connections were dropped. The message is now more accurate, and temporary
6500 errors are given to SMTP connections.
6502 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
6503 (except RSET) with a 421 error, until QUIT is received. However, it was
6504 failing to send a response to QUIT.
6506 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
6507 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
6508 ACL is now run after the helo_try_verify_hosts code.
6510 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
6511 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
6512 case-independent, so other case variants are also recognized). Apparently
6513 some systems use these upper case variants.
6515 25. If more than two messages were waiting for the same host, and a transport
6516 filter was specified for the transport, Exim sent two messages over the
6517 same TCP/IP connection, and then failed with "socket operation on non-
6518 socket" when it tried to send the third.
6520 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
6522 27. The extern definition of crypt16() in expand.c was not being excluded when
6523 the OS had its own crypt16() function.
6525 28. Added bounce_return_body as a new option, and bounce_return_size_limit
6526 as a preferred synonym for return_size_limit, both as an option and as an
6529 29. Added LIBS=-liconv to OS/Makefile-OSF1.
6531 30. Changed the default configuration ACL to relax the local part checking rule
6532 for addresses that are not in any local domains. For these addresses,
6533 slashes and pipe symbols are allowed within local parts, but the sequence
6534 /../ is explicitly forbidden.
6536 31. SPA server authentication was not clearing the challenge buffer before
6539 32. log_message in a "warn" ACL statement was writing to the reject log as
6540 well as to the main log, which contradicts the documentation and doesn't
6541 seem right (because no rejection is happening). So I have stopped it.
6543 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
6544 However, I am unable to do any testing of this.
6546 34. Fixed an infelicity in the appendfile transport. When checking directories
6547 for a mailbox, to see if any needed to be created, it was accidentally
6548 using path names with one or more superfluous leading slashes; tracing
6549 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
6551 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
6552 contents are added to the log line that is written for every discarded
6553 recipient. (Previously a log_message setting was ignored.)
6555 36. The ${quote: operator now quotes the string if it is empty.
6557 37. The install script runs exim in order to find its version number. If for
6558 some reason other than non-existence or emptiness, which it checks, it
6559 could not run './exim', it was installing it with an empty version number,
6560 i.e. as "exim-". This error state is now caught, and the installation is
6563 38. An argument was missing from the function that creates an error message
6564 when Exim fails to connect to the socket for saslauthd authentication.
6565 This could cause Exim to crash, or give a corrupted message.
6567 39. Added isip, isip4, and isip6 to ${if conditions.
6569 40. The ACL variables $acl_xx are now saved with the message, and can be
6570 accessed later in routers, transports, and filters.
6572 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
6573 strings in the file are not string-expanded.
6575 42. If a MAIL command specified a SIZE value that was too large to fit into an
6576 int variable, the check against message_size_limit failed. Such values are
6577 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
6578 day this will have to be increased, but I don't think I want to be around
6579 when emails are that large.
6586 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
6587 iconv() is not standard in FreeBSD.
6589 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
6590 IPv6 enabled. The observed symptom was a segmentation fault on return from
6591 the function os_common_find_running_interfaces() in src/os.c.
6593 3. In the check_special_case() function in daemon.c I had used "errno" as an
6594 argument name, which causes warnings on some systems. This was basically a
6595 typo, since it was named "eno" in the comments!
6597 4. The code that waits for the clock to tick (at a resolution of some fraction
6598 of a second) so as to ensure message-id uniqueness was always waiting for
6599 at least one whole tick, when it could have waited for less. [This is
6600 almost certainly not relevant at current processor speeds, where it is
6601 unlikely to ever wait at all. But we try to future-proof.]
6603 5. The function that sleeps for a time interval that includes fractions of a
6604 second contained a race. It did not block SIGALRM between setting the
6605 timer, and suspending (a couple of lines later). If the interval was short
6606 and the sigsuspend() was delayed until after it had expired, the suspension
6607 never ended. On busy systems this could lead to processes getting stuck for
6610 6. Some uncommon configurations may cause a lookup to happen in a queue runner
6611 process, before it forks any delivery processes. The open lookup caching
6612 mechanism meant that the open file or database connection was passed into
6613 the delivery process. The problem was that delivery processes always tidy
6614 up cached lookup data. This could cause a problem for the next delivery
6615 process started by the queue runner, because the external queue runner
6616 process does not know about the closure. So the next delivery process
6617 still has data in the lookup cache. In the case of a file lookup, there was
6618 no problem because closing a file descriptor in a subprocess doesn't affect
6619 the parent. However, if the lookup was caching a connection to a database,
6620 the connection was closed, and the second delivery process was likely to
6621 see errors such as "PGSQL: query failed: server closed the connection
6622 unexpectedly". The problem has been fixed by closing all cached lookups
6623 in a queue runner before running a delivery process.
6625 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
6626 seem to have the "const" qualifier which it has on other OS. I've
6629 8. Change 4.21/2 was too strict. It is only if there are two authenticators
6630 *of the same type* (client or server) with the same public name that an
6631 error should be diagnosed.
6633 9. When Exim looked up a host name for an IP address, but failed to find the
6634 original IP address when looking up the host name (a safety check), it
6635 output the message "<ip address> does not match any IP for NULL", which was
6636 confusing, to say the least. The bug was that the host name should have
6637 appeared instead of "NULL".
6639 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
6640 user that is built into the binary, and the -C or -D options is used, root
6641 privilege is dropped before the configuration file is read. In addition,
6642 logging is switched to stderr instead of the normal log files. If the
6643 configuration then re-defines the Exim user, the unprivileged environment
6644 is probably not what is expected, so Exim logs a panic warning message (but
6647 However, if deliver_drop_privilege is set, the unprivileged state may well
6648 be exactly what is intended, so the warning has been cut out in that case,
6649 and Exim is allowed to try to write to its normal log files.
6655 1. smtp_return_error_details was not giving details for temporary sender
6656 or receiver verification errors.
6658 2. Diagnose a configuration error if two authenticators have the same public
6661 3. Exim used not to create the message log file for a message until the first
6662 delivery attempt. This could be confusing when incoming messages were held
6663 for policy or load reasons. The message log file is now created at the time
6664 the message is received, and an initial "Received" line is written to it.
6666 4. The automatically generated man page for command line options had a minor
6667 bug that caused no ill effects; however, a more serious problem was that
6668 the procedure for building the man page automatically didn't always
6669 operate. Consequently, release 4.20 contains an out-of-date version. This
6670 shouldn't happen again.
6672 5. When building Exim with embedded Perl support, the script that builds the
6673 Makefile was calling 'perl' to find its compile-time parameters, ignoring
6674 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
6676 6. The freeze_tell option was not being used for messages that were frozen on
6677 arrival, either by an ACL or by local_scan().
6679 7. Added the smtp_incomplete_transaction log selector.
6681 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
6682 was accepting AUTH without a new EHLO.
6684 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
6685 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
6688 10. Logging of TCP/IP connections (when configured) now happens in the main
6689 daemon process instead of the child process, so that the TCP/IP connection
6690 count is more accurate (but it can never be perfect).
6692 11. The use of "drop" in a nested ACL was not being handled correctly in the
6693 outer ACL. Now, if condition failure induced by the nested "drop" causes
6694 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
6695 or "require"), the connection is dropped.
6697 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
6698 that yield "discard" can now be used with an "accept" or a "discard" verb,
6699 but an error is generated for any others (because I can't see a useful way
6700 to define what should happen).
6702 13. When an ACL is read dynamically from a file (or anywhere else), the lines
6703 are now processed in the same way as lines in the Exim configuration file.
6704 In particular, continuation lines are supported.
6706 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
6708 15. Added -ti meaning -t -i.
6710 16. Check for letters, digits, hyphens, and dots in the names of dnslist
6711 domains, and warn by logging if others are found.
6713 17. At least on BSD, alignment is not guaranteed for the array of ifreq's
6714 returned from GIFCONF when Exim is trying to find the list of interfaces on
6715 a host. The code in os.c has been modified to copy each ifreq to an aligned
6716 structure in all cases.
6718 Also, in some cases, the returned ifreq's were being copied to a 'struct
6719 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
6720 means the last couple of bytes of an IPv6 address could be chopped if the
6721 ifreq contained only a normal sockaddr (14 bytes storage).
6723 18. Named domain lists were not supported in the hosts_treat_as_local option.
6724 An entry such as +xxxx was not recognized, and was treated as a literal
6727 19. Ensure that header lines added by a DATA ACL are included in the reject log
6728 if the ACL subsequently rejects the message.
6730 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
6731 MD5 (which is deprecated).
6733 21. When testing a filter file using -bf, Exim was writing a message when it
6734 took the sender from a "From " line in the message, but it was not doing so
6735 when it took $return_path from a Return-Path: header line. It now does.
6737 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
6738 with a valid header line field name (a series of printing characters
6739 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
6741 23. Changed "disc" in the source to "disk" to conform to the documentation and
6742 the book and for uniformity.
6744 24. Ignore Sendmail's -Ooption=value command line item.
6746 25. When execve() failed while trying to run a command in a pipe transport,
6747 Exim was returning EX_UNAVAILABLE (69) from the subprocess. However, this
6748 could be confused with a return value of 69 from the command itself. This
6749 has been changed to 127, the value the shell returns if it is asked to run
6750 a non-existent command. The wording for the related log line suggests a
6751 non-existent command as the problem.
6753 26. If received_header_text expands to an empty string, do not add a Received:
6754 header line to the message. (Well, it adds a token one on the spool, but
6755 marks it "old" so that it doesn't get used or transmitted.)
6757 27. Installed eximstats 1.28 (addition of -nt option).
6759 28. There was no check for failure on the call to getsockname() in the daemon
6760 code. This can fail if there is a shortage of resources on the system, with
6761 ENOMEM, for example. A temporary error is now given on failure.
6763 29. Contrary to the C standard, it seems that in some environments, the
6764 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
6765 program. Exim now does this explicitly; it affects the formatting of
6766 timestamps using strftime().
6768 30. If exiqsumm was given junk data, it threw up some uninitialized variable
6769 complaints. I've now initialized all the variables, to avoid this.
6771 32. Header lines added by a system filter were not being "seen" during
6772 transport-time rewrites.
6774 33. The info_callback() function passed to OpenSSL is set up with type void
6775 (*)(SSL *, int, int), as described somewhere. However, when calling the
6776 function (actually a macro) that sets it up, the type void(*)() is
6777 expected. I've put in a cast to prevent warnings from picky compilers.
6779 34. If a DNS black list lookup found a CNAME record, but there were no A
6780 records associated with the domain it pointed at, Exim crashed.
6782 35. If a DNS black list lookup returned more than one A record, Exim ignored
6783 all but the first. It now scans all returned addresses if a particular IP
6784 value is being sought. In this situation, the contents of the
6785 $dnslist_value variable are a list of all the addresses, separated by a
6788 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
6789 to accept a host name and all its aliases if the forward lookup for any of
6790 them yielded the IP address of the incoming connection. Now it accepts only
6791 those names whose forward lookup yields the correct IP address. Any other
6792 names are discarded. This closes a loophole whereby a rogue DNS
6793 administrator could create reverse DNS records to break through a
6794 wildcarded host restriction in an ACL.
6796 37. If a user filter or a system filter that ran in a subprocess used any of
6797 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
6798 the wrong values were passed to the pipe command ($thisaddress had the
6799 value of $0, $0 had the value of $1, etc). This bug was introduced by
6800 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
6802 38. Improved the line breaking for long SMTP error messages from ACLs.
6803 Previously, if there was no break point between 40 and 75 characters, Exim
6804 left the rest of the message alone. Two changes have been made: (a) I've
6805 reduced the minimum length to 35 characters; (b) if it can't find a break
6806 point between 35 and 75 characters, it looks ahead and uses the first one
6807 that it finds. This may give the occasional overlong line, but at least the
6808 remaining text gets split now.
6810 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
6811 file descriptors might be low, and that setting 1000 would always raise it.
6812 It turns out that in some environments, the limit is already over 1000 and
6813 that lowering it causes trouble. So now Exim takes care not to decrease it.
6815 40. When delivering a message, the value of $return_path is set to $sender_
6816 address at the start of routing (routers may change the value). By an
6817 oversight, this default was not being set up when an address was tested by
6818 -bt or -bv, which affected the outcome if any router or filter referred to
6821 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
6822 log only when "message" or "log_message" are set. However, if one of the
6823 conditions was an address verification, or a call to a nested ACL, the
6824 messages generated by the underlying test were being passed through. This
6825 no longer happens. The underlying message is available in $acl_verify_
6826 message for both "message" and "log_message" expansions, so it can be
6827 passed through if needed.
6829 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
6830 new expansion $bh_ to give the encoded byte string without charset
6831 translation. Translation happens only if iconv() is available; HAVE_ICONV
6832 indicates this at build time. HEADERS_CHARSET gives the charset to
6833 translate to; headers_charset can change it in the configuration, and
6834 "headers charset" can change it in an individual filter file.
6836 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
6837 that creates RFC 2047 encoded "words" labels them as that charset instead
6838 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
6839 expansion operator; (ii) when Exim creates a From: line for a local
6840 message; (iii) when a header line is rewritten to include a "phrase" part.
6842 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
6843 buggy, causing it to skip the first lines of messages whose message ID
6844 ended in 'D'. This would not have bitten before Exim release 4.14, because
6845 message IDs were unlikely to end in 'D' before then. The effect was to have
6846 incorrect size information for certain domains.
6848 45. #include "config.h" was missing at the start of the crypt16.c module. This
6849 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
6852 46. If there was a timeout during a "random" callout check, Exim treated it as
6853 a failure of the random address, and carried on sending RSET and the real
6854 address. If the delay was just some slowness somewhere, the response to the
6855 original RCPT would be taken as a response to RSET and so on, causing
6856 mayhem of various kinds.
6858 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
6859 when I implemented it. It didn't allow for the fact that some option values
6860 may legitimately be negative (e.g. size_addition), and it didn't even do
6861 the right test for positive values.
6863 48. Domain names in DNS records are case-independent. Exim always looks them up
6864 in lower case. Some resolvers return domain names in exactly the case they
6865 appear in the zone file, that is, they may contain uppercase letters. Not
6866 all resolvers do this - some return always lower case. Exim was treating a
6867 change of case by a resolver as a change of domain, similar to a widening
6868 of a domain abbreviation. This triggered its re-routing code and so it was
6869 trying to route what was effectively the same domain again. This normally
6870 caused routing to fail (because the router wouldn't handle the domain
6871 twice). Now Exim checks for this case specially, and just changes the
6872 casing of the domain that it ultimately uses when it transmits the message
6875 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
6878 50. If a filter generated a file delivery with a non-absolute name (possible if
6879 no home directory exists for the router), the forbid_file option was not
6882 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
6883 the existing equality matching.
6885 52. Exim was using ints instead of ino_t variables in some places where it was
6886 dealing with inode numbers.
6888 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
6889 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
6890 called TMPDIR, and if it finds it is different, it changes its value.
6892 54. The smtp_printf() function is now made available to local_scan() so
6893 additional output lines can be written before returning. There is also an
6894 smtp_fflush() function to enable the detection of a dropped connection.
6895 The variables smtp_input and smtp_batched_input are exported to
6898 55. Changed the default runtime configuration: the message "Unknown user"
6899 has been removed from the ACL, and instead placed on the localuser router,
6900 using the cannot_route_message feature. This means that any verification
6901 failures that generate their own messages won't get overridden. Similarly,
6902 the "Unrouteable address" message that was in the ACL for unverifiable
6903 relay addresses has also been removed.
6905 56. Added hosts_avoid_esmtp to the smtp transport.
6907 57. The exicyclog script was not checking for the esoteric option
6908 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
6909 will work only if exicyclog is run under the appropriate euid.
6911 58. Following a discussion on the list, the rules by which Exim recognises line
6912 endings on incoming messages have been changed. The -dropcr and drop_cr
6913 options are now no-ops, retained only for backwards compatibility. The
6914 following line terminators are recognized: LF CRLF CR. However, special
6915 processing applies to CR:
6917 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
6918 nor a local message in the state where . is a terminator.
6920 (ii) If a bare CR is encountered in a header line, an extra space is added
6921 after the line terminator so as not to end the header. The reasoning
6922 behind this is that bare CRs in header lines are most likely either
6923 to be mistakes, or people trying to play silly games.
6925 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
6926 was being incorrectly given as 18 bytes larger than it should have been.
6927 This is a VOB (very old bug).
6929 60. This may never have affected anything current, but just in case it has:
6930 When the local host is found other than at the start of a list of hosts,
6931 the local host, those with the same MX, and any that follow, are discarded.
6932 When the list in question was part of a longer list of hosts, the following
6933 hosts (not currently being processed) were also being discarded. This no
6934 longer happens. I'm not sure if this situation could ever has previously
6937 61. Added the "/MX" feature to lists of hosts in the manualroute and query
6940 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
6941 header. This is something that is recommended in a new Internet Draft, and
6942 is something that is documented as being done by Sendmail. There are two
6943 possible values. For messages generated by the autoreply transport, Exim
6946 Auto-Submitted: auto-replied
6948 whereas for all other generated messages (e.g. bounces) it adds
6950 Auto-Submitted: auto-generated
6952 63. The "personal" condition in filters now includes a test for the
6953 Auto-Submitted: header. If it contains the string "auto-" the message it
6954 not considered personal.
6956 64. Added rcpt_include_affixes as a generic transport option.
6958 65. Added queue_only_override (default true).
6960 66. Added the syslog_duplication option.
6962 67. If what should have been the first header line of a message consisted of
6963 a space followed by a colon, Exim was mis-interpreting it as a header line.
6964 It isn't of course - it is syntactically invalid and should therefore be
6965 treated as the start of the message body. The misbehaviour could have
6966 caused a number of strange effects, including loss of data in subsequent
6967 header lines, and spool format errors.
6969 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
6970 client host had authenticated. This control can now be exercised by an ACL
6971 for more flexibility.
6973 69. By default, callouts do not happen when testing with -bh. There is now a
6974 variant, -bhc, which does actually run the callout code, including
6975 consulting and updating the callout cache.
6977 70. Added support for saslauthd authentication, courtesy of Alexander
6980 71. If statvfs() failed on the spool or log directories while checking their
6981 size for availability, Exim confusingly gave the error "space shortage".
6982 Furthermore, in debugging mode it crashed with a floating point exception.
6983 These checks are done if check_{spool,log}_{space,inodes} are set, and when
6984 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
6985 serious problem, Exim now writes to the main and panic logs when this
6986 happens, with details of the failure. It then refuses to accept the
6987 incoming message, giving the message "spool directory problem" or "log
6988 directory problem" with a 421 code for SMTP messages.
6990 72. When Exim is about to re-exec itself, it ensures that the file descriptors
6991 0, 1, and 2 exist, because some OS complain for execs without them (see
6992 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
6993 descriptors. However, the code omitted to check that the open succeeded,
6994 causing mysterious errors if for some reason the permissions on /dev/null
6995 got screwed. Now Exim writes a message to the main and panic logs, and
6996 bombs out if it can't open /dev/null.
6998 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
6999 interact so that it is all more flexible. It is supposed to remain
7000 backwards compatible. Also added extra_local_interfaces.
7002 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
7003 to bomb out with an assertion failure - to the client this appears as a
7004 connection drop. This problem occurs in the part of the code that was taken
7005 from the Samba project. Fortunately, the assertion is in a very simple
7006 function, so I have fixed this by reproducing the function inline in the
7007 one place where it is called, and arranging for authentication to fail
7008 instead of killing the process with assert().
7010 75. The SPA client code was not working when the server requested OEM rather
7011 than Unicode encoding.
7013 76. Added code to make require_files with a specific uid setting more usable in
7014 the case where statting the file as root fails - usually a non-root-mounted
7015 NFS file system. When this happens and the failure is EACCES, Exim now
7016 forks a subprocess and does the per-uid checking as the relevant uid.
7018 77. Added process_log_path.
7020 78. If log_file_path was not explicitly set, a setting of check_log_space or
7021 check_log_inodes was ignored.
7023 79. If a space check for the spool or log partitions fails, the incident is now
7024 logged. Of course, in the latter case the data may get lost...
7026 80. Added the %p formatting code to string_format() so that it can be used to
7027 print addresses in debug_print(). Adjusted all the address printing in the
7028 debugging in store.c to use %p rather than %d.
7030 81. There was a concern that a line of code in smtp_in.c could overflow a
7031 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
7032 initially expressed, the concern was not well-founded, because trailing
7033 spaces are removed early. However, if the trailing spaces were followed by
7034 a NULL, they did not get removed, so the overflow was possible. Two fixes
7037 (a) I re-wrote the offending code in a cleaner fashion.
7038 (b) If an incoming SMTP command contains a NULL character, it is rejected
7041 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
7042 runs initgroups(), so that if the Exim user is in any additional groups,
7043 they will be used during message reception.
7049 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / exim.git / blob