1 Change log file for Exim from version 4.21
2 ------------------------------------------
3 This document describes *changes* to previous versions, that might
4 affect Exim's operation, with an unchanged configuration file. For new
5 options, and new features, see the NewStuff file next to this ChangeLog.
8 Since Exim version 4.90
9 -----------------------
11 GF/01 DEFER rather than ERROR on redis cluster MOVED response.
12 When redis_servers is set to a list of > 1 element, and the Redis servers
13 in that list are in cluster configuration, convert the REDIS_REPLY_ERROR
14 case of MOVED into a DEFER case instead, thus moving the query onto the
15 next server in the list. For a cluster of N elements, all N servers must
16 be defined in redis_servers.
18 GF/02 Catch and remove uninitialized value warning in exiqsumm
19 Check for existence of @ARGV before looking at $ARGV[0]
21 JH/01 Replace the store_release() internal interface with store_newblock(),
22 which internalises the check required to safely use the old one, plus
23 the allocate and data copy operations duplicated in both (!) of the
26 JH/02 Disallow '/' characters in queue names specified for the "queue=" ACL
27 modifier. This matches the restriction on the commandline.
29 JH/03 Fix pgsql lookup for multiple result-tuples with a single column.
30 Previously only the last row was returned.
32 JH/04 Bug 2217: Tighten up the parsing of DKIM signature headers. Previously
33 we assumed that tags in the header were well-formed, and parsed the
34 element content after inspecting only the first char of the tag.
35 Assumptions at that stage could crash the receive process on malformed
38 JH/05 Bug 2215: Fix crash associated with dnsdb lookup done from DKIM ACL.
39 While running the DKIM ACL we operate on the Permanent memory pool so that
40 variables created with "set" persist to the DATA ACL. Also (at any time)
41 DNS lookups that fail create cache records using the Permanent pool. But
42 expansions release any allocations made on the current pool - so a dnsdb
43 lookup expansion done in the DKIM ACL releases the memory used for the
44 DNS negative-cache, and bad things result. Solution is to switch to the
45 Main pool for expansions.
46 While we're in that code, add checks on the DNS cache during store_reset,
47 active in the testsuite.
48 Problem spotted, and debugging aided, by Wolfgang Breyha.
50 JH/06 Fix issue with continued-connections when the DNS shifts unreliably.
51 When none of the hosts presented to a transport match an already-open
52 connection, close it and proceed with the list. Previously we would
53 queue the message. Spotted by Lena with Yahoo, probably involving
56 JH/07 Bug 2214: Fix SMTP responses resulting from non-accept result of MIME ACL.
57 Previously a spurious "250 OK id=" response was appended to the proper
60 JH/08 The "support for" informational output now, which built with Content
61 Scanning support, has a line for the malware scanner interfaces compiled
62 in. Interface can be individually included or not at build time.
64 JH/09 The "aveserver", "kavdaemon" and "mksd" interfaces are now not included
65 by the template makefile "src/EDITME". The "STREAM" support for an older
66 ClamAV interface method is removed.
68 JH/10 Bug 2223: Fix mysql lookup returns for the no-data case (when the number of
69 rows affected is given instead).
71 JH/11 The runtime Berkeley DB library version is now additionally output by
72 "exim -d -bV". Previously only the compile-time version was shown.
74 JH/12 Bug 2230: Fix cutthrough routing for nonfirst messages in an initiating
75 SMTP connection. Previously, when one had more receipients than the
76 first, an abortive onward connection was made. Move to full support for
77 multiple onward connections in sequence, handling cutthrough connection
78 for all multi-message initiating connections.
80 JH/13 Bug 2229: Fix cutthrough routing for nonstandard port numbers defined by
81 routers. Previously, a multi-recipient message would fail to match the
82 onward-connection opened for the first recipient, and cause its closure.
84 JH/14 Bug 2174: A timeout on connect for a callout was also erroneously seen as
85 a timeout on read on a GnuTLS initiating connection, resulting in the
86 initiating connection being dropped. This mattered most when the callout
87 was marked defer_ok. Fix to keep the two timeout-detection methods
90 JH/15 Relax results from ACL control request to enable cutthrough, in
91 unsupported situations, from error to silently (except under debug)
92 ignoring. This covers use with PRDR, frozen messages, queue-only and
95 HS/01 Fix Buffer overflow in base64d() (CVE-2018-6789)
97 JH/16 Fix bug in DKIM verify: a buffer overflow could corrupt the malloc
98 metadata, resulting in a crash in free().
100 PP/01 Fix broken Heimdal GSSAPI authenticator integration.
101 Broken in f2ed27cf5, missing an equals sign for specified-initialisers.
102 Broken also in d185889f4, with init system revamp.
104 JH/17 Bug 2113: Fix conversation closedown with the Avast malware scanner.
105 Previously we abruptly closed the connection after reading a malware-
106 found indication; now we go on to read the "scan ok" response line,
109 JH/18 Bug 2239: Enforce non-usability of control=utf8_downconvert in the mail
110 ACL. Previously, a crash would result.
112 JH/19 Speed up macro lookups during configuration file read, by skipping non-
113 macro text after a replacement (previously it was only once per line) and
114 by skipping builtin macros when searching for an uppercase lead character.
116 JH/20 DANE support moved from Experimental to mainline. The Makefile control
117 for the build is renamed.
119 JH/21 Fix memory leak during multi-message connections using STARTTLS. A buffer
120 was allocated for every new TLS startup, meaning one per message. Fix
121 by only allocating once (OpenSSL) or freeing on TLS-close (GnuTLS).
123 JH/22 Bug 2236: When a DKIM verification result is overridden by ACL, DMARC
124 reported the original. Fix to report (as far as possible) the ACL
125 result replacing the original.
127 JH/23 Fix memory leak during multi-message connections using STARTTLS under
128 OpenSSL. Certificate information is loaded for every new TLS startup,
129 and the resources needed to be freed.
131 JH/24 Bug 2242: Fix exim_dbmbuild to permit directoryless filenames.
133 JH/25 Fix utf8_downconvert propagation through a redirect router. Previously it
136 JH/26 Bug 2253: For logging delivery lines under PRDR, append the overall
137 DATA response info to the (existing) per-recipient response info for
138 the "C=" log element. It can have useful tracking info from the
139 destination system. Patch from Simon Arlott.
141 JH/27 Bug 2251: Fix ldap lookups that return a single attribute having zero-
142 length value. Previously this would segfault.
144 HS/02 Support Avast multiline protoocol, this allows passing flags to
145 newer versions of the scanner.
147 JH/28 Ensure that variables possibly set during message acceptance are marked
148 dead before release of memory in the daemon loop. This stops complaints
149 about them when the debug_store option is enabled. Discovered specifically
150 for sender_rate_period, but applies to a whole set of variables.
151 Do the same for the queue-runner loop, for variables set from spool
152 message files. Do the same for the SMTP per-message loop, for certain
153 variables indirectly set in ACL operations.
155 JH/29 Bug 2250: Fix a longstanding bug in heavily-pipelined SMTP input (such
156 as a multi-recipient message from a mailinglist manager). The coding had
157 an arbitrary cutoff number of characters while checking for more input;
158 enforced by writing a NUL into the buffer. This corrupted long / fast
159 input. The problem was exposed more widely when more pipelineing of SMTP
160 responses was introduced, and one Exim system was feeding another.
161 The symptom is log complaints of SMTP syntax error (NUL chars) on the
162 receiving system, and refused recipients seen by the sending system
163 (propating to people being dropped from mailing lists).
164 Discovered and pinpointed by David Carter.
166 JH/30 The (EXPERIMENTAL_DMARC) variable $dmarc_ar_header is withdrawn, being
167 replaced by the ${authresults } expansion.
169 JH/31 Bug 2257: Fix pipe transport to not use a socket-only syscall.
171 HS/03 Set a handler for SIGTERM and call exit(3) if running as PID 1. This
172 allows proper process termination in container environments.
178 JH/01 Rework error string handling in TLS interface so that the caller in
179 more cases is responsible for logging. This permits library-sourced
180 string to be attached to addresses during delivery, and collapses
181 pairs of long lines into single ones.
183 PP/01 Allow PKG_CONFIG_PATH to be set in Local/Makefile and use it correctly
184 during configuration. Wildcards are allowed and expanded.
186 JH/02 Rework error string handling in DKIM to pass more info back to callers.
187 This permits better logging.
189 JH/03 Rework the transport continued-connection mechanism: when TLS is active,
190 do not close it down and have the child transport start it up again on
191 the passed-on TCP connection. Instead, proxy the child (and any
192 subsequent ones) for TLS via a unix-domain socket channel. Logging is
193 affected: the continued delivery log lines do not have any DNSSEC, TLS
194 Certificate or OCSP information. TLS cipher information is still logged.
196 JH/04 Shorten the log line for daemon startup by collapsing adjacent sets of
197 identical IP addresses on different listening ports. Will also affect
200 PP/02 Bug 2070: uClibc defines __GLIBC__ without providing glibc headers;
201 add noisy ifdef guards to special-case this sillyness.
202 Patch from Bernd Kuhls.
204 JH/05 Tighten up the checking in isip4 (et al): dotted-quad components larger
205 than 255 are no longer allowed.
207 JH/06 Default openssl_options to include +no_ticket, to reduce load on peers.
208 Disable the session-cache too, which might reduce our load. Since we
209 currrectly use a new context for every connection, both as server and
210 client, there is no benefit for these.
211 GnuTLS appears to not support tickets server-side by default (we don't
212 call gnutls_session_ticket_enable_server()) but client side is enabled
213 by default on recent versions (3.1.3 +) unless the PFS priority string
216 PP/03 Add $SOURCE_DATE_EPOCH support for reproducible builds, per spec at
217 <https://reproducible-builds.org/specs/source-date-epoch/>.
219 JH/07 Fix smtp transport use of limited max_rcpt under mua_wrapper. Previously
220 the check for any unsuccessful recipients did not notice the limit, and
221 erroneously found still-pending ones.
223 JH/08 Pipeline CHUNKING command and data together, on kernels that support
224 MSG_MORE. Only in-clear (not on TLS connections).
226 JH/09 Avoid using a temporary file during transport using dkim. Unless a
227 transport-filter is involved we can buffer the headers in memory for
228 creating the signature, and read the spool data file once for the
229 signature and again for transmission.
231 JH/10 Enable use of sendfile in Linux builds as default. It was disabled in
232 4.77 as the kernel support then wasn't solid, having issues in 64bit
233 mode. Now, it's been long enough. Add support for FreeBSD also.
235 JH/11 Bug 2104: Fix continued use of a transport connection with TLS. In the
236 case where the routing stage had gathered several addresses to send to
237 a host before calling the transport for the first, we previously failed
238 to close down TLS in the old transport process before passing the TCP
239 connection to the new process. The new one sent a STARTTLS command
240 which naturally failed, giving a failed delivery and bloating the retry
241 database. Investigation and fix prototype from Wolfgang Breyha.
243 JH/12 Fix check on SMTP command input synchronisation. Previously there were
244 false-negatives in the check that the sender had not preempted a response
245 or prompt from Exim (running as a server), due to that code's lack of
246 awareness of the SMTP input buffering.
248 PP/04 Add commandline_checks_require_admin option.
249 Exim drops privileges sanely, various checks such as -be aren't a
250 security problem, as long as you trust local users with access to their
251 own account. When invoked by services which pass untrusted data to
252 Exim, this might be an issue. Set this option in main configuration
253 AND make fixes to the calling application, such as using `--` to stop
256 JH/13 Do pipelining under TLS. Previously, although safe, no advantage was
257 taken. Now take care to pack both (client) MAIL,RCPT,DATA, and (server)
258 responses to those, into a single TLS record each way (this usually means
259 a single packet). As a side issue, smtp_enforce_sync now works on TLS
262 PP/05 OpenSSL/1.1: use DH_bits() for more accurate DH param sizes. This
263 affects you only if you're dancing at the edge of the param size limits.
264 If you are, and this message makes sense to you, then: raise the
265 configured limit or use OpenSSL 1.1. Nothing we can do for older
268 JH/14 For the "sock" variant of the malware scanner interface, accept an empty
269 cmdline element to get the documented default one. Previously it was
272 JH/15 Fix a crash in the smtp transport caused when two hosts in succession
273 are unsuable for non-message-specific reasons - eg. connection timeout,
274 banner-time rejection.
276 JH/16 Fix logging of delivery remote port, when specified by router, under
279 PP/06 Repair manualroute's ability to take options in any order, even if one
280 is the name of a transport.
283 HS/01 Cleanup, prevent repeated use of -p/-oMr (CVE-2017-1000369)
285 JH/17 Change the list-building routines interface to use the expanding-string
286 triplet model, for better allocation and copying behaviour.
288 JH/18 Prebuild the data-structure for "builtin" macros, for faster startup.
289 Previously it was constructed the first time a possibly-matching string
290 was met in the configuration file input during startup; now it is done
293 JH/19 Bug 2141: Use the full-complex API for Berkeley DB rather than the legacy-
294 compatible one, to avoid the (poorly documented) possibility of a config
295 file in the working directory redirecting the DB files, possibly correpting
296 some existing file. CVE-2017-10140 assigned for BDB.
298 JH/20 Bug 2147: Do not defer for a verify-with-callout-and-random which is not
299 cache-hot. Previously, although the result was properly cached, the
300 initial verify call returned a defer.
302 JH/21 Bug 2151: Avoid using SIZE on the MAIL for a callout verify, on any but
303 the main verify for receipient in uncached-mode.
305 JH/22 Retire historical build files to an "unsupported" subdir. These are
306 defined as "ones for which we have no current evidence of testing".
308 JH/23 DKIM: enforce the DNS pubkey record "h" permitted-hashes optional field,
309 if present. Previously it was ignored.
311 JH/24 Start using specified-initialisers in C structure init coding. This is
312 a C99 feature (it's 2017, so now considered safe).
314 JH/25 Use one-bit bitfields for flags in the "addr" data structure. Previously
315 if was a fixed-sized field and bitmask ops via macros; it is now more
318 PP/07 GitHub PR 56: Apply MariaDB build fix.
319 Patch provided by Jaroslav Škarvada.
321 PP/08 Bug 2161: Fix regression in sieve quoted-printable handling introduced
322 during Coverity cleanups [4.87 JH/47]
323 Diagnosis and fix provided by Michael Fischer v. Mollard.
325 JH/26 Fix DKIM bug: when the pseudoheader generated for signing was exactly
326 the right size to place the terminating semicolon on its own folded
327 line, the header hash was calculated to an incorrect value thanks to
328 the (relaxed) space the fold became.
330 HS/02 Fix Bug 2130: large writes from the transport subprocess where chunked
331 and confused the parent.
333 JH/27 Fix SOCKS bug: an unitialized pointer was deref'd by the transport process
334 which could crash as a result. This could lead to undeliverable messages.
336 JH/28 Logging: "next input sent too soon" now shows where input was truncated
339 JH/29 Fix queue_run_in_order to ignore the PID portion of the message ID. This
340 matters on fast-turnover and PID-randomising systems, which were getting
341 out-of-order delivery.
343 JH/30 Fix a logging bug on aarch64: an unsafe routine was previously used for
344 a possibly-overlapping copy. The symptom was that "Remote host closed
345 connection in response to HELO" was logged instead of the actual 4xx
348 JH/31 Fix CHUNKING code to properly flush the unwanted chunk after an error.
349 Previously only that bufferd was discarded, resulting in SYMTP command
352 JH/32 DKIM: when a message has multiple signatures matching an identity given
353 in dkim_verify_signers, run the dkim acl once for each. Previously only
354 one run was done. Bug 2189.
356 JH/33 Downgrade an unfound-list name (usually a typo in the config file) from
357 "panic the current process" to "deliberately defer". The panic log is
358 still written with the problem list name; the mail and reject logs now
359 get a temp-reject line for the message that was being handled, saying
360 something like "domains check lookup or other defer". The SMTP 451
361 message is still "Temporary local problem".
363 JH/34 Bug 2199: Fix a use-after-free while reading smtp input for header lines.
364 A crafted sequence of BDAT commands could result in in-use memory beeing
365 freed. CVE-2017-16943.
367 HS/03 Bug 2201: Fix checking for leading-dot on a line during headers reading
368 from SMTP input. Previously it was always done; now only done for DATA
369 and not BDAT commands. CVE-2017-16944.
371 JH/35 Bug 2201: Flush received data in BDAT mode after detecting an error fatal
372 to the message (such as an overlong header line). Previously this was
373 not done and we did not exit BDAT mode. Followon from the previous item
374 though a different problem.
380 JH/01 Bug 1922: Support IDNA2008. This has slightly different conversion rules
381 than -2003 did; needs libidn2 in addition to libidn.
383 JH/02 The path option on a pipe transport is now expanded before use.
385 PP/01 GitHub PR 50: Do not call ldap_start_tls_s on ldapi:// connections.
386 Patch provided by "Björn", documentation fix added too.
388 JH/03 Bug 2003: fix Proxy Protocol v2 handling: the address size field was
389 missing a wire-to-host endian conversion.
391 JH/04 Bug 2004: fix CHUNKING in non-PIPELINEING mode. Chunk data following
392 close after a BDAT command line could be taken as a following command,
393 giving a synch failure. Fix by only checking for synch immediately
394 before acknowledging the chunk.
396 PP/02 GitHub PR 52: many spelling fixes, which include fixing parsing of
397 no_require_dnssec option and creation of _HAVE_TRANSPORT_APPEND_MAILDIR
398 macro. Patches provided by Josh Soref.
400 JH/05 Have the EHLO response advertise VRFY, if there is a vrfy ACL defined.
401 Previously we did not; the RFC seems ambiguous and VRFY is not listed
402 by IANA as a service extension. However, John Klensin suggests that we
405 JH/06 Bug 2017: Fix DKIM verification in -bh test mode. The data feed into
406 the dkim code may be unix-mode line endings rather than smtp wire-format
407 CRLF, so prepend a CR to any bare LF.
409 JH/07 Rationalise the coding for callout smtp conversations and transport ones.
410 As a side-benfit, callouts can now use PIPELINING hence fewer round-trips.
412 JH/08 Bug 2016: Fix DKIM verification vs. CHUNKING. Any BDAT commands after
413 the first were themselves being wrongly included in the feed into dkim
414 processing; with most chunk sizes in use this resulted in an incorrect
415 body hash calculated value.
417 JH/09 Bug 2014: permit inclusion of a DKIM-Signature header in a received
418 DKIM signature block, for verification. Although advised against by
419 standards it is specifically not ruled illegal.
421 JH/10 Bug 2025: Fix reception of (quoted) local-parts with embedded spaces.
423 JH/11 Bug 2029: Fix crash in DKIM verification when a message signature block is
424 missing a body hash (the bh= tag).
426 JH/12 Bug 2018: Re-order Proxy Protocol startup versus TLS-on-connect startup.
427 It seems that HAProxy sends the Proxy Protocol information in clear and
428 only then does a TLS startup, so do the same.
430 JH/13 Bug 2027: Avoid attempting to use TCP Fast Open for non-transport client
431 TCP connections (such as for Spamd) unless the daemon successfully set
432 Fast Open mode on its listening sockets. This fixes breakage seen on
433 too-old kernels or those not configured for Fast Open, at the cost of
434 requiring both directions being enabled for TFO, and TFO never being used
435 by non-daemon-related Exim processes.
437 JH/14 Bug 2000: Reject messages recieved with CHUNKING but with malformed line
438 endings, at least on the first header line. Try to canonify any that get
439 past that check, despite the cost.
441 JH/15 Angle-bracket nesting (an error inserted by broken sendmails) levels are
442 now limited to an arbitrary five deep, while parsing addresses with the
443 strip_excess_angle_brackets option enabled.
445 PP/03 Bug 2018: For Proxy Protocol and TLS-on-connect, do not over-read and
446 instead leave the unprompted TLS handshake in socket buffer for the
447 TLS library to consume.
449 PP/04 Bug 2018: Also handle Proxy Protocol v2 safely.
451 PP/05 FreeBSD compat: handle that Ports no longer create /usr/bin/perl
453 JH/16 Drop variables when they go out of scope. Memory management drops a whole
454 region in one operation, for speed, and this leaves assigned pointers
455 dangling. Add checks run only under the testsuite which checks all
456 variables at a store-reset and panics on a dangling pointer; add code
457 explicitly nulling out all the variables discovered. Fixes one known
458 bug: a transport crash, where a dangling pointer for $sending_ip_address
459 originally assigned in a verify callout, is re-used.
461 PP/06 Drop '.' from @INC in various Perl scripts.
463 PP/07 Switch FreeBSD iconv to always use the base-system libc functions.
465 PP/08 Reduce a number of compilation warnings under clang; building with
466 CC=clang CFLAGS+=-Wno-dangling-else -Wno-logical-op-parentheses
467 should be warning-free.
469 JH/17 Fix inbound CHUNKING when DKIM disabled at runtime.
471 HS/01 Fix portability problems introduced by PP/08 for platforms where
472 realloc(NULL) is not equivalent to malloc() [SunOS et al].
474 HS/02 Bug 1974: Fix missing line terminator on the last received BDAT
475 chunk. This allows us to accept broken chunked messages. We need a more
476 general solution here.
478 PP/09 Wrote util/chunking_fixqueue_finalnewlines.pl to help recover
479 already-broken messages in the queue.
481 JH/18 Bug 2061: Fix ${extract } corrupting an enclosing ${reduce } $value.
483 JH/19 Fix reference counting bug in routing-generated-address tracking.
489 JH/01 Use SIZE on MAIL FROM in a cutthrough connection, if the destination
490 supports it and a size is available (ie. the sending peer gave us one).
492 JH/02 The obsolete acl condition "demime" is removed (finally, after ten
493 years of being deprecated). The replacements are the ACLs
494 acl_smtp_mime and acl_not_smtp_mime.
496 JH/03 Upgrade security requirements imposed for hosts_try_dane: previously
497 a downgraded non-dane trust-anchor for the TLS connection (CA-style)
498 or even an in-clear connection were permitted. Now, if the host lookup
499 was dnssec and dane was requested then the host is only used if the
500 TLSA lookup succeeds and is dnssec. Further hosts (eg. lower priority
501 MXs) will be tried (for hosts_try_dane though not for hosts_require_dane)
502 if one fails this test.
503 This means that a poorly-configured remote DNS will make it incommunicado;
504 but it protects against a DNS-interception attack on it.
506 JH/04 Bug 1810: make continued-use of an open smtp transport connection
507 non-noisy when a race steals the message being considered.
509 JH/05 If main configuration option tls_certificate is unset, generate a
510 self-signed certificate for inbound TLS connections.
512 JH/06 Bug 165: hide more cases of password exposure - this time in expansions
513 in rewrites and routers.
515 JH/07 Retire gnutls_require_mac et.al. These were nonfunctional since 4.80
516 and logged a warning sing 4.83; now they are a configuration file error.
518 JH/08 Bug 1836: Fix crash in VRFY handling when handed an unqualified name
519 (lacking @domain). Apply the same qualification processing as RCPT.
521 JH/09 Bug 1804: Avoid writing msglog files when in -bh or -bhc mode.
523 JH/10 Support ${sha256:} applied to a string (as well as the previous
526 JH/11 Cutthrough: avoid using the callout hints db on a verify callout when
527 a cutthrough deliver is pending, as we always want to make a connection.
528 This also avoids re-routing the message when later placing the cutthrough
529 connection after a verify cache hit.
530 Do not update it with the verify result either.
532 JH/12 Cutthrough: disable when verify option success_on_redirect is used, and
533 when routing results in more than one destination address.
535 JH/13 Cutthrough: expand transport dkim_domain option when testing for dkim
536 signing (which inhibits the cutthrough capability). Previously only
537 the presence of an option was tested; now an expansion evaluating as
538 empty is permissible (obviously it should depend only on data available
539 when the cutthrough connection is made).
541 JH/14 Fix logging of errors under PIPELINING. Previously the log line giving
542 the relevant preceding SMTP command did not note the pipelining mode.
544 JH/15 Fix counting of empty lines in $body_linecount and $message_linecount.
545 Previously they were not counted.
547 JH/16 DANE: treat a TLSA lookup response having all non-TLSA RRs, the same
548 as one having no matching records. Previously we deferred the message
549 that needed the lookup.
551 JH/17 Fakereject: previously logged as a normal message arrival "<="; now
552 distinguished as "(=".
554 JH/18 Bug 1867: make the fail_defer_domains option on a dnslookup router work
555 for missing MX records. Previously it only worked for missing A records.
557 JH/19 Bug 1850: support Radius libraries that return REJECT_RC.
559 JH/20 Bug 1872: Ensure that acl_smtp_notquit is run when the connection drops
560 after the data-go-ahead and data-ack. Patch from Jason Betts.
562 JH/21 Bug 1846: Send DMARC forensic reports for reject and quarantine results,
563 even for a "none" policy. Patch from Tony Meyer.
565 JH/22 Fix continued use of a connection for further deliveries. If a port was
566 specified by a router, it must also match for the delivery to be
569 JH/23 Bug 1874: fix continued use of a connection for further deliveries.
570 When one of the recipients of a message was unsuitable for the connection
571 (has no matching addresses), we lost track of needing to mark it
572 deferred. As a result mail would be lost.
574 JH/24 Bug 1832: Log EHLO response on getting conn-close response for HELO.
576 JH/25 Decoding ACL controls is now done using a binary search; the source code
577 takes up less space and should be simpler to maintain. Merge the ACL
578 condition decode tables also, with similar effect.
580 JH/26 Fix problem with one_time used on a redirect router which returned the
581 parent address unchanged. A retry would see the parent address marked as
582 delivered, so not attempt the (identical) child. As a result mail would
585 JH/27 Fix a possible security hole, wherein a process operating with the Exim
586 UID can gain a root shell. Credit to http://www.halfdog.net/ for
587 discovery and writeup. Ubuntu bug 1580454; no bug raised against Exim
590 JH/28 Enable {spool,log} filesystem space and inode checks as default.
591 Main config options check_{log,spool}_{inodes,space} are now
592 100 inodes, 10MB unless set otherwise in the configuration.
594 JH/29 Fix the connection_reject log selector to apply to the connect ACL.
595 Previously it only applied to the main-section connection policy
598 JH/30 Bug 1897: fix callouts connection fallback from TLS to cleartext.
600 PP/01 Changed default Diffie-Hellman parameters to be Exim-specific, created
601 by me. Added RFC7919 DH primes as an alternative.
603 PP/02 Unbreak build via pkg-config with new hash support when crypto headers
604 are not in the system include path.
606 JH/31 Fix longstanding bug with aborted TLS server connection handling. Under
607 GnuTLS, when a session startup failed (eg because the client disconnected)
608 Exim did stdio operations after fclose. This was exposed by a recent
609 change which nulled out the file handle after the fclose.
611 JH/32 Bug 1909: Fix OCSP proof verification for cases where the proof is
612 signed directly by the cert-signing cert, rather than an intermediate
613 OCSP-signing cert. This is the model used by LetsEncrypt.
615 JH/33 Bug 1914: Ensure socket is nonblocking before draining after SMTP QUIT.
617 HS/01 Fix leak in verify callout under GnuTLS, about 3MB per recipient on
618 an incoming connection.
620 HS/02 Bug 1802: Do not half-close the connection after sending a request
623 HS/03 Use "auto" as the default EC curve parameter. For OpenSSL < 1.0.2
624 fallback to "prime256v1".
626 JH/34 SECURITY: Use proper copy of DATA command in error message.
627 Could leak key material. Remotely exploitable. CVE-2016-9963.
633 JH/01 Bug 1664: Disable OCSP for GnuTLS library versions at/before 3.3.16
634 and 3.4.4 - once the server is enabled to respond to an OCSP request
635 it does even when not requested, resulting in a stapling non-aware
636 client dropping the TLS connection.
638 TF/01 Code cleanup: Overhaul the debug_selector and log_selector machinery to
639 support variable-length bit vectors. No functional change.
641 TF/02 Improve the consistency of logging incoming and outgoing interfaces.
642 The I= interface field on outgoing lines is now after the H= remote
643 host field, same as incoming lines. There is a separate
644 outgoing_interface log selector which allows you to disable the
647 JH/02 Bug 728: Close logfiles after a daemon-process "exceptional" log write.
648 If not running log_selector +smtp_connection the mainlog would be held
649 open indefinitely after a "too many connections" event, including to a
650 deleted file after a log rotate. Leave the per net connection logging
651 leaving it open for efficiency as that will be quickly detected by the
652 check on the next write.
654 HS/01 Bug 1671: Fix post transport crash.
655 Processing the wait-<transport> messages could crash the delivery
656 process if the message IDs didn't exist for some reason. When
657 using 'split_spool_directory=yes' the construction of the spool
658 file name failed already, exposing the same netto behaviour.
660 JH/03 Bug 425: Capture substrings in $regex1, $regex2 etc from regex &
661 mime_regex ACL conditions.
663 JH/04 Bug 1686: When compiled with EXPERIMENTAL_DSN_INFO: Add extra information
664 to DSN fail messages (bounces): remote IP, remote greeting, remote response
665 to HELO, local diagnostic string.
667 JH/05 Downgrade message for a TLS-certificate-based authentication fail from
668 log line to debug. Even when configured with a tls authenticator many
669 client connections are expected to not authenticate in this way, so
670 an authenticate fail is not an error.
672 HS/02 Add the Exim version string to the process info. This way exiwhat
673 gives some more detail about the running daemon.
675 JH/06 Bug 1395: time-limit caching of DNS lookups, to the TTL value. This may
676 matter for fast-change records such as DNSBLs.
678 JH/07 Bug 1678: Always record an interface option value, if set, as part of a
679 retry record, even if constant. There may be multiple transports with
680 different interface settings and the retry behaviour needs to be kept
683 JH/08 Bug 1586: exiqgrep now refuses to run if there are unexpected arguments.
685 JH/09 Bug 1700: ignore space & tab embedded in base64 during decode.
687 JH/10 Bug 840: fix log_defer_output option of pipe transport
689 JH/11 Bug 830: use same host for all RCPTS of a message, even under
690 hosts_randomize. This matters a lot when combined with mua_wrapper.
692 JH/12 Bug 1706: percent and underbar characters are no longer escaped by the
693 ${quote_pgsql:<string>} operator.
695 JH/13 Bug 1708: avoid misaligned access in cached lookup.
697 JH/14 Change header file name for freeradius-client. Relevant if compiling
698 with Radius support; from the Gentoo tree and checked under Fedora.
700 JH/15 Bug 1712: Introduce $prdr_requested flag variable
702 JH/16 Bug 1714: Permit an empty string as expansion result for transport
703 option transport_filter, meaning no filtering.
705 JH/17 Bug 1713: Fix non-PDKIM_DEBUG build. Patch from Jasen Betts.
707 JH/18 Bug 1709: When built with TLS support, the tls_advertise_hosts option now
708 defaults to "*" (all hosts). The variable is now available when not built
709 with TLS, default unset, mainly to enable keeping the testsuite sane.
710 If a server certificate is not supplied (via tls_certificate) an error is
711 logged, and clients will find TLS connections fail on startup. Presumably
712 they will retry in-clear.
713 Packagers of Exim are strongly encouraged to create a server certificate
714 at installation time.
716 HS/03 Add -bP config_file as a synonym for -bP configure_file, for consistency
717 with the $config_file variable.
719 JH/19 Two additional event types: msg:rcpt:defer and msg:rcpt:host:defer. Both
720 in transport context, after the attempt, and per-recipient. The latter type
721 is per host attempted. The event data is the error message, and the errno
722 information encodes the lookup type (A vs. MX) used for the (first) host,
723 and the trailing two digits of the smtp 4xx response.
725 GF/01 Bug 1715: Fix for race condition in exicyclog, where exim could attempt
726 to write to mainlog (or rejectlog, paniclog) in the window between file
727 creation and permissions/ownership being changed. Particularly affects
728 installations where exicyclog is run as root, rather than exim user;
729 result is that the running daemon panics and dies.
731 JH/20 Bug 1701: For MySQL lookups, support MySQL config file option group names.
733 JH/21 Bug 1720: Add support for priority groups and weighted-random proxy
734 selection for the EXPERIMENTAL_SOCKS feature, via new per-proxy options
735 "pri" and "weight". Note that the previous implicit priority given by the
736 list order is no longer honoured.
738 JH/22 Bugs 963, 1721: Fix some corner cases in message body canonicalization
741 JH/23 Move SOCKS5 support from Experimental to mainline, enabled for a build
742 by defining SUPPORT_SOCKS.
744 JH/26 Move PROXY support from Experimental to mainline, enabled for a build
745 by defining SUPPORT_PROXY. Note that the proxy_required_hosts option
746 is renamed to hosts_proxy, and the proxy_{host,target}_{address,port}.
747 variables are renamed to proxy_{local,external}_{address,port}.
749 JH/27 Move Internationalisation support from Experimental to mainline, enabled
750 for a build by defining SUPPORT_I18N
752 JH/28 Bug 1745: Fix redis lookups to handle (quoted) spaces embedded in parts
753 of the query string, and make ${quote_redis:} do that quoting.
755 JH/29 Move Events support from Experimental to mainline, enabled by default
756 and removable for a build by defining DISABLE_EVENT.
758 JH/30 Updated DANE implementation code to current from Viktor Dukhovni.
760 JH/31 Fix bug with hosts_connection_nolog and named-lists which were wrongly
761 cached by the daemon.
763 JH/32 Move Redis support from Experimental to mainline, enabled for a build
764 by defining LOOKUP_REDIS. The libhiredis library is required.
766 JH/33 Bug 1748: Permit ACL dnslists= condition in non-smtp ACLs if explicit
767 keys are given for lookup.
769 JH/34 Bug 1192: replace the embedded copy of PolarSSL RSA routines in the DKIM
770 support, by using OpenSSL or GnuTLS library ones. This means DKIM is
771 only supported when built with TLS support. The PolarSSL SHA routines
772 are still used when the TLS library is too old for convenient support.
774 JH/35 Require SINGLE_DH_USE by default in OpenSSL (main config option
775 openssl_options), for security. OpenSSL forces this from version 1.1.0
776 server-side so match that on older versions.
778 JH/36 Bug 1778: longstanding bug in memory use by the ${run } expansion: A fresh
779 allocation for $value could be released as the expansion processing
780 concluded, but leaving the global pointer active for it.
782 JH/37 Bug 1769: Permit a VRFY ACL to override the default 252 response,
783 and to use the domains and local_parts ACL conditions.
785 JH/38 Fix cutthrough bug with body lines having a single dot. The dot was
786 incorrectly not doubled on cutthrough transmission, hence seen as a
787 body-termination at the receiving system - resulting in truncated mails.
788 Commonly the sender saw a TCP-level error, and retransmitted the message
789 via the normal store-and-forward channel. This could result in duplicates
790 received - but deduplicating mailstores were liable to retain only the
791 initial truncated version.
793 JH/39 Bug 1781: Fix use of DKIM private-keys having trailing '=' in the base-64.
795 JH/40 Fix crash in queryprogram router when compiled with EXPERIMENTAL_SRS.
797 JH/41 Bug 1792: Fix selection of headers to sign for DKIM: bottom-up. While
798 we're in there, support oversigning also; bug 1309.
800 JH/42 Bug 1796: Fix error logged on a malware scanner connection failure.
802 HS/04 Add support for keep_environment and add_environment options.
804 JH/43 Tidy coding issues detected by gcc --fsanitize=undefined. Some remain;
805 either intentional arithmetic overflow during PRNG, or testing config-
808 JH/44 Bug 1800: The combination of a -bhc commandline option and cutthrough
809 delivery resulted in actual delivery. Cancel cutthrough before DATA
812 JH/45 Fix cutthrough, when connection not opened by verify and target hard-
813 rejects a recipient: pass the reject to the originator.
815 JH/46 Multiple issues raised by Coverity. Some were obvious or plausible bugs.
816 Many were false-positives and ignorable, but it's worth fixing the
819 JH/47 Fix build on HP-UX and older Solaris, which need (un)setenv now also
820 for the new environment-manipulation done at startup. Move the routines
821 from being local to tls.c to being global via the os.c file.
823 JH/48 Bug 1807: Fix ${extract } for the numeric/3-string case. While preparsing
824 an extract embedded as result-arg for a map, the first arg for extract
825 is unavailable so we cannot tell if this is a numbered or keyed
826 extraction. Accept either.
832 JH/01 Bug 1545: The smtp transport option "retry_include_ip_address" is now
835 JH/02 The smtp transport option "multi_domain" is now expanded.
837 JH/03 The smtp transport now requests PRDR by default, if the server offers
840 JH/04 Certificate name checking on server certificates, when exim is a client,
841 is now done by default. The transport option tls_verify_cert_hostnames
842 can be used to disable this per-host. The build option
843 EXPERIMENTAL_CERTNAMES is withdrawn.
845 JH/05 The value of the tls_verify_certificates smtp transport and main options
846 default to the word "system" to access the system default CA bundle.
847 For GnuTLS, only version 3.0.20 or later.
849 JH/06 Verification of the server certificate for a TLS connection is now tried
850 (but not required) by default. The verification status is now logged by
851 default, for both outbound TLS and client-certificate supplying inbound
854 JH/07 Changed the default rfc1413 lookup settings to disable calls. Few
857 JH/08 The EXPERIMENTAL_DSN compile option is no longer needed; all Delivery
858 Status Notification (bounce) messages are now MIME format per RFC 3464.
859 Support for RFC 3461 DSN options NOTIFY,ENVID,RET,ORCPT can be advertised
860 under the control of the dsn_advertise_hosts option, and routers may
861 have a dsn_lasthop option.
863 JH/09 A timeout of 2 minutes is now applied to all malware scanner types by
864 default, modifiable by a malware= option. The list separator for
865 the options can now be changed in the usual way. Bug 68.
867 JH/10 The smtp_receive_timeout main option is now expanded before use.
869 JH/11 The incoming_interface log option now also enables logging of the
870 local interface on delivery outgoing connections.
872 JH/12 The cutthrough-routing facility now supports multi-recipient mails,
873 if the interface and destination host and port all match.
875 JH/13 Bug 344: The verify = reverse_host_lookup ACL condition now accepts a
878 JH/14 Bug 1573: The spam= ACL condition now additionally supports Rspamd.
879 Patch from Andrew Lewis.
881 JH/15 Bug 670: The spamd_address main option (for the spam= ACL condition)
882 now supports optional time-restrictions, weighting, and priority
883 modifiers per server. Patch originally by <rommer@active.by>.
885 JH/16 The spamd_address main option now supports a mixed list of local
886 and remote servers. Remote servers can be IPv6 addresses, and
887 specify a port-range.
889 JH/17 Bug 68: The spamd_address main option now supports an optional
890 timeout value per server.
892 JH/18 Bug 1581: Router and transport options headers_add/remove can
893 now have the list separator specified.
895 JH/19 Bug 392: spamd_address, and clamd av_scanner, now support retry
898 JH/20 Bug 1571: Ensure that $tls_in_peerdn is set, when verification fails
901 JH/21 Support for the A6 type of dns record is withdrawn.
903 JH/22 Bug 608: The result of a QUIT or not-QUIT toplevel ACL now matters
904 rather than the verbs used.
906 JH/23 Bug 1572: Increase limit on SMTP confirmation message copy size
907 from 255 to 1024 chars.
909 JH/24 Verification callouts now attempt to use TLS by default.
911 HS/01 DNSSEC options (dnssec_require_domains, dnssec_request_domains)
912 are generic router options now. The defaults didn't change.
914 JH/25 Bug 466: Add RFC2322 support for MIME attachment filenames.
915 Original patch from Alexander Shikoff, worked over by JH.
917 HS/02 Bug 1575: exigrep falls back to autodetection of compressed
918 files if ZCAT_COMMAND is not executable.
920 JH/26 Bug 1539: Add timeout/retry options on dnsdb lookups.
922 JH/27 Bug 286: Support SOA lookup in dnsdb lookups.
924 JH/28 Bug 1588: Do not use the A lookup following an AAAA for setting the FQDN.
925 Normally benign, it bites when the pair was led to by a CNAME;
926 modern usage is to not canonicalize the domain to a CNAME target
927 (and we were inconsistent anyway for A-only vs AAAA+A).
929 JH/29 Bug 1632: Removed the word "rejected" from line logged for ACL discards.
931 JH/30 Check the forward DNS lookup for DNSSEC, in addition to the reverse,
932 when evaluating $sender_host_dnssec.
934 JH/31 Check the HELO verification lookup for DNSSEC, adding new
935 $sender_helo_dnssec variable.
937 JH/32 Bug 1397: Enable ECDHE on OpenSSL, just the NIST P-256 curve.
939 JH/33 Bug 1346: Note MAIL cmd seen in -bS batch, to avoid smtp_no_mail log.
941 JH/34 Bug 1648: Fix a memory leak seen with "mailq" and large queues.
943 JH/35 Bug 1642: Fix support of $spam_ variables at delivery time. Was
944 documented as working, but never had. Support all but $spam_report.
946 JH/36 Bug 1659: Guard checking of input smtp commands again pseudo-command
947 added for tls authenticator.
949 HS/03 Add perl_taintmode main config option
955 TL/01 When running the test suite, the README says that variables such as
956 no_msglog_check are global and can be placed anywhere in a specific
957 test's script, however it was observed that placement needed to be near
958 the beginning for it to behave that way. Changed the runtest perl
959 script to read through the entire script once to detect and set these
960 variables, reset to the beginning of the script, and then run through
961 the script parsing/test process like normal.
963 TL/02 The BSD's have an arc4random API. One of the functions to induce
964 adding randomness was arc4random_stir(), but it has been removed in
965 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
966 function when detected.
968 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
969 cause callback expansion.
971 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
972 syntax errors in an expansion can be treated as a string instead of
973 logging or causing an error, due to the internal use of bool_lax
974 instead of bool when processing it.
976 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
977 server certificates when making smtp deliveries.
979 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
981 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
983 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
985 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
986 Merged patch from Sebastian Wiedenroth.
988 JH/05 Fix results-pipe from transport process. Several recipients, combined
989 with certificate use, exposed issues where response data items split
990 over buffer boundaries were not parsed properly. This eventually
991 resulted in duplicates being sent. This issue only became common enough
992 to notice due to the introduction of connection certificate information,
993 the item size being so much larger. Found and fixed by Wolfgang Breyha.
995 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
996 size buffer was used, resulting in syntax errors when an expansion
999 JH/07 Add support for directories of certificates when compiled with a GnuTLS
1000 version 3.3.6 or later.
1002 JH/08 Rename the TPDA experimental facility to Event Actions. The #ifdef
1003 is EXPERIMENTAL_EVENT, the main-configuration and transport options
1004 both become "event_action", the variables become $event_name, $event_data
1005 and $event_defer_errno. There is a new variable $verify_mode, usable in
1006 routers, transports and related events. The tls:cert event is now also
1007 raised for inbound connections, if the main configuration event_action
1010 TL/06 In test suite, disable OCSP for old versions of openssl which contained
1011 early OCSP support, but no stapling (appears to be less than 1.0.0).
1013 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
1014 server certificate names available under the smtp transport option
1015 "tls_verify_cert_hostname" now do not permit multi-component wildcard
1018 JH/10 Time-related extraction expansions from certificates now use the main
1019 option "timezone" setting for output formatting, and are consistent
1020 between OpenSSL and GnuTLS compilations. Bug 1541.
1022 JH/11 Fix a crash in mime ACL when meeting a zero-length, quoted or RFC2047-
1023 encoded parameter in the incoming message. Bug 1558.
1025 JH/12 Bug 1527: Autogrow buffer used in reading spool files. Since they now
1026 include certificate info, eximon was claiming there were spoolfile
1029 JH/13 Bug 1521: Fix ldap lookup for single-attr request, multiple-attr return.
1031 JH/14 Log delivery-related information more consistently, using the sequence
1032 "H=<name> [<ip>]" wherever possible.
1034 TL/07 Bug 1547: Omit RFCs from release. Draft and RFCs have licenses which
1035 are problematic for Debian distribution, omit them from the release
1038 JH/15 Updates and fixes to the EXPERIMENTAL_DSN feature.
1040 JH/16 Fix string representation of time values on 64bit time_t architectures.
1043 JH/17 Fix a null-indirection in certextract expansions when a nondefault
1044 output list separator was used.
1049 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
1050 checkers that were complaining about end of non-void function with no
1053 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
1054 This was a regression introduced in 4.83 by another bugfix.
1056 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
1058 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
1059 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
1065 TF/01 Correctly close the server side of TLS when forking for delivery.
1067 When a message was received over SMTP with TLS, Exim failed to clear up
1068 the incoming connection properly after forking off the child process to
1069 deliver the message. In some situations the subsequent outgoing
1070 delivery connection happened to have the same fd number as the incoming
1071 connection previously had. Exim would try to use TLS and fail, logging
1072 a "Bad file descriptor" error.
1074 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
1075 utilities have not been installed.
1077 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
1078 temporary space as the ACL may create new global variables.
1080 TL/01 LDAP support uses per connection or global context settings, depending
1081 upon the detected version of the libraries at build time.
1083 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
1084 to extract and use the src ip:port in logging and expansions as if it
1085 were a direct connection from the outside internet. PPv2 support was
1086 updated based on HAProxy spec change in May 2014.
1088 JH/02 Add ${listextract {number}{list}{success}{fail}}.
1090 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
1091 Properly escape header and check for NULL return.
1093 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
1096 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
1098 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
1099 characters in header names, implemented as a verify condition.
1100 Contributed by Michael Fischer v. Mollard.
1102 TL/05 Rename SPF condition results err_perm and err_temp to standardized
1103 results permerror and temperror. Previous values are deprecated but
1104 still accepted. In a future release, err_perm and err_temp will be
1105 completely removed, which will be a backward incompatibility if the
1106 ACL tests for either of these two old results. Patch contributed by
1107 user bes-internal on the mailing list.
1109 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
1111 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
1112 selectors, in both main and reject logs.
1114 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
1117 JH/07 Add malware type "sock" for talking to simple daemon.
1119 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
1121 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
1122 routers/transports under cutthrough routing.
1124 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
1125 numbers. Touch up "bool" conditional to keep the same definition.
1127 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
1129 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
1131 JH/12 Expand items in router/transport headers_add or headers_remove lists
1132 individually rather than the list as a whole. Bug 1452.
1134 Required for reasonable handling of multiple headers_ options when
1135 they may be empty; requires that headers_remove items with embedded
1136 colons must have them doubled (or the list-separator changed).
1138 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
1139 view the policy declared in the DMARC record. Currently, $dmarc_status
1140 is a combined value of both the record presence and the result of the
1143 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
1145 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
1146 dnslookup router and the smtp transport (applying to the forward
1149 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
1150 of ldap servers used for a specific lookup. Patch provided by Heiko
1153 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
1154 New variable $lookup_dnssec_authenticated for observability.
1156 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
1157 Patch submitted by Lars Timman.
1159 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
1161 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
1162 Requires trusted mode and valid format message id, aborts otherwise.
1163 Patch contributed by Heiko Schlichting.
1165 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
1166 certextract with support for various fields. Bug 1358.
1168 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
1169 is requested by default, modifiable by smtp transport option
1172 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
1173 operate on certificate variables to give certificate fingerprints
1174 Also new ${sha256:cert_variable}.
1176 JH/23 The PRDR feature is moved from being Experimental into the mainline.
1178 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
1179 Christian Aistleitner.
1181 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
1183 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
1184 file. Patch from Wolfgang Breyha.
1186 JH/25 Expand the coverage of the delivery $host and $host_address to
1187 client authenticators run in verify callout. Bug 1476.
1189 JH/26 Port service names are now accepted for tls_on_connect_ports, to
1190 align with daemon_smtp_ports. Bug 72.
1192 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
1193 support and error reporting did not work properly.
1195 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
1196 and is readable. Patch from Andrew Colin Kissa.
1198 TL/14 Enhance documentation of ${run expansion and how it parses the
1199 commandline after expansion, particularly in the case when an
1200 unquoted variable expansion results in an empty value.
1202 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
1204 PP/02 Fix internal collision of T_APL on systems which support RFC3123
1205 by renaming away from it. Addresses GH issue 15, reported by
1208 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
1210 TL/15 SECURITY: prevent double expansion in math comparison functions
1211 (can expand unsanitized data). Not remotely exploitable.
1218 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
1220 PP/02 Make -n do something, by making it not do something.
1221 When combined with -bP, the name of an option is not output.
1223 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
1226 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
1227 $sender_host_name and config options to manage this, and basic check
1230 PP/05 DSCP support for outbound connections and control modifier for inbound.
1232 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
1233 (Only plugin which currently uses this is kerberos4, which nobody should
1234 be using, but we should make it available and other future plugins might
1235 conceivably use it, even though it would break NAT; stuff *should* be
1236 using channel bindings instead).
1238 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
1239 name; added for Sendmail compatibility; requires admin caller.
1240 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
1241 just ignore it); requires trusted caller.
1242 Also parse but ignore: -Ac -Am -X<logfile>
1245 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
1247 TL/02 Add +smtp_confirmation as a default logging option.
1249 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
1250 Patch by Magnus Holmgren from 2007-02-20.
1252 TL/04 Bugzilla 1281 - Spec typo.
1253 Bugzilla 1283 - Spec typo.
1254 Bugzilla 1290 - Spec grammar fixes.
1256 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
1258 TL/06 Add Experimental DMARC support using libopendmarc libraries.
1260 TL/07 Fix an out of order global option causing a segfault. Reported to dev
1261 mailing list by by Dmitry Isaikin.
1263 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
1265 JH/02 Support "G" suffix to numbers in ${if comparisons.
1267 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
1269 NM/01 Bugzilla 1197 - Spec typo
1270 Bugzilla 1196 - Spec examples corrections
1272 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
1274 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
1275 gnutls_enable_pkcs11, but renamed to more accurately indicate its
1278 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
1279 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
1281 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
1282 "acl {{name}{arg}...}", and optional args on acl condition
1285 JH/05 Permit multiple router/transport headers_add/remove lines.
1287 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
1289 JH/07 Avoid using a waiting database for a single-message-only transport.
1290 Performance patch from Paul Fisher. Bugzilla 1262.
1292 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
1295 JH/09 Add $headers_added variable, with content from use of ACL modifier
1296 add_header (but not yet added to the message). Bugzilla 199.
1298 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
1299 Pulled from Bugzilla 817 by Wolfgang Breyha.
1301 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
1303 (nb: this is the same fix as in Exim 4.80.1)
1305 JH/11 Add A= logging on delivery lines, and a client_set_id option on
1308 JH/12 Add optional authenticated_sender logging to A= and a log_selector
1311 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
1313 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
1314 advertise SMTP AUTH mechanism to us, instead of a generic
1315 protocol violation error. Also, make Exim more robust to bad
1316 data from the Dovecot auth socket.
1318 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
1320 When a queue runner is handling a message, Exim first routes the
1321 recipient addresses, during which it prunes them based on the retry
1322 hints database. After that it attempts to deliver the message to
1323 any remaining recipients. It then updates the hints database using
1326 So if a recipient address works intermittently, it can get repeatedly
1327 deferred at routing time. The retry hints record remains fresh so the
1328 address never reaches the final cutoff time.
1330 This is a fairly common occurrence when a user is bumping up against
1331 their storage quota. Exim had some logic in its local delivery code
1332 to deal with this. However it did not apply to per-recipient defers
1333 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
1335 This change adds a proper retry rule check during routing so that the
1336 final cutoff time is checked against the message's age. We only do
1337 this check if there is an address retry record and there is not a
1338 domain retry record; this implies that previous attempts to handle
1339 the address had the retry_use_local_parts option turned on. We use
1340 this as an approximation for the destination being like a local
1341 delivery, as in LMTP.
1343 I suspect this new check makes the old local delivery cutoff check
1344 redundant, but I have not verified this so I left the code in place.
1346 TF/02 Correct gecos expansion when From: is a prefix of the username.
1348 Test 0254 submits a message to Exim with the header
1352 When I ran the test suite under the user fanf2, Exim expanded
1353 the header to contain my full name, whereas it should have added
1354 a Resent-Sender: header. It erroneously treats any prefix of the
1355 username as equal to the username.
1357 This change corrects that bug.
1359 GF/01 DCC debug and logging tidyup
1360 Error conditions log to paniclog rather than rejectlog.
1361 Debug lines prefixed by "DCC: " to remove any ambiguity.
1363 TF/03 Avoid unnecessary rebuilds of lookup-related code.
1365 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
1366 Bug spotted by Jeremy Harris; was flawed since initial commit.
1367 Would have resulted in OCSP responses post-SNI triggering an Exim
1368 NULL dereference and crash.
1370 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
1372 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
1373 Bug detection, analysis and fix by Samuel Thibault.
1374 Bugzilla 1331, Debian bug #698092.
1376 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
1378 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
1379 Server implementation by Todd Lyons, client by JH.
1380 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
1381 config variable "prdr_enable" controls whether the server
1382 advertises the facility. If the client requests PRDR a new
1383 acl_data_smtp_prdr ACL is called once for each recipient, after
1384 the body content is received and before the acl_smtp_data ACL.
1385 The client is controlled by both of: a hosts_try_prdr option
1386 on the smtp transport, and the server advertisement.
1387 Default client logging of deliveries and rejections involving
1388 PRDR are flagged with the string "PRDR".
1390 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
1391 fclose(). Diagnosis by Todd Lyons.
1393 PP/17 Update configure.default to handle IPv6 localhost better.
1394 Patch by Alain Williams (plus minor tweaks).
1397 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
1398 This is now consistent with GnuTLS, and is now documented: the
1399 previous undocumented portable approach to treating the option as
1400 unset was to force an expansion failure. That still works, and
1401 an empty string is now equivalent.
1403 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
1404 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
1405 not performing validation itself.
1407 PP/20 Added force_command boolean option to pipe transport.
1408 Patch from Nick Koston, of cPanel Inc.
1410 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
1413 TF/04 Added udpsend ACL modifier and hexquote expansion operator
1415 PP/21 Fix eximon continuous updating with timestamped log-files.
1416 Broken in a format-string cleanup in 4.80, missed when I repaired the
1417 other false fix of the same issue.
1418 Report and fix from Heiko Schlichting.
1421 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
1422 Report from Prashanth Katuri.
1424 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
1425 It's SecureTransport, so affects any MacOS clients which use the
1426 system-integrated TLS libraries, including email clients.
1428 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
1429 using a MIME ACL for non-SMTP local injection.
1430 Report and assistance in diagnosis by Warren Baker.
1432 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
1434 JH/16 Fix comparisons for 64b. Bugzilla 1385.
1436 TL/09 Add expansion variable $authenticated_fail_id to keep track of
1437 last id that failed so it may be referenced in subsequent ACL's.
1439 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
1442 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
1443 ldap library initialization, allowing self-signed CA's to be
1444 used. Also properly sets require_cert option later in code by
1445 using NULL (global ldap config) instead of ldap handle (per
1446 session). Bug diagnosis and testing by alxgomz.
1448 TL/12 Enhanced documentation in the ratelimit.pl script provided in
1449 the src/util/ subdirectory.
1451 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
1452 renamed to Transport Post Delivery Action by Jeremy Harris, as
1455 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
1456 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
1457 redis_servers = needs to be configured which will be used by the redis
1458 lookup. Patch from Warren Baker, of The Packet Hub.
1460 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
1462 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
1463 hostname or reverse DNS when processing a host list. Used suggestions
1464 from multiple comments on this bug.
1466 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
1468 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
1469 Missed a few lines, added it to make the runtest require no keyboard
1472 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
1473 contains upper case chars. Make router use caseful_local_part.
1475 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
1476 support when GnuTLS has been built with p11-kit.
1482 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
1484 This, or similar/improved, will also be change PP/11 of 4.82.
1490 PP/01 Handle short writes when writing local log-files.
1491 In practice, only affects FreeBSD (8 onwards).
1492 Bugzilla 1053, with thanks to Dmitry Isaikin.
1494 NM/01 Bugzilla 949 - Documentation tweak
1496 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
1499 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
1501 PP/02 Implemented gsasl authenticator.
1503 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
1505 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
1506 `pkg-config foo` for cflags/libs.
1508 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
1509 with rest of GSASL and with heimdal_gssapi.
1511 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
1512 `pkg-config foo` for cflags/libs for the TLS implementation.
1514 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
1515 properties get this fed in as external SSF. A number of robustness
1516 and debugging improvements to the cyrus_sasl authenticator.
1518 PP/08 cyrus_sasl server now expands the server_realm option.
1520 PP/09 Bugzilla 1214 - Log authentication information in reject log.
1521 Patch by Jeremy Harris.
1523 PP/10 Added dbmjz lookup type.
1525 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
1527 PP/12 MAIL args handles TAB as well as SP, for better interop with
1528 non-compliant senders.
1529 Analysis and variant patch by Todd Lyons.
1531 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
1532 Bug report from Lars Müller <lars@samba.org> (via SUSE),
1533 Patch from Dirk Mueller <dmueller@suse.com>
1535 PP/13 tls_peerdn now print-escaped for spool files.
1536 Observed some $tls_peerdn in wild which contained \n, which resulted
1537 in spool file corruption.
1539 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
1540 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
1541 or write after TLS renegotiation, which otherwise led to messages
1544 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
1545 as a tracking header (ie: a signed header comes before the signature).
1546 Patch from Wolfgang Breyha.
1548 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
1549 comma-sep list; embedded commas doubled.
1551 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
1553 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
1555 Report and patch from Dmitry Banschikov.
1557 PP/16 Removed "dont_insert_empty_fragments" from "openssl_options".
1558 Removed SSL_clear() after SSL_new() which led to protocol negotiation
1559 failures. We appear to now support TLS1.1+ with Exim.
1561 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
1562 lets Exim select keys and certificates based upon TLS SNI from client.
1563 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
1564 before an outbound SMTP session. New log_selector, +tls_sni.
1566 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
1567 NULL dereference. Report and patch from Alun Jones.
1569 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
1570 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
1571 Not seeing resolver debug output on NetBSD, but suspect this is a
1572 resolver implementation change.
1574 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
1575 Left warnings. Added "eximon gdb" invocation mode.
1577 PP/21 Defaulting "accept_8bitmime" to true, not false.
1579 PP/22 Added -bw for inetd wait mode support.
1581 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
1582 locate the relevant includes and libraries. Made this the default.
1584 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
1585 Bugzilla 1246, report and most of solution from Tomasz Kusy.
1587 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
1588 This may cause build issues on older platforms.
1590 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
1591 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
1592 gnutls_require_mac & gnutls_require_protocols (no longer supported).
1593 Added SNI support via GnuTLS too.
1594 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
1596 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
1598 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
1599 Applied second patch from Janne, implementing suggestion to default
1600 multiple-strings-in-record handling to match SPF spec.
1602 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
1604 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
1605 read-only, out of scope).
1606 Patch from Wolfgang Breyha, report from Stuart Northfield.
1608 PP/29 Fix three issues highlighted by clang analyser static analysis.
1609 Only crash-plausible issue would require the Cambridge-specific
1610 iplookup router and a misconfiguration.
1611 Report from Marcin Mirosław.
1613 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
1615 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
1616 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
1617 As part of this, removing so much warning spew let me fix some minor
1618 real issues in debug logging.
1620 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
1621 assignment on my part. Fixed.
1623 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
1624 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
1625 Janne Snabb (who went above and beyond: thank you).
1627 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
1628 string otherwise requires a connection and a bunch more work and it's
1629 relatively easy to get wrong. Should also expose TLS library linkage
1632 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
1633 64-bit ${eval} (JH/03).
1635 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
1636 GNU libc to support some of the 64-bit stuff, should not lead to
1637 conflicts. Defined before os.h is pulled in, so if a given platform
1638 needs to override this, it can.
1640 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
1641 protection layer was required, which is not implemented.
1642 Bugzilla 1254, patch from Wolfgang Breyha.
1644 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
1645 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
1646 tls_dhparam take prime identifiers. Also unbreak combination of
1647 OpenSSL+DH_params+TLSSNI.
1649 PP/39 Disable SSLv2 by default in OpenSSL support.
1655 PP/01 Solaris build fix for Oracle's LDAP libraries.
1656 Bugzilla 1109, patch from Stephen Usher.
1658 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
1660 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
1663 TF/02 Fix a couple more cases where we did not log the error message
1664 when unlink() failed. See also change 4.74-TF/03.
1666 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
1667 lock up or crash if it happened to be inside a call to libc when it
1668 got a SIGUSR1 from exiwhat.
1670 The SIGUSR1 handler appends the current process status to the process
1671 log which is later printed by exiwhat. It used to use the general
1672 purpose logging code to do this, but several functions it calls are
1673 not safe for signals.
1675 The new output code in the SIGUSR1 handler is specific to the process
1676 log, and simple enough that it's easy to inspect for signal safety.
1677 Removing some special cases also simplifies the general logging code.
1678 Removing the spurious timestamps from the process log simplifies
1681 TF/04 Improved ratelimit ACL condition.
1683 The /noupdate option has been deprecated in favour of /readonly which
1684 has clearer semantics. The /leaky, /strict, and /readonly update modes
1685 are mutually exclusive. The update mode is no longer included in the
1686 database key; it just determines when the database is updated. (This
1687 means that when you upgrade Exim will forget old rate measurements.)
1689 Exim now checks that the per_* options are used with an update mode that
1690 makes sense for the current ACL. For example, when Exim is processing a
1691 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
1692 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
1693 must specify per_mail/readonly. If you omit the update mode it defaults to
1694 /leaky where that makes sense (as before) or /readonly where required.
1696 The /noupdate option is now undocumented but still supported for
1697 backwards compatibility. It is equivalent to /readonly except that in
1698 ACLs where /readonly is required you may specify /leaky/noupdate or
1699 /strict/noupdate which are treated the same as /readonly.
1701 A useful new feature is the /count= option. This is a generalization
1702 of the per_byte option, so that you can measure the throughput of other
1703 aggregate values. For example, the per_byte option is now equivalent
1704 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
1706 The per_rcpt option has been generalized using the /count= mechanism
1707 (though it's more complicated than the per_byte equivalence). When it is
1708 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
1709 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
1710 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
1711 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
1712 Note that using per_rcpt with a non-readonly update mode in more than
1713 one ACL will cause the recipients to be double-counted. (The per_mail
1714 and per_byte options don't have this problem.)
1716 The handling of very low rates has changed slightly. If the computed rate
1717 is less than the event's count (usually one) then this event is the first
1718 after a long gap. In this case the rate is set to the same as this event's
1719 count, so that the first message of a spam run is counted properly.
1721 The major new feature is a mechanism for counting the rate of unique
1722 events. The new per_addr option counts the number of different
1723 recipients that someone has sent messages to in the last time period. It
1724 behaves like per_rcpt if all the recipient addresses are different, but
1725 duplicate recipient addresses do not increase the measured rate. Like
1726 the /count= option this is a general mechanism, so the per_addr option
1727 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
1728 example, measure the rate that a client uses different sender addresses
1729 with the options per_mail/unique=$sender_address. There are further
1730 details in the main documentation.
1732 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
1734 TF/06 Removed a few PCRE remnants.
1736 TF/07 Automatically extract Exim's version number from tags in the git
1737 repository when doing development or release builds.
1739 PP/02 Raise smtp_cmd_buffer_size to 16kB.
1740 Bugzilla 879. Patch from Paul Fisher.
1742 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
1743 Heavily based on revision 40f9a89a from Simon Arlott's tree.
1746 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
1748 PP/05 Variable $av_failed, true if the AV scanner deferred.
1749 Bugzilla 1078. Patch from John Horne.
1751 PP/06 Stop make process more reliably on build failure.
1752 Bugzilla 1087. Patch from Heiko Schlittermann.
1754 PP/07 Make maildir_use_size_file an _expandable_ boolean.
1755 Bugzilla 1089. Patch from Heiko Schlittermann.
1757 PP/08 Handle ${run} returning more data than OS pipe buffer size.
1758 Bugzilla 1131. Patch from Holger Weiß.
1760 PP/09 Handle IPv6 addresses with SPF.
1761 Bugzilla 860. Patch from Wolfgang Breyha.
1763 PP/10 GnuTLS: support TLS 1.2 & 1.1.
1765 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
1768 PP/11 match_* no longer expand right-hand-side by default.
1769 New compile-time build option, EXPAND_LISTMATCH_RHS.
1770 New expansion conditions, "inlist", "inlisti".
1772 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
1774 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
1776 PP/14 fix log_write() format string regression from TF/03.
1777 Bugzilla 1152. Patch from Dmitry Isaikin.
1783 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
1785 PP/02 Harmonised TLS library version reporting; only show if debugging.
1786 Layout now matches that introduced for other libraries in 4.74 PP/03.
1788 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
1790 PP/04 New "dns_use_edns0" global option.
1792 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
1795 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
1796 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
1798 TK/01 Updated PolarSSL code to 0.14.2.
1799 Bugzilla 1097. Patch from Andreas Metzler.
1801 PP/07 Catch divide-by-zero in ${eval:...}.
1802 Fixes bugzilla 1102.
1804 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
1807 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
1808 format-string attack -- SECURITY: remote arbitrary code execution.
1810 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
1811 time unintentionally subject to list matching rules, letting the header
1812 cause arbitrary Exim lookups (of items which can occur in lists, *not*
1813 arbitrary string expansion). This allowed for information disclosure.
1815 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
1816 INT_MIN/-1 -- value coerced to INT_MAX.
1822 NM/01 Workaround for PCRE version dependency in version reporting
1825 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
1826 This fixes portability to compilers other than gcc, notably
1827 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
1829 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
1830 makefiles for portability to HP-UX and POSIX correctness.
1832 PP/01 Permit LOOKUP_foo enabling on the make command-line.
1833 Also via indented variable definition in the Makefile.
1834 (Debugging by Oliver Heesakkers).
1836 PP/02 Restore caching of spamd results with expanded spamd_address.
1837 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
1839 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
1840 Improves build reliability. Fix from: Frank Elsner
1842 NM/02 Fix wide character breakage in the rfc2047 coding
1843 Fixes bug 1064. Patch from Andrey N. Oktyabrski
1845 NM/03 Allow underscore in dnslist lookups
1846 Fixes bug 1026. Patch from Graeme Fowler
1848 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
1849 Code patches from Adam Ciarcinski of NetBSD.
1851 NM/04 Fixed exiqgrep to cope with mailq missing size issue
1854 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
1855 is logged, to avoid truncation. Patch from John Horne.
1857 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
1858 Patch from Jakob Hirsch.
1860 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
1861 SQL string expansion failure details.
1862 Patch from Andrey Oktyabrski.
1864 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
1865 Patch from Simon Arlott.
1867 PP/09 New lookups functionality failed to compile on old gcc which rejects
1868 extern declarations in function scope.
1869 Patch from Oliver Fleischmann
1871 PP/10 Use sig_atomic_t for flags set from signal handlers.
1872 Check getgroups() return and improve debugging.
1873 Fixed developed for diagnosis in bug 927 (which turned out to be
1876 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
1877 Patch from Mark Zealey.
1879 PP/12 Bugzilla 1056: Improved spamd server selection.
1880 Patch from Mark Zealey.
1882 PP/13 Bugzilla 1086: Deal with maildir quota file races.
1883 Based on patch from Heiko Schlittermann.
1885 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
1886 Patch from Uwe Doering, sign-off by Michael Haardt.
1888 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
1889 variable declaration deep within a block. Bug and patch from
1892 PP/15 lookups-Makefile IRIX compatibility coercion.
1894 PP/16 Make DISABLE_DKIM build knob functional.
1896 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
1897 Patch by Simon Arlott
1899 TF/03 Fix valgrind.h portability to C89 compilers that do not support
1900 variable argument macros. Our copy now differs from upstream.
1906 TF/01 Failure to get a lock on a hints database can have serious
1907 consequences so log it to the panic log.
1909 TF/02 Log LMTP confirmation messages in the same way as SMTP,
1910 controlled using the smtp_confirmation log selector.
1912 TF/03 Include the error message when we fail to unlink a spool file.
1914 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
1915 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
1916 for maintaining out-of-tree patches for some time.
1918 PP/01 Bugzilla 139: Documentation and portability issues.
1919 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
1920 Handle per-OS dynamic-module compilation flags.
1922 PP/02 Let /dev/null have normal permissions.
1923 The 4.73 fixes were a little too stringent and complained about the
1924 permissions on /dev/null. Exempt it from some checks.
1925 Reported by Andreas M. Kirchwitz.
1927 PP/03 Report version information for many libraries, including
1928 Exim version information for dynamically loaded libraries. Created
1929 version.h, now support a version extension string for distributors
1930 who patch heavily. Dynamic module ABI change.
1932 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
1933 privilege escalation vulnerability whereby the Exim run-time user
1934 can cause root to append content of the attacker's choosing to
1937 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
1940 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
1941 If dropping privileges for untrusted macros, we disabled normal logging
1942 on the basis that it would fail; for the Exim run-time user, this is not
1943 the case, and it resulted in successful deliveries going unlogged.
1944 Fixed. Reported by Andreas Metzler.
1950 PP/01 Date: & Message-Id: revert to normally being appended to a message,
1951 only prepend for the Resent-* case. Fixes regression introduced in
1952 Exim 4.70 by NM/22 for Bugzilla 607.
1954 PP/02 Include check_rfc2047_length in configure.default because we're seeing
1955 increasing numbers of administrators be bitten by this.
1957 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
1959 PP/03 Bugzilla 994: added openssl_options main configuration option.
1961 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
1963 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
1965 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
1967 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
1968 without a peer certificate, leading to a segfault because of an
1969 assumption that peers always have certificates. Be a little more
1970 paranoid. Problem reported by Martin Tscholak.
1972 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
1973 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
1974 NB: ClamAV planning to remove STREAM in "middle of 2010".
1975 CL also introduces -bmalware, various -d+acl logging additions and
1976 more caution in buffer sizes.
1978 PP/09 Implemented reverse_ip expansion operator.
1980 PP/10 Bugzilla 937: provide a "debug" ACL control.
1982 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
1984 PP/12 Bugzilla 973: Implement --version.
1986 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
1988 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
1990 PP/15 Bugzilla 816: support multiple condition rules on Routers.
1992 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
1993 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
1994 ignore trailing whitespace.
1996 JJ/02 prevent non-panic DKIM error from being sent to paniclog
1998 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
2001 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
2002 Notification from Dr Andrew Aitchison.
2004 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
2005 ExtendedDetectionInfo response format.
2006 Notification from John Horne.
2008 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
2011 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
2012 XSL and documented dependency on system catalogs, with examples of how
2015 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
2018 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
2019 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
2020 configuration file which is writeable by the Exim user or group.
2022 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
2023 of configuration files to cover files specified with the -C option if
2024 they are going to be used with root privileges, not just the default
2027 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
2028 option (effectively making it always true).
2030 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
2031 files to be used while preserving root privileges.
2033 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
2034 that rogue child processes cannot use them.
2036 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
2037 run-time user, instead of root.
2039 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
2040 Exim run-time user without dropping privileges.
2042 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
2043 result string, instead of calling string_vformat() twice with the same
2046 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
2047 for other users. Others should always drop root privileges if they use
2048 -C on the command line, even for a whitelisted configure file.
2050 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
2052 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
2058 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
2059 $data_path, and $header_path variables; fixed documentation bugs and
2062 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
2063 exipick to access non-standard spools, including the "frozen" queue
2066 NM/01 Bugzilla 965: Support mysql stored procedures.
2067 Patch from Alain Williams
2069 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
2071 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
2072 Patch from Andreas Metzler
2074 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
2075 Patch from Kirill Miazine
2077 NM/05 Bugzilla 671: Added umask to procmail example.
2079 JJ/03 installed exipick 20100323.0, fixing doc bug
2081 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
2082 directory. Notification and patch from Dan Rosenberg.
2084 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
2086 TK/02 Improve log output when DKIM signing operation fails.
2088 MH/01 Treat the transport option dkim_domain as a colon separated
2089 list, not as a single string, and sign the message with each element,
2090 omitting multiple occurences of the same signer.
2092 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
2093 Bugzilla 985, 986. Patch by Simon Arlott
2095 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
2096 Patch by Simon Arlott
2098 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
2099 MBX locking. Notification from Dan Rosenberg.
2105 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
2107 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
2109 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
2111 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
2113 NM/04 Bugzilla 915: Flag broken perl installation during build.
2119 TK/01 Added patch by Johannes Berg that expands the main option
2120 "spamd_address" if it starts with a dollar sign.
2122 TK/02 Write list of recipients to X-Envelope-Sender header when building
2123 the mbox-format spool file for content scanning (suggested by Jakob
2126 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
2127 (http://www.dcc-servers.net/) support via dccifd. Activated by
2128 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
2130 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
2131 by Mark Daniel Reidel <mr@df.eu>.
2133 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
2134 When building exim an external PCRE library is now needed -
2135 PCRE is a system library on the majority of modern systems.
2136 See entry on PCRE_LIBS in EDITME file.
2138 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
2139 conversation. Added nologin parameter to request.
2140 Patch contributed by Kirill Miazine.
2142 TF/01 Do not log submission mode rewrites if they do not change the address.
2144 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
2146 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
2147 log files in place. Contributed by Roberto Lima.
2149 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
2151 TF/03 Bugzilla 615: When checking the local_parts router precondition
2152 after a local_part_suffix or local_part_prefix option, Exim now
2153 does not use the address's named list lookup cache, since this
2154 contains cached lookups for the whole local part.
2156 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
2157 Robert Millan. Documentation is in experimental-spec.txt.
2159 TF/04 Bugzilla 668: Fix parallel build (make -j).
2161 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
2163 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
2164 Patch provided by Jan Srzednicki.
2166 TF/05 Leading white space used to be stripped from $spam_report which
2167 wrecked the formatting. Now it is preserved.
2169 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
2170 that they are available at delivery time.
2172 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
2174 TF/08 TLS error reporting now respects the incoming_interface and
2175 incoming_port log selectors.
2177 TF/09 Produce a more useful error message if an SMTP transport's hosts
2178 setting expands to an empty string.
2180 NM/06 Bugzilla 744: EXPN did not work under TLS.
2181 Patch provided by Phil Pennock.
2183 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
2184 Patch provided by Richard Godbee.
2186 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
2187 acl_smtp_notquit, added index entry.
2189 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
2190 Patch provided by Eugene Bujak.
2192 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
2193 accept(). Patch provided by Maxim Dounin.
2195 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
2196 Patch provided by Phil Pennock.
2198 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
2200 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
2201 Patch provided by Brad "anomie" Jorsch.
2203 NM/14 Bugzilla 622: Added timeout setting to transport filter.
2204 Patch provided by Dean Brooks.
2206 TK/05 Add native DKIM support (does not depend on external libraries).
2208 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
2209 Patch provided by Graeme Fowler.
2211 NM/16 Bugzilla 851: Documentation example syntax fix.
2213 NM/17 Changed NOTICE file to remove references to embedded PCRE.
2215 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
2218 NM/19 Bugzilla 745: TLS version reporting.
2219 Patch provided by Phil Pennock.
2221 NM/20 Bugzilla 167: bool: condition support.
2222 Patch provided by Phil Pennock.
2224 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
2225 clients. Patch provided by Phil Pennock.
2227 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
2228 Patch provided by Brad "anomie" Jorsch.
2230 NM/23 Bugzilla 687: Fix misparses in eximstats.
2231 Patch provided by Heiko Schlittermann.
2233 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
2234 Patch provided by Heiko Schlittermann.
2236 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
2237 plus update to original patch.
2239 NM/26 Bugzilla 799: Documentation correction for ratelimit.
2241 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
2242 Patch provided by David Brownlee.
2244 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
2246 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
2248 NM/30 Bugzilla 888: TLS documentation bugfixes.
2250 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
2252 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
2253 Unlike the original bugzilla I have changed all shell scripts in src tree.
2255 NM/33 Bugzilla 898: Transport filter timeout fix.
2256 Patch by Todd Rinaldo.
2258 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
2259 Patch by Serge Demonchaux.
2261 NM/35 Bugzilla 39: Base64 decode bug fixes.
2262 Patch by Jakob Hirsch.
2264 NM/36 Bugzilla 909: Correct connect() call in dcc code.
2266 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
2268 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
2270 NM/39 Bugzilla 911: Fixed MakeLinks build script.
2276 TK/01 Add preliminary DKIM support. Currently requires a forked version of
2277 ALT-N's libdkim that I have put here:
2278 http://duncanthrax.net/exim-experimental/
2280 Note to Michael Haardt: I had to rename some vars in sieve.c. They
2281 were called 'true' and it seems that C99 defines that as a reserved
2282 keyword to be used with 'bool' variable types. That means you could
2283 not include C99-style headers which use bools without triggering
2284 build errors in sieve.c.
2286 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
2287 as mailq or other aliases. Changed the --help handling significantly
2288 to do whats expected. exim_usage() emits usage/help information.
2290 SC/01 Added the -bylocaldomain option to eximstats.
2292 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
2294 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
2296 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
2302 PH/01 Another patch from the Sieve maintainer.
2304 PH/02 When an IPv6 address is converted to a string for single-key lookup
2305 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
2306 dots are used instead of colons so that keys in lsearch files need not
2307 contain colons. This was done some time before quoting was made available
2308 in lsearch files. However, iplsearch files do require colons in IPv6 keys
2309 (notated using the quote facility) so as to distinguish them from IPv4
2310 keys. This meant that lookups for IP addresses in host lists did not work
2311 for iplsearch lookups.
2313 This has been fixed by arranging for IPv6 addresses to be expressed with
2314 colons if the lookup type is iplsearch. This is not incompatible, because
2315 previously such lookups could never work.
2317 The situation is now rather anomalous, since one *can* have colons in
2318 ordinary lsearch keys. However, making the change in all cases is
2319 incompatible and would probably break a number of configurations.
2321 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
2324 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
2325 conversion specification without a maximum field width, thereby enabling
2326 a rogue spamd server to cause a buffer overflow. While nobody in their
2327 right mind would setup Exim to query an untrusted spamd server, an
2328 attacker that gains access to a server running spamd could potentially
2329 exploit this vulnerability to run arbitrary code as the Exim user.
2331 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
2332 $primary_hostname instead of what libspf2 thinks the hosts name is.
2334 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
2335 a directory entry by the name of the lookup key. Previously, if a
2336 symlink pointed to a non-existing file or a file in a directory that
2337 Exim lacked permissions to read, a lookup for a key matching that
2338 symlink would fail. Now it is enough that a matching directory entry
2339 exists, symlink or not. (Bugzilla 503.)
2341 PH/03 The body_linecount and body_zerocount variables are now exported in the
2344 PH/04 Added the $dnslist_matched variable.
2346 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
2347 This means they are set thereafter only if the connection becomes
2350 PH/06 Added the client_condition to authenticators so that some can be skipped
2351 by clients under certain conditions.
2353 PH/07 The error message for a badly-placed control=no_multiline_responses left
2354 "_responses" off the end of the name.
2356 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
2358 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
2359 (without spaces) instead of just copying the configuration text.
2361 PH/10 Added the /noupdate option to the ratelimit ACL condition.
2363 PH/11 Added $max_received_linelength.
2365 PH/12 Added +ignore_defer and +include_defer to host lists.
2367 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
2368 way in which PCRE > 7.0 is built.
2370 PH/14 Implemented queue_only_load_latch.
2372 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
2373 MAIL command. The effect was to mangle the value on 64-bit systems.
2375 PH/16 Another patch from the Sieve maintainer.
2377 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
2379 PH/18 If a system quota error occurred while trying to create the file for
2380 a maildir delivery, the message "Mailbox is full" was not appended to the
2381 bounce if the delivery eventually timed out. Change 4.67/27 below applied
2382 only to a quota excession during the actual writing of the file.
2384 PH/19 It seems that peer DN values may contain newlines (and other non-printing
2385 characters?) which causes problems in log lines. The DN values are now
2386 passed through string_printing() before being added to log lines.
2388 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
2389 and InterBase are left for another time.)
2391 PH/21 Added message_body_newlines option.
2393 PH/22 Guard against possible overflow in moan_check_errorcopy().
2395 PH/23 POSIX allows open() to be a macro; guard against that.
2397 PH/24 If the recipient of an error message contained an @ in the local part
2398 (suitably quoted, of course), incorrect values were put in $domain and
2399 $local_part during the evaluation of errors_copy.
2405 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
2406 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
2409 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
2410 issue a MAIL command.
2412 PH/02 In an ACL statement such as
2414 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
2416 if a client was not listed at all, or was listed with a value other than
2417 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
2418 the condition was not true (as it should be), so access was not denied.
2419 The bug was that the ! inversion was incorrectly passed on to the second
2420 item. This has been fixed.
2422 PH/03 Added additional dnslists conditions == and =& which are different from
2423 = and & when the dns lookup returns more than one IP address.
2425 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
2426 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
2428 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
2429 FSYNC, which compiles an option called disable_fsync that allows for
2430 bypassing fsync(). The documentation is heavily laced with warnings.
2432 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
2434 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
2435 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
2436 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
2437 including adding "make clean"; (3) Added -fPIC when compiling the test
2438 dynamically loaded module, to get rid of a warning.
2440 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
2441 message fails, move_frozen_messages = true and ignore_bounce_errors_after
2442 = 0s. The bug is otherwise harmless.
2444 PH/07 There was a bug in the dovecot authenticator such that the value of
2445 $auth1 could be overwritten, and so not correctly preserved, after a
2446 successful authentication. This usually meant that the value preserved by
2447 the server_setid option was incorrect.
2449 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
2451 PH/09 Installed PCRE release 7.0.
2453 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
2454 run for batched SMTP input. It is now run at the start of every message
2455 in the batch. While fixing this I discovered that the process information
2456 (output by running exiwhat) was not always getting set for -bs and -bS
2457 input. This is fixed, and it now also says "batched" for BSMTP.
2459 PH/11 Added control=no_pipelining.
2461 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
2462 patch, slightly modified), and move the expansion of helo_data till after
2463 the connection is made in the smtp transport (so it can use these
2466 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
2468 PH/14 Added log_selector = +pid.
2470 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
2472 PH/16 Add ${if forany and ${if forall.
2474 PH/17 Added dsn_from option to vary the From: line in DSNs.
2476 PH/18 Flush SMTP output before performing a callout, unless control =
2477 no_callout_flush is set.
2479 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
2480 was true (the default) a successful delivery failed to delete the retry
2481 item, thus causing premature timeout of the address. The bug is now
2484 PH/20 Added hosts_avoid_pipelining to the smtp transport.
2486 PH/21 Long custom messages for fakedefer and fakereject are now split up
2487 into multiline responses in the same way that messages for "deny" and
2488 other ACL rejections are.
2490 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
2491 with slight modification.
2493 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
2494 draft, changing the syntax and factoring some duplicate code".
2496 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
2497 for deliveries of the second and subsequent messages over the same SMTP
2500 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
2501 ${reduce, with only minor "tidies".
2503 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
2505 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
2506 expansion side effects.
2508 PH/27 When a message times out after an over-quota error from an Exim-imposed
2509 quota, the bounce message says "mailbox is full". This message was not
2510 being given when it was a system quota that was exceeded. It now should
2513 MH/03 Made $recipients available in local_scan(). local_scan() already has
2514 better access to the recipient list through recipients_list[], but
2515 $recipients can be useful in postmaster-provided expansion strings.
2517 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
2518 in the case of a MAIL command with additional options following the
2519 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
2520 were accidentally chopped off.
2522 PH/29 SMTP synchronization checks are implemented when a command is read -
2523 there is a check that no more input is waiting when there shouldn't be
2524 any. However, for some commands, a delay in an ACL can mean that it is
2525 some time before the response is written. In this time, more input might
2526 arrive, invalidly. So now there are extra checks after an ACL has run for
2527 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
2528 pipelining has not been advertised.
2530 PH/30 MH's patch to allow iscntrl() characters to be list separators.
2532 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
2533 returned in the SMTP response when smtp_return_error_details was false.
2534 This has been fixed.
2536 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
2537 instead of the C I/O that was originally supplied, because problems were
2538 reported on Solaris.
2540 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
2541 Exim which did not show up earlier: it was assuming that a call to
2542 SSL_CTX_set_info_callback() might give an error value. In fact, there is
2543 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
2544 was a macro that became an assignment, so it seemed to work. This has
2545 changed to a proper function call with a void return, hence the compile
2546 error. Exim's code has been fixed.
2548 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
2551 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
2553 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
2555 PH/37 If a message is not accepted after it has had an id assigned (e.g.
2556 because it turns out to be too big or there is a timeout) there is no
2557 "Completed" line in the log. When some messages of this type were
2558 selected by exigrep, they were listed as "not completed". Others were
2559 picked up by some special patterns. I have improved the selection
2560 criteria to be more general.
2562 PH/38 The host_find_failed option in the manualroute router can now be set
2563 to "ignore", to completely ignore a host whose IP address cannot be
2564 found. If all hosts are ignored, the behaviour is controlled by the new
2565 host_all_ignored option.
2567 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
2568 homing or because of multiple MX records with /mx) generated more than
2569 one IP address, and the following item turned out to be the local host,
2570 all the secondary addresses of the first item were incorrectly removed
2571 from the list, along with the local host and any following hosts (which
2572 is what is supposed to happen).
2574 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
2575 whoever called Exim into the -H file. In the case of the daemon it was
2576 behaving confusingly. When first started, it used values for whoever
2577 started the daemon, but after a SIGHUP it used the Exim user (because it
2578 calls itself on a restart). I have changed the code so that it now always
2581 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
2582 message are rejected with the same error (e.g. no authentication or bad
2583 sender address), and a DATA command is nevertheless sent (as can happen
2584 with PIPELINING or a stupid MUA), the error message that was given to the
2585 RCPT commands is included in the rejection of the DATA command. This is
2586 intended to be helpful for MUAs that show only the final error to their
2589 PH/42 Another patch from the Sieve maintainer.
2591 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
2592 Eximstats - Fixed some broken HTML links and added missing column headers
2594 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
2595 columns for Rejects, Temp Rejects, Ham, and Spam rows.
2597 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
2599 PH/43 Yet another patch from the Sieve maintainer.
2601 PH/44 I found a way to check for a TCP/IP connection going away before sending
2602 the response to the final '.' that terminates a message, but only in the
2603 case where the client has not sent further data following the '.'
2604 (unfortunately, this is allowed). However, in many cases there won't be
2605 any further data because there won't be any more messages to send. A call
2606 to select() can be used: if it shows that the input is "ready", there is
2607 either input waiting, or the socket has been closed. An attempt to read
2608 the next input character can distinguish the two cases. Previously, Exim
2609 would have sent an OK response which the client would never have see.
2610 This could lead to message repetition. This fix should cure that, at
2611 least in a lot of common cases.
2613 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
2614 advertised in response to EHLO.
2620 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
2621 fixed by 4.65/MH/01 (is this a record?) are fixed:
2623 (i) An empty string was always treated as zero by the numeric comparison
2624 operators. This behaviour has been restored.
2626 (ii) It is documented that the numeric comparison operators always treat
2627 their arguments as decimal numbers. This was broken in that numbers
2628 starting with 0 were being interpreted as octal.
2630 While fixing these problems I realized that there was another issue that
2631 hadn't been noticed. Values of message_size_limit (both the global option
2632 and the transport option) were treated as octal if they started with 0.
2633 The documentation was vague. These values are now always treated as
2634 decimal, and I will make that clear in the documentation.
2640 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
2641 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
2644 MH/01 Don't check that the operands of numeric comparison operators are
2645 integers when their expansion is in "skipping" mode (fixes bug
2646 introduced by 4.64-PH/07).
2648 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
2649 child addresses, Exim now panics and dies. Previously, because the count
2650 is held in a short int, deliveries were likely to be lost. As such a
2651 large number of recipients for a single message is ridiculous
2652 (performance will be very, very poor), I have chosen to impose a limit
2653 rather than extend the field.
2659 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
2660 leftover -K file (the existence of which was triggered by #402).
2661 While we were at it, introduced process PID as part of the -K
2662 filename. This should rule out race conditions when creating
2665 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
2666 processing considerably. Previous code took too long for large mails,
2667 triggering a timeout which in turn triggers #401.
2669 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
2670 in the DK code in transports.c. sendfile() is not really portable,
2671 hence the _LINUX specificness.
2673 TF/01 In the add_headers option to the mail command in an Exim filter,
2674 there was a bug that Exim would claim a syntax error in any
2675 header after the first one which had an odd number of characters
2678 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
2679 callout verification, Exim cached a "reject" for the entire domain. This
2680 is correct for most verifications, but it is not correct for a recipient
2681 verification with use_sender or use_postmaster set, because in that case
2682 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
2683 case of MAIL FROM:<> rejection from other early rejections (e.g.
2684 rejection of HELO). When verifying a recipient using a non-null MAIL
2685 address, the cache is ignored if it shows MAIL FROM:<> rejection.
2686 Whatever the result of the callout, the value of the domain cache is
2687 left unchanged (for any other kind of callout, getting as far as trying
2688 RCPT means that the domain itself is ok).
2690 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
2693 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
2694 manifest itself as EPIPE rather than ECONNECT. When tidying away a
2695 session, the daemon ignores ECONNECT errors and logs others; it now
2696 ignores EPIPE as well.
2698 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
2699 (quoted-printable decoding).
2701 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
2702 later the small subsequent patch to fix an introduced bug.
2704 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
2706 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
2708 PH/08 An error is now given if message_size_limit is specified negative.
2710 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
2711 to be given (somewhat) arbitrary names.
2713 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
2716 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
2717 miscellaneous code fixes
2719 PH/10 Added the log_reject_target ACL modifier to specify where to log
2722 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
2723 hostname. This is wrong, because it relates to the incoming message (and
2724 probably the interface on which it is arriving) and not to the outgoing
2725 callout (which could be using a different interface). This has been
2726 changed to use the value of the helo_data option from the smtp transport
2727 instead - this is what is used when a message is actually being sent. If
2728 there is no remote transport (possible with a router that sets up host
2729 addresses), $smtp_active_hostname is used.
2731 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
2732 tweaks were necessary in order to get it to work (see also 21 below):
2733 (a) The code assumed that strncpy() returns a negative number on buffer
2734 overflow, which isn't the case. Replaced with Exim's string_format()
2736 (b) There were several signed/unsigned issues. I just did the minimum
2737 hacking in of casts. There is scope for a larger refactoring.
2738 (c) The code used strcasecmp() which is not a standard C function.
2739 Replaced with Exim's strcmpic() function.
2740 (d) The code set only $1; it now sets $auth1 as well.
2741 (e) A simple test gave the error "authentication client didn't specify
2742 service in request". It would seem that Dovecot has changed its
2743 interface. Fortunately there's a specification; I followed it and
2744 changed what the client sends and it appears to be working now.
2746 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
2749 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
2750 address A is aliased to B and C, where B exists and C does not. Without
2751 -v the output is "A verified" because verification stops after a
2752 successful redirection if more than one address is generated. However,
2753 with -v the child addresses are also verified. Exim was outputting "A
2754 failed to verify" and then showing the successful verification for C,
2755 with its parentage. It now outputs "B failed to verify", showing B's
2756 parentage before showing the successful verification of C.
2758 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
2759 look up a TXT record in a specific list after matching in a combined
2762 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
2763 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
2764 they consult the DNS. I had assumed they would set it the way they
2765 wanted; and indeed my experiments on Linux seem to show that in some
2766 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
2767 To be on the safe side, however, I have now made the interface to
2768 host_find_byname() similar to host_find_bydns(), with an argument
2769 containing the DNS resolver options. The host_find_byname() function now
2770 sets these options at its start, just as host_find_bydns() does. The smtp
2771 transport options dns_qualify_single and dns_search_parents are passed to
2772 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
2773 of host_find_byname() use the default settings of RES_DEFNAMES
2774 (qualify_single) but not RES_DNSRCH (search_parents).
2776 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
2777 spool_read_header() do less string testing, by means of a preliminary
2778 switch on the second character of optional "-foo" lines. (This is
2779 overdue, caused by the large number of possibilities that now exist.
2780 Originally there were few.) While I was there, I also converted the
2781 str(n)cmp tests so they don't re-test the leading "-" and the first
2782 character, in the hope this might squeeze out yet more improvement.
2784 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
2785 flag allowing group syntax was set by the header_syntax check but not
2786 turned off, possible causing trouble later; (2) The flag was not being
2787 set at all for the header_verify test, causing "group"-style headers to
2788 be rejected. I have now set it in this case, and also caused header_
2789 verify to ignore an empty address taken from a group. While doing this, I
2790 came across some other cases where the code for allowing group syntax
2791 while scanning a header line wasn't quite right (mostly, not resetting
2792 the flag correctly in the right place). These bugs could have caused
2793 trouble for malformed header lines. I hope it is now all correct.
2795 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
2796 with the "reply" argument non-NULL. The code, however (which originally
2797 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
2798 but it didn't always do it. This confused somebody who was copying the
2799 code for some other use. I have removed all the tests.
2801 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
2802 feature that was used to support insecure browsers during the U.S. crypto
2803 embargo. It requires special client support, and Exim is probably the
2804 only MTA that supported it -- and would never use it because real RSA is
2805 always available. This code has been removed, because it had the bad
2806 effect of slowing Exim down by computing (never used) parameters for the
2807 RSA_EXPORT functionality.
2809 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
2810 authenticator to fail if there's a tab character in the incoming data
2811 (there should never be unless someone is messing about, as it's supposed
2812 to be base64-encoded). Also added, on Timo's advice, the "secured" option
2813 if the connection is using TLS or if the remote IP is the same as the
2814 local IP, and the "valid-client-cert option" if a client certificate has
2817 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
2818 authenticators. This can be used for authorization after authentication
2819 succeeds. (In the case of plaintext, it servers for both authentication
2822 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
2823 if any retry times were supplied.
2825 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
2826 connection, where there is no client IP address to check. In this
2827 situation, the verify now always succeeds.
2829 PH/25 Applied John Jetmore's -Mset patch.
2831 PH/26 Added -bem to be like -Mset, but loading a message from a file.
2833 PH/27 In a string expansion for a processed (not raw) header when multiple
2834 headers of the same name were present, leading whitespace was being
2835 removed from all of them, but trailing whitespace was being removed only
2836 from the last one. Now trailing whitespace is removed from each header
2837 before concatenation. Completely empty headers in a concatenation (as
2838 before) are ignored.
2840 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
2841 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
2843 PH/29 [Removed. This was a change that I later backed out, and forgot to
2844 correct the ChangeLog entry (that I had efficiently created) before
2845 committing the later change.]
2847 PH/30 Exim was sometimes attempting to deliver messages that had suffered
2848 address errors (4xx response to RCPT) over the same connection as other
2849 messages routed to the same hosts. Such deliveries are always "forced",
2850 so retry times are not inspected. This resulted in far too many retries
2851 for the affected addresses. The effect occurred only when there were more
2852 hosts than the hosts_max_try setting in the smtp transport when it had
2853 the 4xx errors. Those hosts that it had tried were not added to the list
2854 of hosts for which the message was waiting, so if all were tried, there
2855 was no problem. Two fixes have been applied:
2857 (i) If there are any address or message errors in an SMTP delivery, none
2858 of the hosts (tried or untried) are now added to the list of hosts
2859 for which the message is waiting, so the message should not be a
2860 candidate for sending over the same connection that was used for a
2861 successful delivery of some other message. This seems entirely
2862 reasonable: after all the message is NOT "waiting for some host".
2863 This is so "obvious" that I'm not sure why it wasn't done
2864 previously. Hope I haven't missed anything, but it can't do any
2865 harm, as the worst effect is to miss an optimization.
2867 (ii) If, despite (i), such a delivery is accidentally attempted, the
2868 routing retry time is respected, so at least it doesn't keep
2869 hammering the server.
2871 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
2872 in ${readsocket because some servers need this prod.
2874 PH/32 Added some extra debug output when updating a wait-xxx database.
2876 PH/33 The hint "could be header name not terminated by colon", which has been
2877 given for certain expansion errors for a long time, was not being given
2878 for the ${if def:h_colon_omitted{... case.
2880 PH/34 The spec says: "With one important exception, whenever a domain list is
2881 being scanned, $domain contains the subject domain." There was at least
2882 one case where this was not true.
2884 PH/35 The error "getsockname() failed: connection reset by peer" was being
2885 written to the panic log as well as the main log, but it isn't really
2886 panic-worthy as it just means the connection died rather early on. I have
2887 removed the panic log writing for the ECONNRESET error when getsockname()
2890 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
2891 runs only) independently of the message's sender address. This meant
2892 that, if the 4xx error was in fact related to the sender, a different
2893 message to the same recipient with a different sender could confuse
2894 things. In particular, this can happen when sending to a greylisting
2895 server, but other circumstances could also provoke similar problems.
2896 I have changed the default so that the retry time for these errors is now
2897 based a combination of the sender and recipient addresses. This change
2898 can be overridden by setting address_retry_include_sender=false in the
2901 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
2902 remote server are returned as part of bounce messages. This was not
2903 happening for LMTP over a pipe (the lmtp transport), but now it is the
2904 same for both kinds of LMTP.
2906 PH/38 Despite being documented as not happening, Exim was rewriting addresses
2907 in header lines that were in fact CNAMEs. This is no longer the case.
2909 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
2910 and queue runs started by the daemon processed all messages. This has
2911 been fixed so that -R and -S can now usefully be given with -q<time>.
2913 PH/40 Import PCRE release 6.7 (fixes some bugs).
2915 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
2917 PH/42 Give an error if -q is specified more than once.
2919 PH/43 Renamed the variables $interface_address and $interface_port as
2920 $received_ip_address and $received_port, to make it clear that these
2921 values apply to message reception, and not to the outgoing interface when
2922 a message is delivered. (The old names remain recognized, of course.)
2924 PH/44 There was no timeout on the connect() call when using a Unix domain
2925 socket in the ${readsocket expansion. There now is.
2927 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
2928 be meaningful with "accept".
2930 SC/01 Eximstats V1.43
2931 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
2933 SC/02 Eximstats V1.44
2934 Use a glob alias rather than an array ref in the generated
2935 parser. This improves both readability and performance.
2937 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
2938 Collect SpamAssassin and rejection statistics.
2939 Don't display local sender or destination tables unless
2940 there is data to show.
2941 Added average volumes into the top table text output.
2943 SC/04 Eximstats V1.46
2944 Collect data on the number of addresses (recipients)
2945 as well as the number of messages.
2947 SC/05 Eximstats V1.47
2948 Added 'Message too big' to the list of mail rejection
2949 reasons (thanks to Marco Gaiarin).
2951 SC/06 Eximstats V1.48
2952 Mainlog lines which have GMT offsets and are too short to
2953 have a flag are now skipped.
2955 SC/07 Eximstats V1.49 (Alain Williams)
2956 Added the -emptyok flag.
2958 SC/08 Eximstats V1.50
2959 Fixes for obtaining the IP address from reject messages.
2961 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
2962 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
2963 whitespace changes from 4.64-PH/27
2965 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
2968 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
2969 are found, allow negative numbers in numeric criteria)
2971 JJ/06 exipick.20061117.2, added new $message_body_missing variable
2973 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
2974 to match changes made in 4.64-PH/43
2976 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
2978 PH/47 Put in an explicit test for a DNS lookup of an address record where the
2979 "domain" is actually an IP address, and force a failure. This locks out
2980 those revolvers/nameservers that support "A-for-A" lookups, in
2981 contravention of the specifications.
2983 PH/48 When a host name was looked up from an IP address, and the subsequent
2984 forward lookup of the name timed out, the host name was left in
2985 $sender_host_name, contrary to the specification.
2987 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
2988 restricted to single-key lookups, Exim was not diagnosing an error if
2989 * or *@ was used with a query-style lookup.
2991 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
2993 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
2994 long ago, but noone interested enough thought of it. Let's just say that
2995 the "1.1" means that there are some new functions that weren't there at
2996 some point in the past.
2998 PH/51 Error processing for expansion failure of helo_data from an smtp
2999 transport during callout processing was broken.
3001 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
3002 tested/used via the -bh/-bhc/-bs options.
3004 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
3005 bug, fixed in subsequent PCRE releases).
3007 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
3008 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
3010 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
3016 SC/01 Use a glob alias rather than an array ref in eximstats generated
3017 parser. This improves both readability and performance.
3019 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
3020 Don't display local sender or destination tables in eximstats unless
3021 there is data to show.
3022 Added average volumes into the eximstats top table text output.
3024 SC/03 Collect data on the number of addresses (recipients) as well
3025 as the number of messages in eximstats.
3027 TF/01 Correct an error in the documentation for the redirect router. Exim
3028 does (usually) call initgroups() when daemonizing.
3030 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
3031 with consistent privilege compared to when running as a daemon.
3033 TF/03 Note in the spec that $authenticated_id is not set for local
3034 submissions from trusted users.
3036 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
3037 Thanks to Dean Brooks <dean@iglou.com> for the patch.
3039 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
3040 by adding some example configuration directives to the default
3041 configuration file. A little bit of work is required to uncomment the
3042 directives and define how usernames and passwords are checked, but
3043 there is now a framework to start from.
3045 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
3046 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
3047 without this. I don't know how relevant this is to other LDAP libraries.
3049 PH/02 Add the verb name to the "unknown ACL verb" error.
3051 PH/03 Magnus Holmgren's patch for filter_prepend_home.
3053 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
3055 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
3056 directory not expanded when it should be if an expanded home directory
3057 was set for the address (which is overridden by the transport).
3059 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
3062 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
3063 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
3064 because it is too late at that time, and has no effect.
3066 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
3067 security issue with \' (bugzilla #107). I could not use the
3068 PQescapeStringConn() function, because it needs a PGconn value as one of
3071 PH/08 When testing addresses using -bt, indicate those final addresses that
3072 are duplicates that would not cause an additional delivery. At least one
3073 person was confused, thinking that -bt output corresponded to deliveries.
3074 (Suppressing duplicates isn't a good idea as you lose the information
3075 about possibly different redirections that led to the duplicates.)
3077 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
3078 systems where poll() doesn't work, in particular OS X.
3080 PH/10 Added more information to debugging output for retry time not reached.
3082 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
3083 operations in malware.c.
3085 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
3088 PH/13 If write_rejectlog was set false when logging was sent to syslog with
3089 syslog_duplication set false, log lines that would normally be written
3090 both the the main log and to the reject log were not written to syslog at
3093 PH/14 In the default configuration, change the use of "message" in ACL warn
3094 statements to "add_header".
3096 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
3097 not followed by a command (e.g. "seen endif").
3099 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
3100 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
3103 PH/17 Added extra conditions to the default value of delay_warning_condition
3107 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
3108 { match{$h_precedence:}{(?i)bulk|list|junk} } \
3109 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
3112 The Auto-Submitted: and various List- headers are standardised, whereas I
3113 don't think Precedence: ever was.
3115 PH/18 Refactored debugging code in route_finduser() to show more information,
3116 in particular, the error code if getpwnam() issues one.
3118 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
3119 This is apparently needed in addition to the PH/07 change above to avoid
3120 any possible encoding problems.
3122 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
3123 but not after initializing Perl.
3125 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
3126 output them only if debugging. By default they are written stderr,
3127 apparently, which is not desirable.
3129 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
3132 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
3135 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
3137 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
3138 authenticated or an ident call has been made. Suppress the default
3139 values for $authenticated_id and $authenticated_sender (but permit -oMai
3140 and -oMas) when testing with -bh.
3142 PH/24 Re-jigged the order of the tests in the default configuration so that the
3143 tests for valid domains and recipients precede the DNS black list and CSA
3144 tests, on the grounds that those ones are more expensive.
3146 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
3147 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
3148 This bug exists in every version of Exim that I still have, right back to
3151 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
3152 However, an attempt to turn on case-sensitivity in a regex key by
3153 including (?-i) didn't work because the subject string was already
3154 lowercased, and the effects were non-intuitive. It turns out that a
3155 one-line patch can be used to allow (?-i) to work as expected.
3161 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
3162 other effects) broke the use of negated acl sub-conditions.
3164 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
3167 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
3168 "Deny" causes Exim to reject the incoming connection with a 554 error.
3169 Unfortunately, if there is a major crisis, such as a disk failure,
3170 tcp-wrappers gives "deny", whereas what one would like would be some
3171 kind of temporary error. A kludge has been added to help with this.
3172 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
3173 554 error is used if errno is still zero or contains ENOENT (which occurs
3174 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
3177 PH/03 Add -lutil to the default FreeBSD LIBS setting.
3179 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
3180 errors. Otherwise a message that provokes a temporary error (when other
3181 messages do not) can cause a whole host to time out.
3183 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
3184 addresses were routed directly to files or pipes from a redirect router.
3185 File deliveries just didn't batch; pipe deliveries might have suffered
3188 PH/06 A failure to get a lock for a hints database would erroneously always say
3189 "Failed to get write lock", even when it was really a read lock.
3191 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
3192 of 0600. This has been changed to use the value of the lockfile_mode
3193 option (which defaults to 0600).
3195 PH/08 Applied small patch from the Sieve maintainer.
3197 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
3198 folder from quota calculations, a direct delivery into this folder messed
3199 up the contents of the maildirsize file. This was because the regex was
3200 used only to exclude .Trash (or whatever) when the size of the mailbox
3201 was calculated. There was no check that a delivery was happening into an
3202 excluded directory. This bug has been fixed by ignoring all quota
3203 processing for deliveries into excluded directories.
3205 PH/10 Added the maildirfolder_create_regex option to appendfile.
3211 PH/01 The code for finding all the local interface addresses on a FreeBSD
3212 system running IPv6 was broken. This may well have applied to all BSD
3213 systems, as well as to others that have similar system calls. The broken
3214 code found IPv4 interfaces correctly, but gave incorrect values for the
3215 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
3216 that it would not match correctly against @[] and not recognize the IPv6
3219 PH/02 The ipliteral router was not recognizing addresses of the form user@
3220 [ipv6:....] because it didn't know about the "ipv6:" prefix.
3222 PH/03 Added disable_ipv6.
3224 PH/04 Changed $reply_address to use the raw form of the headers instead of the
3225 decoded form, because it is most often used to construct To: headers
3226 lines in autoreplies, and the decoded form may well be syntactically
3227 invalid. However, $reply_address has leading white space removed, and all
3228 newlines turned into spaces so that the autoreply transport does not
3231 PH/05 If group was specified without a user on a router, and no group or user
3232 was specified on a transport, the group from the router was ignored.
3234 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
3235 for visible compile-time settings that can be used to change these
3236 numbers, for those that want even more. Backwards compatibility with old
3237 spool files has been maintained. However, going back to a previous Exim
3238 release will lost any variables that are in spool files.
3240 PH/07 Two small changes when running in the test harness: increase delay when
3241 passing a TCP/IP connection to a new process, in case the original
3242 process has to generate a bounce, and remove special handling of
3243 127.0.0.2 (sic), which is no longer necessary.
3245 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
3246 be the same on different OS.
3248 PH/09 Moved a debug statement in filter processing to avoid a race problem when
3251 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
3252 whether --show-vars was specified or not
3254 JJ/02 exipick: Added support for new ACL variable spool format introduced
3257 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
3258 syntactically invalid From: or Reply-to: line, and a filter used this to
3259 generate an autoreply, and therefore failed to obtain an address for the
3260 autoreply, Exim could try to deliver to a non-existent relative file
3261 name, causing unrelated and misleading errors. What now happens is that
3262 it logs this as a hard delivery error, but does not attempt to create a
3265 PH/11 The exinext utility has a -C option for testing purposes, but although
3266 the given file was scanned by exinext itself; it wasn't being passed on
3267 when Exim was called.
3269 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
3270 an end-of-file indication when reading a command response.
3272 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
3273 compiled. In many other places in Exim, IPv6 addresses are always
3274 recognized, so I have changed this. It also means that IPv4 domain
3275 literals of the form [IPV4:n.n.n.n] are now always recognized.
3277 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
3278 used if the router is not running as root, for example, when verifying at
3279 ACL time, or when using -bh. The debugging output from this situation was
3280 non-existent - all you got was a failure to exec. I have made two
3283 (a) Failures to set uid/gid, the current directory, or a process leader
3284 in a subprocess such as that created by queryprogram now generate
3285 suitable debugging output when -d is set.
3287 (b) The queryprogram router detects when it is not running as root,
3288 outputs suitable debugging information if -d is set, and then runs
3289 the subprocess without attempting to change uid/gid.
3291 PH/15 Minor change to Makefile for building test_host (undocumented testing
3294 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
3295 additional section of a DNS packet that returns MX or SRV records.
3296 Instead, it always explicitly searches for A/AAAA records. This avoids
3297 major problems that occur when a DNS server includes only records of one
3298 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
3299 fixed another bug: if SRV records were looked up and the corresponding
3300 address records were *not* found in the additional section, the port
3301 values from the SRV records were lost.
3303 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
3304 using the correct key (the original address) when searching the retry
3305 rules in order to find which one to use for generating the retry hint.
3307 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
3308 adding the default one. Similarly, if it contains a Reply-To: header, the
3309 errors_reply_to option, if set, is not used.
3311 PH/19 When calculating a retry time, Exim used to measure the "time since
3312 failure" by looking at the "first failed" field in the retry record. Now
3313 it does not use this if it is later than than the arrival time of the
3314 message. Instead it uses the arrival time. This makes for better
3315 behaviour in cases where some deliveries succeed, thus re-setting the
3316 "first failed" field. An example is a quota failure for a huge message
3317 when small messages continue to be delivered. Without this change, the
3318 "time since failure" will always be short, possible causing more frequent
3319 delivery attempts for the huge message than are intended.
3320 [Note: This change was subsequently modified - see PH/04 for 4.62.]
3322 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
3323 $1, $2, $3) because the numerical variables can be reset during some
3324 expansion items (e.g. "match"), thereby losing the authentication data.
3326 PH/21 Make -bV show the size of off_t variables so that the test suite can
3327 decide whether to run tests for quotas > 2G.
3329 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
3330 mailbox_size, and mailbox_filecount in the appendfile transport. If a
3331 filecount value is greater than 2G or if a quota value is greater than 2G
3332 on a system where the size of off_t is not greater than 4, a panic error
3335 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
3336 never match. The debug and -bh output now contains an explicit error
3337 message indicating a malformed IPv4 address or mask.
3339 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
3340 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
3341 PH/23 above applies.
3343 PH/25 Do not write to syslog when running in the test harness. The only
3344 occasion when this arises is a failure to open the main or panic logs
3345 (for which there is an explicit test).
3347 PH/26 Added the /no_tell option to "control=freeze".
3349 PH/27 If a host name lookup failed very early in a connection, for example, if
3350 the IP address matched host_lookup and the reverse lookup yielded a name
3351 that did not have a forward lookup, an error message of the form "no IP
3352 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
3353 could be logged. Now it outputs the IP address instead of "NULL".
3355 PH/28 An enabling patch from MH: add new function child_open_exim2() which
3356 allows the sender and the authenticated sender to be set when
3357 submitting a message from within Exim. Since child_open_exim() is
3358 documented for local_scan(), the new function should be too.
3360 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
3361 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
3362 results in an empty string is now treated as unset.
3364 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
3366 PH/31 Added sender_verify_fail logging option.
3368 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
3369 needed by RFC 822 but not by RFC 2822 was commented out. I have now
3370 tidied the source and removed it altogether.
3372 PH/33 When a queue run was abandoned because the load average was too high, a
3373 log line was always written; now it is written only if the queue_run log
3374 selector is set. In addition, the log line for abandonment now contains
3375 information about the queue run such as the pid. This is always present
3376 in "start" and "stop" lines but was omitted from the "abandon" line.
3378 PH/34 Omit spaces between a header name and the colon in the error message that
3379 is given when verify = headers_syntax fails (if there are lots of them,
3380 the message gets confusing).
3382 PH/35 Change the default for dns_check_names_pattern to allow slashes within
3383 names, as there are now some PTR records that contain slashes. This check
3384 is only to protect against broken name servers that fall over on strange
3385 characters, so the fact that it applies to all lookups doesn't matter.
3387 PH/36 Now that the new test suite is complete, we can remove some of the
3388 special code in Exim that was needed for the old test suite. For example,
3389 sorting DNS records because real resolvers return them in an arbitrary
3390 order. The new test suite's fake resolver always returns records in the
3393 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
3394 bounces) except when queue_only is set, to avoid logging races between
3395 the different processes.
3397 PH/38 Panic-die if .include specifies a non-absolute path.
3399 PH/39 A tweak to the "H" retry rule from its user.
3401 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
3402 a label. They prevented compilation on older perls.
3404 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
3405 a warning to be raised on newish perls.
3407 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
3408 on queue. Changes to match documented behaviour of showing count of
3409 messages matching specified criteria.
3411 PH/40 Changed the default ident timeout from 30s to 5s.
3413 PH/41 Added support for the use of login_cap features, on those BSD systems
3414 that have them, for controlling the resources used by pipe deliveries.
3416 PH/42 The content-scanning code uses fopen() to create files in which to put
3417 message data. Previously it was not paying any attention to the mode of
3418 the files. Exim runs with umask(0) because the rest of the code creates
3419 files with open(), and sets the required mode explicitly. Thus, these
3420 files were ending up world-writeable. This was not a big issue, because,
3421 being within the spool directory, they were not world-accessible. I have
3422 created a function called modefopen, which takes an additional mode
3423 argument. It sets umask(777), creates the file, chmods it to the required
3424 mode, then resets the umask. All the relevant calls to fopen() in the
3425 content scanning code have been changed to use this function.
3427 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
3428 to 24 hours. This avoids potential overflow problems when processing G
3429 and H retry rules. I suspect nobody ever tinkers with this value.
3431 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
3433 PH/45 When the plaintext authenticator is running as a client, the server's
3434 challenges are checked to ensure they are valid base64 strings. By
3435 default, the authentication attempt is cancelled if an invalid string is
3436 received. Setting client_ignore_invalid_base64 true ignores these errors.
3437 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
3438 they are received. Thus, the responses can be made to depend on the
3439 challenges. If an invalid string is ignored, an empty string is placed in
3442 PH/46 Messages that are created by the autoreply transport now contains a
3443 References: header, in accordance with RFCs 2822 and 3834.
3445 PH/47 Added authenticated_sender_force to the smtp transport.
3447 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
3449 PH/49 Installed latest patch from the Sieve maintainer.
3451 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
3452 was also set, the appendfile transport was unnecessarily scanning a
3453 directory of message files (e.g. for maildir delivery) to find the count
3454 of files (along with the size), even though it did not need this
3455 information. It now does the scan only if it needs to find either the
3456 size of the count of files.
3458 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
3460 PH/52 Two bugs concerned with error handling when the smtp transport is
3463 (i) Exim was not creating retry information for temporary errors given
3464 for individual recipients after the DATA command when the smtp transport
3465 was used in LMTP mode. This meant that they could be retried too
3466 frequently, and not timed out correctly.
3468 (ii) Exim was setting the flag that allows error details to be returned
3469 for LMTP errors on RCPT commands, but not for LMTP errors for individual
3470 recipients that were returned after the DATA command.
3472 PH/53 This is related to PH/52, but is more general: for any failing address,
3473 when detailed error information was permitted to be returned to the
3474 sender, but the error was temporary, then after the final timeout, only
3475 "retry timeout exceeded" was returned. Now it returns the full error as
3476 well as "retry timeout exceeded".
3478 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
3479 do this, and (what is worse) MTAs that accept it.
3481 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
3482 will now be deprecated.
3484 PH/56 New os.c-cygwin from the Cygwin maintainer.
3486 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
3487 formats (previously only available in exim formats via -bpr, -bpru,
3488 and -bpra. Now also available in native and exiqgrep formats)
3490 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
3491 with very large, slow to parse queues
3493 JJ/08 exipick: added ! as generic prefix to negate any criteria format
3495 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
3497 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
3498 responses to authentication challenges, though it was showing the
3499 challenges; (ii) I've removed the CR characters from the debug output for
3502 PH/58 Allow for the insertion of a newline as well as a space when a string
3503 is turned into more than one encoded-word during RFC 2047 encoding. The
3504 Sieve code now uses this.
3506 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
3507 data_4xx, lost_connection, tls_required.
3509 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
3510 message was being sent as an SMTP response.
3512 PH/61 Add -l and -k options to exicyclog.
3514 PH/62 When verifying, if an address was redirected to one new address, so that
3515 verification continued, and the new address failed or deferred after
3516 having set something in $address_data, the value of $address_data was not
3517 passed back to the ACL. This was different to the case when no
3518 redirection occurred. The value is now passed back in both cases.
3520 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
3521 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
3522 use login_cap.h, so on its own it isn't the distinguishing feature. The
3523 new name refers directly to the setclassresources() function.
3525 PH/65 Added configuration files for NetBSD3.
3527 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
3529 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
3530 is preferred over IPv4.
3532 PH/68 The bounce_return_message and bounce_return_body options were not being
3533 honoured for bounces generated during the reception of non-SMTP messages.
3534 In particular, this applied to messages rejected by the ACL. This bug has
3535 been fixed. However, if bounce_return_message is true and bounce_return_
3536 body is false, the headers that are returned for a non-SMTP message
3537 include only those that have been read before the error was detected.
3538 (In the case of an ACL rejection, they have all been read.)
3540 PH/69 The HTML version of the specification is now built in a directory called
3541 spec_html instead of spec.html, because the latter looks like a path with
3542 a MIME-type, and this confuses some software.
3544 PH/70 Catch two compiler warnings in sieve.c.
3546 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
3547 function verify_get_ident() calls ip_connect() to connect a socket, but
3548 if the "connect()" function timed out, ip_connect() used to close the
3549 socket. However, verify_get_ident() also closes the socket later, and in
3550 between Exim writes to the log, which may get opened at this point. When
3551 the socket was closed in ip_connect(), the log could get the same file
3552 descriptor number as the socket. This naturally causes chaos. The fix is
3553 not to close the socket in ip_connect(); the socket should be closed by
3554 the function that creates it. There was only one place in the code where
3555 this was missing, in the iplookup router, which I don't think anybody now
3556 uses, but I've fixed it anyway.
3558 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
3559 well as to direct DNS lookups. Otherwise the handling of names in host
3560 lists is inconsistent and therefore confusing.
3566 PH/01 Two changes to the default runtime configuration:
3568 (1) Move the checks for relay_from_hosts and authenticated clients from
3569 after to before the (commented out) DNS black list checks.
3571 (2) Add control=submission to the relay_from_hosts and authenticated
3572 clients checks, on the grounds that messages accepted by these
3573 statements are most likely to be submissions.
3575 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
3577 (1) Generate an error if the third argument for the ${prvs expansion is
3580 (2) Treat a missing third argument of ${prvscheck as if it were an empty
3583 (3) Reset the variables that are obtained from the first argument of
3584 ${prvscheck and used in the second argument before leaving the code,
3585 because their memory is reclaimed, so using them afterwards may do
3588 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
3589 one (it's much easier than Tom thought :-).
3591 (5) Because of (4), we can now allow for the use of $prvscheck_result
3592 inside the third argument.
3594 PH/03 For some reason, the default setting of PATH when running a command from
3595 a pipe transport was just "/usr/bin". I have changed it to
3598 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
3599 anything to be listed in the output from -bV.
3601 PH/05 When a filter generated an autoreply, the entire To: header line was
3602 quoted in the delivery log line, like this:
3604 => >A.N.Other <ano@some.domain> <original@ddress> ...
3606 This has been changed so that it extracts the operative address. There
3607 may be more than one such address. If so, they are comma-separated, like
3610 => >ano@some.domain,ona@other.domain <original@ddress> ...
3612 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
3613 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
3614 not being looked up in the rDNS to get a host name, Exim was showing the
3615 IP address twice in Received: lines, even though the IP addresses were
3616 identical. For example:
3618 Received: from [1.2.3.4] (helo=[1.2.3.4])
3620 However, if the real host name was known, it was omitting the HELO data
3621 if it matched the actual IP address. This has been tidied up so that it
3622 doesn't show the same IP address twice.
3624 PH/07 When both +timestamp and +memory debugging was on, the value given by
3625 $tod_xxx expansions could be wrong, because the tod_stamp() function was
3626 called by the debug printing, thereby overwriting the timestamp buffer.
3627 Debugging no longer uses the tod_stamp() function when +timestamp is set.
3629 PH/08 When the original message was included in an autoreply transport, it
3630 always said "this is a copy of the message, including all the headers",
3631 even if body_only or headers_only was set. It now gives an appropriate
3634 PH/09 Applied a patch from the Sieve maintainer which:
3636 o fixes some comments
3637 o adds the (disabled) notify extension core
3638 o adds some debug output for the result of if/elsif tests
3639 o points to the current vacation draft in the documentation
3640 and documents the missing references header update
3644 o fixes a bug in processing the envelope test (when testing
3645 multiple envelope elements, the last element determined the
3648 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
3649 Electronic Mail") by including:
3651 Auto-submitted: auto-generated
3653 in the messages that it generates (bounce messages and others, such as
3654 warnings). In the case of bounce messages for non-SMTP messages, there was
3655 also a typo: it was using "Auto_submitted" (underscore instead of
3656 hyphen). Since every message generated by Exim is necessarily in response
3657 to another message, thes have all been changed to:
3659 Auto-Submitted: auto-replied
3661 in accordance with these statements in the RFC:
3663 The auto-replied keyword:
3665 - SHOULD be used on messages sent in direct response to another
3666 message by an automatic process,
3668 - MUST NOT be used on manually-generated messages,
3670 - MAY be used on Delivery Status Notifications (DSNs) and Message
3671 Disposition Notifications (MDNs),
3673 - MUST NOT be used on messages generated by automatic or periodic
3674 processes, except for messages which are automatic responses to
3677 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
3678 to the default Received: header definition.
3680 PH/12 Added log selector acl_warn_skipped (default on).
3682 PH/13 After a successful wildlsearch lookup, discard the values of numeric
3683 variables because (a) they are in the wrong storage pool and (b) even if
3684 they were copied, it wouldn't work properly because of the caching.
3686 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
3687 checking when decoding. Apparently there are clients that generate
3688 overlong encoded strings. Why am I not surprised?
3690 PH/15 If the first argument of "${if match_address" was not empty, but did not
3691 contain an "@" character, Exim crashed. Now it writes a panic log message
3692 and treats the condition as false.
3694 PH/16 In autoreply, treat an empty string for "once" the same as unset.
3696 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
3697 extension "envelope-auth". The code is finished and in agreement with
3698 other implementations, but there is no documentation so far and in fact,
3699 nobody wrote the draft yet. This extension is currently #undef'ed, thus
3700 not changing the active code.
3702 Print executed "if" and "elsif" statements when debugging is used. This
3703 helps a great deal to understand what a filter does.
3705 Document more things not specified clearly in RFC3028. I had all this
3706 sorted out, when out of a sudden new issues came to my mind. Oops."
3708 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
3711 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
3712 canonical form (as documented). However, after a host name lookup from
3713 the IP address, check_host() was doing a simple string comparison with
3714 addresses acquired from the DNS when checking that the found name did
3715 have the original IP as one of its addresses. Since any found IPv6
3716 addresses are likely to be in abbreviated form, the comparison could
3717 fail. Luckily, there already exists a function for doing the comparison
3718 by converting both addresses to binary, so now that is used instead of
3719 the text comparison.
3721 PH/20 There was another similar case to PH/19, when a complete host name was
3722 given in a host list; looking up its IP address could give an abbreviated
3723 form, whereas the current host's name might or might not be abbreviated.
3724 The same fix has been applied.
3730 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
3731 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
3734 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
3735 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
3737 PH/03 Typo: missing ".o" in src/pcre/Makefile.
3739 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
3740 header line, restrict the check to what is listed in RFCs 2369 and 2929.
3741 Also, for "Auto-Submitted", treat anything other than "no" as
3742 non-personal, in accordance with RFC 3834. (Previously it treated
3743 anything starting "auto-" as non-personal.)
3745 TF/01 The control=submission/name=... option had a problem with syntax
3746 errors if the name included a slash character. The /name= option
3747 now slurps the rest of the string, so it can include any characters
3748 but it must come last in the list of options (after /sender_retain
3751 PH/05 Some modifications to the interface to the fake nameserver for the new
3759 TK/01 Added the "success_on_redirect" address verification option. See
3760 NewStuff for rationale and an example.
3762 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
3764 PH/02 Patch to exigrep to allow it to work on syslog lines.
3766 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
3767 fread() to skip over the body file's header line, because in Cygwin the
3768 header line is locked and is inaccessible.
3770 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
3771 co-exist for some time) to make it clear that it is the Exim ID that is
3772 referenced, not the Message-ID: header line.
3774 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
3775 string_format() function, because snprintf() does not exist on all
3778 PH/06 The use of forbid_filter_existstest now also locks out the use of the
3779 ${stat: expansion item.
3781 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
3782 protocol synchronization error", to keep the pedants happy.
3784 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
3785 well as for IRIX systems, when gcc is being used. See the host.c source
3788 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
3790 PH/10 Named domain lists were not working if used in a queue_smtp_domains
3793 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
3794 transport and to the smtp transport in LMTP mode.
3796 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
3798 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
3799 run a filter in a subprocess. This could lead to confusion in subsequent
3800 lookups in the parent process. There should also be a search_tidyup() at
3801 the end of the subprocess.
3803 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
3804 only if the host matched helo_try_verify_hosts, which caused the
3805 verification to occur when the EHLO/HELO command was issued. The ACL just
3806 tested the remembered result. Now, if a previous verification attempt has
3807 not happened, "verify = helo" does it there and then.
3809 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
3811 TK/03 Fix log output including CR from clamd.
3813 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
3814 exist provoked a memory error which could cause a segfault.
3816 PH/15 Installed PCRE 6.2
3818 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
3820 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
3821 of the problem. Specifically, suggested +O2 rather than +O1 for the
3824 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
3826 PH/20 If a delivery was routed to a non-standard port by means of an SRV
3827 record, the port was not correctly logged when the outgoing_port log
3828 selector was set (it logged the transort's default port).
3830 PH/21 Added support for host-specific ports to manualroute, queryprogram,
3831 fallback_hosts, and "hosts" in the smtp transport.
3833 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
3834 host errors such as "Connection refused".
3836 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
3837 authentication with radiusclient 0.4.9:
3839 - Error returned from rc_read_config was caught wrongly
3840 - Username/password not passed on to radius server due to wrong length.
3842 The presumption is that some radiusclient API changes for 4.51/PH/17
3843 were not taken care of correctly. The code is still untested by me (my
3844 Linux distribution still has 0.3.2 of radiusclient), but it was
3845 contributed by a Radius user.
3847 PH/24 When doing a callout, the value of $domain wasn't set correctly when
3848 expanding the "port" option of the smtp transport.
3850 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
3851 while reading a MIME header. Thanks to Tom Hughes for a patch.
3853 PH/24 Include config.h inside local_scan.h so that configuration settings are
3856 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
3857 that in an ACL for RCPT (for example), you can examine exactly what was
3860 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
3861 commands, but it was not correctly comparing the address with the actual
3862 client host address. Thus, it would show the EHLO address in Received:
3863 header lines when this was not necessary.
3865 PH/27 Added the % operator to ${eval:}.
3867 PH/28 Exim tries to create and chdir to its spool directory when it starts;
3868 it should be ignoring failures (because with -C, for example, it has lost
3869 privilege). It wasn't ignoring creation failures other than "already
3872 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
3875 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
3876 because an input line was too long, either on its own, or by virtue of
3877 too many continuations, the temporary file was not being removed, and the
3878 return code was incorrect.
3880 PH/31 Missing "BOOL" in function definition in filtertest.c.
3882 PH/32 Applied Sieve patches from the maintainer.
3884 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
3886 PH/33 Added "verify = not_blind".
3888 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
3889 Local/Makefile (with some defaults set). These are used in built scripts
3890 such as exicyclog, but they have never been used in the exim_install
3891 script (though there are many overriding facilities there). I have
3892 arranged that the exim_install script now takes note of these two
3895 PH/35 Installed configuration files for Dragonfly.
3897 PH/36 When a locally submitted message by a trusted user did not contain a
3898 From: header, and the sender address was obtained from -f or from an SMTP
3899 MAIL command, and the trusted user did not use -F to supply a sender
3900 name, $originator_name was incorrectly used when constructing a From:
3901 header. Furthermore, $originator_name was used for submission mode
3902 messages from external hosts without From: headers in a similar way,
3903 which is clearly wrong.
3905 PH/37 Added control=suppress_local_fixups.
3907 PH/38 When log_selector = +received_sender was set, and the addition of the
3908 sender made the log line's construction buffer exactly full, or one byte
3909 less than full, an overflow happened when the terminating "\n" was
3912 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
3913 when the result of a list match is failure because a DNS lookup failed.
3915 PH/40 RM_COMMAND is now used in the building process.
3917 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
3918 the "build-* directories that it finds.
3920 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
3921 address in a domain literal was a prefix of an interface address.
3923 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
3924 when verifying a sender address, unless rewrite_headers is false.
3926 PH/44 Wrote a long comment about why errors_to addresses are verified as
3927 recipients, not senders.
3929 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
3930 the ratelimit ACL was added.
3932 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
3934 PH/46 Added extra information about PostgreSQL errors to the error string.
3936 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
3937 avoiding the need to install special zones in a real server. This is
3938 backwards compatible; if it can't find the fake resolver, it drops back.
3939 Thus, both old and new test suites can be run.
3941 TF/02 Added util/ratelimit.pl
3943 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
3944 clock is set back in time.
3946 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
3947 Candler <B.Candler@pobox.com>.
3949 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
3950 OK for addresses that are the result of redirections.
3952 PH/48 A number of further additions for the benefit of the new test suite,
3953 including a fake gethostbyname() that interfaces to the fake DNS resolver
3956 TF/06 The fix for widen_domains has also been applied to qualify_single and
3957 search_parents which are the other dnslookup options that can cause
3960 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
3963 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
3965 TF/07 Exim produced the error message "an SRV record indicated no SMTP
3966 service" if it encountered an MX record with an empty target hostname.
3967 The message is now "an MX or SRV record indicated no SMTP service".
3969 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
3970 if the DNS of the sending site is misconfigured. This is quite a
3971 common situation. This change restores the behaviour of treating a
3972 helo verification defer as a failure.
3974 PH/51 If self=fail was set on a router, the bounce message did not include the
3975 actual error message.
3981 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
3983 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
3984 command itself ended in error, the underlying message about the transport
3985 filter timeout was being overwritten with the pipe command error. Now the
3986 underlying error message should be appended to the second error message.
3988 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
3990 PH/02 Reduce the amount of output that "make" produces by default. Full output
3991 can still be requested.
3993 PH/03 The warning log line about a condition test deferring for a "warn" verb
3994 was being output only once per connection, rather than after each
3995 occurrence (because it was using the same function as for successful
3996 "warn" verbs). This seems wrong, so I have changed it.
3998 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
3999 it should not have, which might have caused a crash in the right
4000 circumstances, but probably never did.
4002 PH/04 Installed a modified version of Tony Finch's patch to make submission
4003 mode fix the return path as well as the Sender: header line, and to
4004 add a /name= option so that you can make the user's friendly name appear
4007 TF/03 Added the control = fakedefer ACL modifier.
4009 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
4010 Mark Lowes for thorough testing.
4012 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
4014 TK/03 Merged latest SRS patch from Miles Wilton.
4016 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
4017 with the definition in sysexits.h (which is #included earlier).
4018 Fortunately, Exim does not actually use EX_OK. The code used to try to
4019 preserve the sysexits.h value, by assuming that macro definitions were
4020 scanned for macro replacements. I have been disabused of this notion,
4021 so now the code just undefines EX_OK before #including unistd.h.
4023 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
4024 in the smtp transport. When a block could not be written in a single
4025 write() function, the timeout was being re-applied to each part-write.
4026 This seems wrong - if the receiver was accepting one byte at a time it
4027 would take for ever. The timeout is now adjusted when this happens. It
4028 doesn't have to be particularly precise.
4030 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
4031 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
4033 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
4034 without a domain if the check to <postmaster@domain> fails.
4036 SC/01 Eximstats: added -xls and the ability to specify output files
4037 (patch written by Frank Heydlauf).
4039 SC/02 Eximstats: use FileHandles for outputting results.
4041 SC/03 Eximstats: allow any combination of xls, txt, and html output.
4043 SC/04 Eximstats: fixed display of large numbers with -nvr option
4045 SC/05 Eximstats: fixed merging of reports with empty tables.
4047 SC/06 Eximstats: added the -include_original_destination flag
4049 SC/07 Eximstats: removed tabs and trailing whitespace.
4051 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
4053 TK/06 MBOX spool code: Add real "From " MBOX separator line
4054 so the .eml file is really in mbox format (even though
4055 most programs do not really care). Patch from Alex Miller.
4057 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
4058 The latter is generated from $received_to and is only set if the
4059 message has one envelope recipient. SA can use these headers,
4060 obviously out-of-the-box. Patch from Alex Miller.
4062 PH/08 The ${def test on a variable was returning false if the variable's
4063 value was "0", contrary to what the specification has always said!
4064 The result should be true unless the variable is empty.
4066 PH/09 The syntax error of a character other than { following "${if
4067 def:variable_name" (after optional whitespace) was not being diagnosed.
4068 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
4069 accidental colon was present, for example, could give incorrect results.
4071 PH/10 Tidied the code in a number of places where the st_size field of a stat()
4072 result is used (not including appendfile, where other changes are about
4075 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
4076 This involved changing a lot of size variables from int to off_t. It
4077 should work with maildirs and everything.
4079 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
4080 spamd dying while we are connected to it.
4082 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
4085 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
4086 function for BDB 4.3.
4088 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
4090 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
4091 the sources for PCRE have been re-arranged and more files are now
4094 PH/15 The code I had for printing potentially long long variables in PH/11
4095 above was not the best (it lost precision). The length of off_t variables
4096 is now inspected at build time, and an appropriate printing format (%ld
4097 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
4098 to be "long long int" or "long int". This is needed for the internal
4099 formatting function string_vformat().
4101 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
4102 the configuration file to be ":syslog", then the script "guesses" where
4103 the logs files are, rather than using the compiled in default. In our
4104 case the guess is not the same as the compiled default, so the script
4105 suddenly stopped working when I started to use syslog. The patch checks
4106 to see if log_file_path is "". If so, it attempts to read it from exim
4107 with no configuration file to get the compiled in version, before it
4108 falls back to the previous guessing code."
4110 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
4111 implementing BATV in an Exim configuration. See NewStuff for the gory
4114 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
4115 Makefile that are specific to HP-UX.
4117 PH/18 If the "use_postmaster" option was set for a recipient callout together
4118 with the "random" option, the postmaster address was used as the MAIL
4119 FROM address for the random test, but not for the subsequent recipient
4120 test. It is now used for both.
4122 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
4123 patch removes a few documentation additions to RFC 3028, because the
4124 latest draft now contains them. It adds the new en;ascii-case comparator
4125 and a new error check for 8bit text in MIME parts. Comparator and
4126 require names are now matched exactly. I enabled the subaddress
4127 extension, but it is not well tested yet (read: it works for me)."
4129 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
4130 rework some of the code of TK/09 above to avoid the hardwired use of
4131 "%lld" and "long long". Replaced the call to snprintf() with a call to
4134 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
4135 records point to non-existent hosts", "retry timeout exceeded", and
4136 "retry time not reached for any host after a long failure period".
4138 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
4139 experimental DomainKeys support:
4141 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
4142 (2) On an error such as an illegally used "control", the wrong name for
4143 the control was given.
4145 These problems did NOT occur unless DomainKeys support was compiled.
4147 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
4149 PH/24 Added ${if match_ip condition.
4151 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
4152 databases so that it will be absolutely obvious if a crash occurs in the
4153 DB library. This is a regular occurrence (often caused by mis-matched
4156 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
4157 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
4158 picked up on a user's system that detects such things. There doesn't seem
4159 to be a gcc warning option for this - only an attribute that has to be
4160 put on the function's prototype. It seems that in Fedora Core 4 they have
4161 set this on a number of new functions. No doubt there will be more in due
4164 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
4165 specify a transport. However, if an address that was verified by such a
4166 router was the subject of a callout, Exim crashed because it tried to
4167 read the rcpt_include_affixes from the non-existent transport. Now it
4168 just assumes that the setting of that option is false. This bug was
4169 introduced by 4.51/PH/31.
4171 PH/28 Changed -d+all to exclude +memory, because that information is very
4172 rarely of interest, but it makes the output a lot bigger. People tend to
4173 do -d+all out of habit.
4175 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
4176 code in os-type was giving problems when libc.so lives in lib64, like on
4179 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
4180 aren't the modern standard, and it seems that some systems' include files
4181 don't always have them. Exim was already checking for some of the newer
4182 ones like T_AAAA, and defining it itself. I've added checks for all the
4183 record types that Exim uses.
4185 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
4186 not automatically generating a new one, as it is supposed to. This
4187 prevented TLS from working. If the file did exist, but contained invalid
4188 data, a new version was generated, as expected. It was only the case of a
4189 non-existent file that was broken.
4191 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
4192 with a change in libdomainkeys > 0.64.
4194 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
4195 from DNS. If the selector record carries the flag, it now has
4196 precedence over the domain-wide flag.
4198 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
4200 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
4201 the use of a port name that isn't defined in /etc/services) occurred, the
4202 message was deferred as in a normal delivery, and thus remained on the
4203 spool, instead of being failed because of the mua_wrapper setting. This
4204 is now fixed, and I tidied up some of the mua_wrapper messages at the
4207 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
4208 the messages in a hash of arrays rather than using individual hashes.
4209 This is a bit cleaner and results in dramatic memory savings, albeit
4210 at a slight CPU cost.
4212 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
4213 as requested by Marc Sherman.
4215 SC/10 Eximstats: added histograms for user specified patterns as requested
4218 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
4220 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
4221 fopen() in the content-scanning modules that did not already have it.
4227 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
4228 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
4230 TK/02 Fix ACL "control" statement not being available in MIME ACL.
4232 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
4234 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
4235 to test Sieve filters that use "vacation".
4237 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
4238 that changes the way the GnuTLS parameters are stored in the cache file.
4239 The new format can be generated externally. For backward compatibility,
4240 if the data in the cache doesn't make sense, Exim assumes it has read an
4241 old-format file, and it generates new data and writes a new file. This
4242 means that you can't go back to an older release without removing the
4245 PH/03 A redirect router that has both "unseen" and "one_time" set does not
4246 work if there are any delivery delays because "one_time" forces the
4247 parent to be marked "delivered", so its unseen clone is never tried
4248 again. For this reason, Exim now forbids the simultaneous setting of
4251 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
4252 redirected to themselves ("homonym" addresses). Read the long ChangeLog
4253 entry if you want to know the details. The fix, however, neglected to
4254 consider the case when local delivery batching is involved. The test for
4255 "previously delivered" was not happening when checking to see if an
4256 address could be batched with a previous (undelivered) one; under
4257 certain circumstances this could lead to multiple deliveries to the same
4260 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
4261 in its include files, and this causes problems building Exim.
4263 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
4264 header_syntax) but Exim was just ignoring anything given after a slash.
4265 In particular, this caused confusion with an attempt to use "verify =
4266 reverse_host_lookup/defer_ok". An error is now given when options are
4267 supplied for verify items that do not have them. (Maybe reverse_host_
4268 lookup should have a defer_ok option, but that's a different point.)
4270 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
4271 defined by RFC 821) to 2048, because there were problems with some AUTH
4272 commands, and RFC 1869 says the size should be increased for extended
4273 SMTP commands that take arguments.
4275 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
4278 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
4279 "unknown" error; now it says that the functionality isn't in the binary.
4281 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
4282 an address' error message when a string expansion fails (syntax or
4283 whatever). Otherwise the password may appear in the log. Following change
4284 PH/42 below, there is no longer a chance of it appearing in a bounce
4287 PH/11 Installed exipick version 20050225.0 from John Jetmore.
4289 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
4290 of its addresses was ever tried. (Bugzilla bug #2.)
4292 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
4293 the result incorrectly in the debug output. (It correctly added a newline
4294 to what was transported.)
4296 TF/01 Added $received_time.
4298 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
4299 commented out examples of how to interface to a virus scanner and to
4300 SpamAssassin. Also added commented examples of av_scanner and
4301 spamd_address settings.
4303 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
4304 and controls are allowed in which ACLs. There were a couple of minor
4305 errors. Some of the entries in the conditions table (which is a table of
4306 where they are NOT allowed) were getting very unwieldy; rewrote them as a
4307 negation of where the condition IS allowed.
4309 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
4311 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
4312 header file does not have a version number, so I've had to invent a new
4313 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
4314 API. The code is untested by me (my Linux distribution still has 0.3.2 of
4315 radiusclient), but it was contributed by a Radius user.
4317 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
4318 files or directories, for OpenSSL.
4320 PH/19 When an Exim process that is running as root has to create an Exim log
4321 file, it does so in a subprocess that runs as exim:exim so as to get the
4322 ownership right at creation (otherwise, other Exim processes might see
4323 the file with the wrong ownership). There was no test for failure of this
4324 fork() call, which would lead to the process getting stuck as it waited
4325 for a non-existent subprocess. Forks do occasionally fail when resources
4326 run out. I reviewed all the other calls to fork(); they all seem to check
4329 PH/20 When checking for unexpected SMTP input at connect time (before writing
4330 the banner), Exim was not dealing correctly with a non-positive return
4331 from the read() function. If the client had disconnected by this time,
4332 the result was a log entry for a synchronization error with an empty
4333 string after "input=" when read() returned zero. If read() returned -1
4334 (an event I could not check), uninitialized data bytes were printed.
4335 There were reports of junk text (parts of files, etc) appearing after
4338 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
4340 PH/22 Added support for macro redefinition, and (re)definition in between
4341 driver and ACL definitions.
4343 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
4344 forgetting to use the resulting value; it was using the unexpanded value.
4346 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
4347 hadn't been configured. The fix is from Juergen Kreileder, who
4348 understands it better than I do:
4350 "Here's what I see happening with three configured cyrus_sasl
4351 authenticators configured (plain, login, cram-md5):
4353 On startup auth_cyrus_sasl_init() gets called for each of these.
4354 This means three calls to sasl_listmech() without a specified mech_list.
4355 => SASL tests which mechs of all available mechs actually work
4356 => three warnings about OTP not working
4357 => the returned list contains: plain, login, cram-md5, digest-md5, ...
4359 With the patch, sasl_listmech() also gets called three times. But now
4360 SASL's mech_list option is set to the server_mech specified in the the
4361 authenticator. Or in other words, the answer from sasl_listmech()
4362 gets limited to just the mech you're testing for (which is different
4364 => the return list contains just 'plain' or 'login', 'cram-md5' or
4365 nothing depending on the value of ob->server_mech.
4367 I've just tested the patch: Authentication still works fine,
4368 unavailable mechs specified in the exim configuration are still
4369 caught, and the auth.log warnings about OTP are gone."
4371 PH/25 When debugging is enabled, the contents of the command line are added
4372 to the debugging output, even when log_selector=+arguments is not
4375 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
4376 answer is "GNU", and only if the return is "GNU/something" is the answer
4379 PH/27 $acl_verify_message is now set immediately after the failure of a
4380 verification in an ACL, and so is available in subsequent modifiers. In
4381 particular, the message can be preserved by coding like this:
4383 warn !verify = sender
4384 set acl_m0 = $acl_verify_message
4386 Previously, $acl_verify_message was set only while expanding "message"
4387 and "log_message" when a very denied access.
4389 PH/28 Modified OS/os.c-Linux with
4391 -#ifndef OS_LOAD_AVERAGE
4392 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
4394 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
4395 nomenclature these days.)
4397 PH/29 Installed patch from the Sieve maintainer that adds the options
4398 sieve_useraddress and sieve_subaddress to the redirect router.
4400 PH/30 In these circumstances:
4401 . Two addresses routed to the same list of hosts;
4402 . First host does not offer TLS;
4403 . First host accepts first address;
4404 . First host gives temporary error to second address;
4405 . Second host offers TLS and a TLS session is established;
4406 . Second host accepts second address.
4407 Exim incorrectly logged both deliveries with the TLS parameters (cipher
4408 and peerdn, if requested) that were in fact used only for the second
4411 PH/31 When doing a callout as part of verifying an address, Exim was not paying
4412 attention to any local part prefix or suffix that was matched by the
4413 router that accepted the address. It now behaves in the same way as it
4414 does for delivery: the affixes are removed from the local part unless
4415 rcpt_include_affixes is set on the transport.
4417 PH/32 Add the sender address, as F=<...>, to the log line when logging a
4418 timeout during the DATA phase of an incoming message.
4420 PH/33 Sieve envelope tests were broken for match types other than :is. I have
4421 applied a patch sanctioned by the Sieve maintainer.
4423 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
4424 the uid or gid is negative. A case of a negative gid caused this to be
4425 noticed. The fix allows for either to be negative.
4427 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
4428 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
4431 PH/36 The change PH/12 above was broken. Fixed it.
4433 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
4434 the grounds that routing the same address twice would always produce the
4435 same answer. This might have been true once, but it is certainly no
4436 longer true now. Routing a child address may depend on the previous
4437 routing that produced that child. Some complicated redirection strategies
4438 went wrong when messages had multiple recipients, and made Exim's
4439 behaviour dependent on the order in which the addresses were given.
4441 I have moved the duplicate checking until after the routing is complete.
4442 Exim scans the addresses that are assigned to local and remote
4443 transports, and removes any duplicates. This means that more work will be
4444 done, as duplicates will always all be routed, but duplicates are
4445 presumably rare, so I don't expect this is of any significance.
4447 For deliveries to pipes, files, and autoreplies, the duplicate checking
4448 still happens during the routing process, since they are not going to be
4451 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
4452 It corrects a timeout issue with spamd. This is Ian's comment: "The
4453 background is that sometimes spamd either never reads data from a
4454 connection it has accepted, or it never writes response data. The exiscan
4455 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
4456 blindly assumes that writes won't block so it may never time out."
4458 PH/39 Allow G after quota size as well as K and M.
4460 PH/40 The value set for $authenticated_id in an authenticator may not contain
4461 binary zeroes or newlines because the value is written to log lines and
4462 to spool files. There was no check on this. Now the value is run through
4463 the string_printing() function so that such characters are converted to
4464 printable escape sequences.
4466 PH/41 $message_linecount is a new variable that contains the total number of
4467 lines in the message. Compare $body_linecount, which is the count for the
4470 PH/42 Exim no longer gives details of delivery errors for specific addresses in
4471 bounce and delay warning messages, except in certain special cases, which
4474 (a) An SMTP error message from a remote host;
4475 (b) A message specified in a :fail: redirection;
4476 (c) A message specified in a "fail" command in a system filter;
4477 (d) A message specified in a FAIL return from the queryprogram router;
4478 (e) A message specified by the cannot_route_message router option.
4480 In these cases only, Exim does include the error details in bounce and
4481 warning messages. There are also a few cases where bland messages such
4482 as "unrouteable address" or "local delivery error" are given.
4484 PH/43 $value is now also set for the "else" part of a ${run expansion.
4486 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
4487 being worked on, but at least Exim now implements the latest version to
4490 PH/45 In a pipe transport, although a timeout while waiting for the pipe
4491 process to complete was treated as a delivery failure, a timeout while
4492 writing the message to the pipe was logged, but erroneously treated as a
4493 successful delivery. Such timeouts include transport filter timeouts. For
4494 consistency with the overall process timeout, these timeouts are now
4495 treated as errors, giving rise to delivery failures by default. However,
4496 there is now a new Boolean option for the pipe transport called
4497 timeout_defer, which, if set TRUE, converts the failures into defers for
4498 both kinds of timeout. A transport filter timeout is now identified in
4501 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
4502 systems where "make" and "gmake" are different, calling "gmake" at top
4503 level broke things. I've arranged for the value of $(MAKE) to be passed
4504 from the Makefile to this script so that it can call the same version of
4508 A note about Exim versions 4.44 and 4.50
4509 ----------------------------------------
4511 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
4512 changes of various kinds. As a consequence, a big documentation update was
4513 needed. This delayed the release for rather longer than seemed good, especially
4514 in the light of a couple of (minor) security issues. Therefore, the changes
4515 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
4516 release. So 4.44 and 4.50 are in effect two different branches that both start
4519 I have left the 4.50 change log unchanged; it contains all the changes since
4520 4.43. The change log for 4.44 is below; many of its items are identical to
4521 those for 4.50. This seems to be the most sensible way to preserve the
4522 historical information.
4528 1. Minor wording change to the doc/README.SIEVE file.
4530 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
4531 computation of the current number of files was incorrect.
4533 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4534 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4537 4. Give more explanation in the error message when the command for a transport
4538 filter fails to execute.
4540 5. There are several places where Exim runs a non-Exim command in a
4541 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4542 was being done only for the command run by the queryprogram router. It is
4543 now done for all such subprocesses. The other cases are: ${run, transport
4544 filters, and the commands run by the lmtp and pipe transports.
4546 6. Added CONFIGURE_GROUP build-time option.
4548 7. Some older OS have a limit of 256 on the maximum number of file
4549 descriptors. Exim was using setrlimit() to set 1000 as a large value
4550 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4551 systems. I've change it so that if it can't get 1000, it tries for 256.
4553 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4554 was an oversight, and furthermore, ever since the addition of extra
4555 controls (e.g. 4.43/32), the checks on when to allow different forms of
4556 "control" were broken. There should now be diagnostics for all cases when a
4557 control that does not make sense is encountered.
4559 9. Added the /retain_sender option to "control=submission".
4561 10. $recipients is now available in the predata ACL (oversight).
4563 11. Tidy the search cache before the fork to do a delivery from a message
4564 received from the command line. Otherwise the child will trigger a lookup
4565 failure and thereby defer the delivery if it tries to use (for example) a
4566 cached ldap connection that the parent has called unbind on.
4568 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4569 of $address_data from the recipient verification was clobbered by the
4570 sender verification.
4572 13. The value of address_data from a sender verification is now available in
4573 $sender_address_data in subsequent conditions in the ACL statement.
4575 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
4577 15. Added a new option "connect=<time>" to callout options, to set a different
4580 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4581 was its contents. (It was OK if the option was not defined at all.)
4583 17. A "Completed" log line is now written for messages that are removed from
4584 the spool by the -Mrm option.
4586 18. New variables $sender_verify_failure and $recipient_verify_failure contain
4587 information about exactly what failed.
4589 19. Added -dd to debug only the daemon process.
4591 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
4592 handles timeouts, both on the server side and network timeouts. Renamed the
4593 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
4595 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
4596 It is now set to "smtps".
4598 22. $host_address is now set to the target address during the checking of
4599 ignore_target_hosts.
4601 23. When checking ignore_target_hosts for an ipliteral router, no host name was
4602 being passed; this would have caused $sender_host_name to have been used if
4603 matching the list had actually called for a host name (not very likely,
4604 since this list is usually IP addresses). A host name is now passed as
4607 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4608 code that specifies a non-restarting handler (typically sigaction() in
4609 modern systems) in an attempt to fix a rare and obscure crash bug.
4611 25. Narrowed the window for a race in the daemon that could cause it to ignore
4612 SIGCHLD signals. This is not a major problem, because they are used only to
4613 wake it up if nothing else does.
4615 26. A malformed maildirsize file could cause Exim to calculate negative values
4616 for the mailbox size or file count. Odd effects could occur as a result.
4617 The maildirsize information is now recalculated if the size or filecount
4620 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4621 support for a long time. Removed HAVE_SYS_VFS_H.
4623 28. Installed the latest version of exipick from John Jetmore.
4625 29. In an address list, if the pattern was not a regular expression, an empty
4626 subject address (from a bounce message) matched only if the pattern was an
4627 empty string. Non-empty patterns were not even tested. This was the wrong
4628 because it is perfectly reasonable to use an empty address as part of a
4629 database query. An empty address is now tested by patterns that are
4630 lookups. However, all the other forms of pattern expect the subject to
4631 contain a local part and a domain, and therefore, for them, an empty
4632 address still always fails if the pattern is not itself empty.
4634 30. Exim went into a mad DNS loop when attempting to do a callout where the
4635 host was specified on an smtp transport, and looking it up yielded more
4636 than one IP address.
4638 31. Re-factored the code for checking spool and log partition space into a
4639 function that finds that data and another that does the check. The former
4640 is then used to implement four new variables: $spool_space, $log_space,
4641 $spool_inodes, and $log_inodes.
4643 32. The RFC2047 encoding function was originally intended for short strings
4644 such as real names; it was not keeping to the 75-character limit for
4645 encoded words that the RFC imposes. It now respects the limit, and
4646 generates multiple encoded words if necessary. To be on the safe side, I
4647 have increased the buffer size for the ${rfc2047: expansion operator from
4650 33. It is now permitted to omit both strings after an "if" condition; if the
4651 condition is true, the result is "true". As before, when the second string
4652 is omitted, a false condition yields an empty string. This makes it less
4653 cumbersome to write custom ACL and router conditions.
4655 34. Failure to deliver a bounce message always caused it to be frozen, even if
4656 there was an errors_to setting on the router. The errors_to setting is now
4659 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4660 canonical form (fully expanded) before being placed in
4661 $sender_host_address.
4663 36. The table in the code that translates DNS record types into text (T_A to
4664 "A" for instance) was missing entries for NS and CNAME. It is just possible
4665 that this could have caused confusion if both these types were looked up
4666 for the same domain, because the text type is used as part of Exim's
4667 per-process caching. But the chance of anyone hitting this buglet seems
4670 37. The dnsdb lookup has been extended in a number of ways.
4672 (1) There is a new type, "zns", which walks up the domain tree until it
4673 finds some nameserver records. It should be used with care.
4675 (2) There is a new type, "mxh", which is like "mx" except that it returns
4676 just the host names, not the priorities.
4678 (3) It is now possible to give a list of domains (or IP addresses) to be
4679 looked up. The behaviour when one of the lookups defers can be
4680 controlled by a keyword.
4682 (4) It is now possible to specify the separator character for use when
4683 multiple records are returned.
4685 38. The dnslists ACL condition has been extended: it is now possible to supply
4686 a list of IP addresses and/or domains to be looked up in a particular DNS
4689 39. Added log_selector=+queue_time_overall.
4691 40. When running the queue in the test harness, wait just a tad after forking a
4692 delivery process, to get repeatability of debugging output.
4694 41. Include certificate and key file names in error message when GnuTLS fails
4695 to set them up, because the GnuTLS error message doesn't include the name
4696 of the failing file when there is a problem reading it.
4698 42. Allow both -bf and -bF in the same test run.
4700 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
4702 44. The "Exiscan patch" is now merged into the mainline Exim source.
4704 45. Sometimes the final signoff response after QUIT could fail to get
4705 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4706 < 0 before doing a fflush(). This bug looks as though it goes back to the
4707 introduction of TLS in release 3.20, but "sometimes" must have been rare
4708 because the tests only now provoked it.
4710 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
4711 (this can affect the format of dates).
4713 47. exim_tidydb, when checking for the continued existence of a message for
4714 which it has found a message-specific retry record, was not finding
4715 messages that were in split spool directories. Consequently, it was
4716 deleting retry records that should have stayed in existence.
4718 48. Steve fixed some bugs in eximstats.
4720 49. The SPA authentication driver was not abandoning authentication and moving
4721 on to the next authenticator when an expansion was forced to fail,
4722 contradicting the general specification for all authenticators. Instead it
4723 was generating a temporary error. It now behaves as specified.
4725 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
4726 (the order specifies the preference for clients). The order is now AES256,
4727 AES128, 3DES, ARCFOUR128.
4729 51. Small patch to Sieve code - explicitly set From: when generating an
4732 52. Exim crashed if a remote delivery caused a very long error message to be
4733 recorded - for instance if somebody sent an entire SpamAssassin report back
4734 as a large number of 550 error lines. This bug was coincidentally fixed by
4735 increasing the size of one of Exim's internal buffers (big_buffer) that
4736 happened as part of the Exiscan merge. However, to be on the safe side, I
4737 have made the code more robust (and fixed the comments that describe what
4740 53. Now that there can be additional text after "Completed" in log lines (if
4741 the queue_time_overall log selector is set), a one-byte patch to exigrep
4742 was needed to allow it to recognize "Completed" as not the last thing in
4745 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
4746 patch that reportedly fixes this has been added. I am not expert enough to
4747 create a test for it. This is what the patch creator wrote:
4749 "I found a little strange behaviour of ldap code when working with
4750 Windows 2003 AD Domain, where users was placed in more than one
4751 Organization Units. When I tried to give exim partial DN, the exit code
4752 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
4753 But simultaneously result of request was absolutely normal ldap result,
4754 so I produce this patch..."
4756 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
4757 REFERENCE, so I have modified the code to exclude the patch when that macro
4760 55. Some experimental protocols are using DNS PTR records for new purposes. The
4761 keys for these records are domain names, not reversed IP addresses. The
4762 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4763 leaves it alone. Component reversal etc. now happens only for IP addresses.
4766 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
4768 57. Double the size of the debug message buffer (to 2048) so that more of very
4769 long debug lines gets shown.
4771 58. The exicyclog utility now does better if the number of log files to keep
4772 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
4774 59. Two changes related to the smtp_active_hostname option:
4776 (1) $smtp_active_hostname is now available as a variable.
4777 (2) The default for smtp_banner uses $smtp_active_hostname instead
4778 of $primary_hostname.
4780 60. The host_aton() function is supposed to be passed a string that is known
4781 to be a valid IP address. However, in the case of IPv6 addresses, it was
4782 not checking this. This is a hostage to fortune. Exim now panics and dies
4783 if the condition is not met. A case was found where this could be provoked
4784 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4785 components; fortuitously, this particular loophole had already been fixed
4786 by change 4.50/55 above.
4788 If there are any other similar loopholes, the new check in host_aton()
4789 itself should stop them being exploited. The report I received stated that
4790 data on the command line could provoke the exploit when Exim was running as
4791 exim, but did not say which command line option was involved. All I could
4792 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4793 running as the user.
4796 61. There was a buffer overflow vulnerability in the SPA authentication code
4797 (which came originally from the Samba project). I have added a test to the
4798 spa_base64_to_bits() function which I hope fixes it.
4801 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
4802 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
4804 63. The daemon start-up calls getloadavg() while still root for those OS that
4805 need the first call to be done as root, but it missed one case: when
4806 deliver_queue_load_max is set with deliver_drop_privilege. This is
4807 necessary for the benefit of the queue runner, because there is no re-exec
4808 when deliver_drop_privilege is set.
4810 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
4811 This has been fixed.
4813 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4814 was in use, was not putting the data itself into the right store pool;
4815 consequently, it could be overwritten for a subsequent message in the same
4816 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4819 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
4821 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
4822 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
4823 the function were treating the return as a boolean value, which happened to
4824 work because 0=false and not-0=true, but is not correct code.
4826 68. The host_aton() function was not handling scoped IPv6 addresses (those
4827 with, for example, "%eth0" on the end) correctly.
4829 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
4830 negated items (that is, ~something) in unsigned ints. Some compilers
4831 apparently mutter when there is no cast.
4833 70. If an address verification called from an ACL failed, and did not produce a
4834 user-specific message (i.e. there was only a "system" message), nothing was
4835 put in $acl_verify_message. In this situation, it now puts the system
4838 71. Change 4.23/11 added synchronization checking at the start of an SMTP
4839 session; change 4.31/43 added the unwanted input to the log line - except
4840 that it did not do this in the start of session case. It now does.
4842 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
4843 This is wrong and can cause the other end to generate a synchronization
4844 error if it is another Exim or anything else that does the synchronization
4845 check. A QUIT command is no longer sent after a timeout.
4847 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
4848 during host lookups.
4850 74. The defer_ok option of callout verification was not working if it was used
4851 when verifying addresses in header lines, that is, for this case:
4853 verify = header_sender/callout=defer_ok
4855 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
4856 those file descriptors could be used for SMTP connections. If anything
4857 wrote to stderr (the example that came up was "warn" in embedded Perl), it
4858 could be sent to the SMTP client, causing chaos. The daemon now opens
4859 stdin, stdout, and stderr to /dev/null when it puts itself into the
4862 76. Arrange for output from Perl's "warn" command to be written to Exim's main
4863 log by default. The user can override this with suitable Perl magic.
4865 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
4866 the log message when discard triggers, was not working for the DATA ACL or
4867 for the non-SMTP ACL.
4869 78. Error message wording change in sieve.c.
4871 79. If smtp_accept_max_per_host was set, the number of connections could be
4872 restricted to fewer than expected, because the daemon was trying to set up
4873 a new connection before checking whether the processes handling previous
4874 connections had finished. The check for completed processes is now done
4875 earlier. On busy systems, this bug wouldn't be noticed because something
4876 else would have woken the daemon, and it would have reaped the completed
4879 80. If a message was submitted locally by a user whose login name contained one
4880 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
4881 It caused a spool format error. I have fixed the spool reading code. A
4882 related problem was that the "from" clause in the Received: line became
4883 illegal because of the space(s). It is now covered by ${quote_local_part.
4885 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
4888 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
4890 83. Patch from Sieve maintainer to fix unterminated string problem in
4891 "vacation" handling.
4893 84. Some minor changes to the Linux configuration files to help with other
4894 OS variants using glibc.
4896 85. One more patch for Sieve to update vacation handling to latest spec.
4899 ----------------------------------------------------
4900 See the note above about the 4.44 and 4.50 releases.
4901 ----------------------------------------------------
4907 1. Change 4.43/35 introduced a bug that caused file counts to be
4908 incorrectly computed when quota_filecount was set in an appendfile
4911 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4912 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4915 3. Give more explanation in the error message when the command for a transport
4916 filter fails to execute.
4918 4. There are several places where Exim runs a non-Exim command in a
4919 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4920 was being done only for the command run by the queryprogram router. It is
4921 now done for all such subprocesses. The other cases are: ${run, transport
4922 filters, and the commands run by the lmtp and pipe transports.
4924 5. Some older OS have a limit of 256 on the maximum number of file
4925 descriptors. Exim was using setrlimit() to set 1000 as a large value
4926 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4927 systems. I've change it so that if it can't get 1000, it tries for 256.
4929 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4930 was an oversight, and furthermore, ever since the addition of extra
4931 controls (e.g. 4.43/32), the checks on when to allow different forms of
4932 "control" were broken. There should now be diagnostics for all cases when a
4933 control that does not make sense is encountered.
4935 7. $recipients is now available in the predata ACL (oversight).
4937 8. Tidy the search cache before the fork to do a delivery from a message
4938 received from the command line. Otherwise the child will trigger a lookup
4939 failure and thereby defer the delivery if it tries to use (for example) a
4940 cached ldap connection that the parent has called unbind on.
4942 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4943 of $address_data from the recipient verification was clobbered by the
4944 sender verification.
4946 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4947 was its contents. (It was OK if the option was not defined at all.)
4949 11. A "Completed" log line is now written for messages that are removed from
4950 the spool by the -Mrm option.
4952 12. $host_address is now set to the target address during the checking of
4953 ignore_target_hosts.
4955 13. When checking ignore_target_hosts for an ipliteral router, no host name was
4956 being passed; this would have caused $sender_host_name to have been used if
4957 matching the list had actually called for a host name (not very likely,
4958 since this list is usually IP addresses). A host name is now passed as
4961 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4962 code that specifies a non-restarting handler (typically sigaction() in
4963 modern systems) in an attempt to fix a rare and obscure crash bug.
4965 15. Narrowed the window for a race in the daemon that could cause it to ignore
4966 SIGCHLD signals. This is not a major problem, because they are used only to
4967 wake it up if nothing else does.
4969 16. A malformed maildirsize file could cause Exim to calculate negative values
4970 for the mailbox size or file count. Odd effects could occur as a result.
4971 The maildirsize information is now recalculated if the size or filecount
4974 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4975 support for a long time. Removed HAVE_SYS_VFS_H.
4977 18. Updated exipick to current release from John Jetmore.
4979 19. Allow an empty sender to be matched against a lookup in an address list.
4980 Previously the only cases considered were a regular expression, or an
4983 20. Exim went into a mad DNS lookup loop when doing a callout where the
4984 host was specified on the transport, if the DNS lookup yielded more than
4987 21. The RFC2047 encoding function was originally intended for short strings
4988 such as real names; it was not keeping to the 75-character limit for
4989 encoded words that the RFC imposes. It now respects the limit, and
4990 generates multiple encoded words if necessary. To be on the safe side, I
4991 have increased the buffer size for the ${rfc2047: expansion operator from
4994 22. Failure to deliver a bounce message always caused it to be frozen, even if
4995 there was an errors_to setting on the router. The errors_to setting is now
4998 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4999 canonical form (fully expanded) before being placed in
5000 $sender_host_address.
5002 24. Updated eximstats to version 1.33
5004 25. Include certificate and key file names in error message when GnuTLS fails
5005 to set them up, because the GnuTLS error message doesn't include the name
5006 of the failing file when there is a problem reading it.
5008 26. Expand error message when OpenSSL has problems setting up cert/key files.
5011 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
5012 (this can affect the format of dates).
5014 28. exim_tidydb, when checking for the continued existence of a message for
5015 which it has found a message-specific retry record, was not finding
5016 messages that were in split spool directories. Consequently, it was
5017 deleting retry records that should have stayed in existence.
5019 29. eximstats updated to version 1.35
5020 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
5021 1.35 - bugfix such that pie charts by volume are generated correctly
5023 30. The SPA authentication driver was not abandoning authentication and moving
5024 on to the next authenticator when an expansion was forced to fail,
5025 contradicting the general specification for all authenticators. Instead it
5026 was generating a temporary error. It now behaves as specified.
5028 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
5029 (the order specifies the preference for clients). The order is now AES256,
5030 AES128, 3DES, ARCFOUR128.
5032 31. Small patch to Sieve code - explicitly set From: when generating an
5035 32. Exim crashed if a remote delivery caused a very long error message to be
5036 recorded - for instance if somebody sent an entire SpamAssassin report back
5037 as a large number of 550 error lines. This bug was coincidentally fixed by
5038 increasing the size of one of Exim's internal buffers (big_buffer) that
5039 happened as part of the Exiscan merge. However, to be on the safe side, I
5040 have made the code more robust (and fixed the comments that describe what
5043 33. Some experimental protocols are using DNS PTR records for new purposes. The
5044 keys for these records are domain names, not reversed IP addresses. The
5045 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
5046 leaves it alone. Component reversal etc. now happens only for IP addresses.
5049 34. The host_aton() function is supposed to be passed a string that is known
5050 to be a valid IP address. However, in the case of IPv6 addresses, it was
5051 not checking this. This is a hostage to fortune. Exim now panics and dies
5052 if the condition is not met. A case was found where this could be provoked
5053 from a dnsdb PTR lookup with an IPv6 address that had more than 8
5054 components; fortuitously, this particular loophole had already been fixed
5055 by change 4.50/55 or 4.44/33 above.
5057 If there are any other similar loopholes, the new check in host_aton()
5058 itself should stop them being exploited. The report I received stated that
5059 data on the command line could provoke the exploit when Exim was running as
5060 exim, but did not say which command line option was involved. All I could
5061 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
5062 running as the user.
5065 35. There was a buffer overflow vulnerability in the SPA authentication code
5066 (which came originally from the Samba project). I have added a test to the
5067 spa_base64_to_bits() function which I hope fixes it.
5070 36. The daemon start-up calls getloadavg() while still root for those OS that
5071 need the first call to be done as root, but it missed one case: when
5072 deliver_queue_load_max is set with deliver_drop_privilege. This is
5073 necessary for the benefit of the queue runner, because there is no re-exec
5074 when deliver_drop_privilege is set.
5076 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
5077 was in use, was not putting the data itself into the right store pool;
5078 consequently, it could be overwritten for a subsequent message in the same
5079 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
5082 38. Sometimes the final signoff response after QUIT could fail to get
5083 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
5084 < 0 before doing a fflush(). This bug looks as though it goes back to the
5085 introduction of TLS in release 3.20, but "sometimes" must have been rare
5086 because the tests only now provoked it.
5092 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
5093 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
5094 Now it returns an integer. A number of calls were still expecting a T/F
5095 return. Fortuitously, in all cases, the tests worked in OK situations,
5096 which is the norm. However, things would have gone wrong on any write
5097 failures on the smtp file descriptor. This function is used when sending
5098 messages over SMTP and also when doing verify callouts.
5100 2. When Exim is called to do synchronous delivery of a locally submitted
5101 message (the -odf or -odi options), it no longer closes stderr before doing
5104 3. Implemented the mua_wrapper option.
5106 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
5108 5. Implemented the functions header_remove(), header_testname(),
5109 header_add_at_position(), and receive_remove_recipient(), and exported them
5112 6. If an ACL "warn" statement specified the addition of headers, Exim already
5113 inserted X-ACL-Warn: at the start if there was no header name. However, it
5114 was not making this test for the second and subsequent header lines if
5115 there were newlines in the string. This meant that an invalid header could
5116 be inserted if Exim was badly configured.
5118 7. Allow an ACL "warn" statement to add header lines at the start or after all
5119 the Received: headers, as well as at the end.
5121 8. Added the rcpt_4xx retry error code.
5123 9. Added postmaster_mailfrom=xxx to callout verification option.
5125 10. Added mailfrom=xxxx to the callout verification option, for verify=
5128 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
5129 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
5131 12. Inserted some casts to stop certain compilers warning when using pointer
5132 differences as field lengths or precisions in printf-type calls (mostly
5133 affecting debugging statements).
5135 13. Added optional readline() support for -be (dynamically loaded).
5137 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
5138 same clock tick as a message's arrival, so that its received time was the
5139 same as the "first fail" time on the retry record, and that message
5140 remained on the queue past the ultimate address timeout, every queue runner
5141 would try a delivery (because it was past the ultimate address timeout) but
5142 after another failure, the ultimate address timeout, which should have then
5143 bounced the address, did not kick in. This was a "< instead of <=" error;
5144 in most cases the first failure would have been in the next clock tick
5145 after the received time, and all would be well.
5147 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
5148 being recognized when the domain list was tested by the match_domain
5149 condition in an expansion string.
5151 16. Added the ${str2b64: operator.
5153 17. Exim was always calling setrlimit() to set a large limit for the number of
5154 processes, without checking whether the existing limit was already
5155 adequate. (It did check for the limit on file descriptors.) Furthermore,
5156 errors from getrlimit() and setrlimit() were being ignored. Now they are
5157 logged to the main and panic logs, but Exim does carry on, to try to do its
5158 job under whatever limits there are.
5160 18. Imported PCRE 5.0.
5162 19. Trivial typo in log message " temporarily refused connection" (the leading
5165 20. If the log selector return_path_on_delivery was set and an address was
5166 redirected to /dev/null, the delivery process crashed because it assumed
5167 that a return path would always be set for a "successful" delivery. In this
5168 case, the whole delivery is bypassed as an optimization, and therefore no
5171 21. Internal re-arrangement: the function for sending a challenge and reading
5172 a response while authentication was assuming a zero-terminated challenge
5173 string. It's now changed to take a pointer and a length, to allow for
5174 binary data in such strings.
5176 22. Added the cyrus_sasl authenticator (code supplied by MBM).
5178 23. Exim was not respecting finduser_retries when seeking the login of the
5179 uid under which it was called; it was always trying 10 times. (The default
5180 setting of finduser_retries is zero.) Also, it was sleeping after the final
5181 failure, which is pointless.
5183 24. Implemented tls_on_connect_ports.
5185 25. Implemented acl_smtp_predata.
5187 26. If the domain in control=submission is set empty, Exim assumes that the
5188 authenticated id is a complete email address when it generates From: or
5189 Sender: header lines.
5191 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
5192 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
5193 chown and chgrp in /bin and hostname in /usr/bin.
5195 28. Exim was keeping the "process log" file open after each use, just as it
5196 does for the main log. This opens the possibility of it remaining open for
5197 long periods when the USR1 signal hits a daemon. Occasional processlog
5198 errors were reported, that could have been caused by this. Anyway, it seems
5199 much more sensible not to leave this file open at all, so that is what now
5202 29. The long-running daemon process does not normally write to the log once it
5203 has entered its main loop, and it closes the log before doing so. This is
5204 so that log files can straightforwardly be renamed and moved. However,
5205 there are a couple of unusual error situations where the daemon does write
5206 log entries, and I had neglected to close the log afterwards.
5208 30. The text of an SMTP error response that was received during a remote
5209 delivery was being truncated at 512 bytes. This is too short for some of
5210 the long messages that one sometimes sees. I've increased the limit to
5213 31. It is now possible to make retry rules that apply only when a message has a
5214 specific sender, in particular, an empty sender.
5216 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
5217 it possible to be selective about when SMTP synchronization is enforced.
5219 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
5221 32. Implemented hosts_connection_nolog.
5223 33. Added an ACL for QUIT.
5225 34. Setting "delay_warning=" to disable warnings was not working; it gave a
5228 35. Added mailbox_size and mailbox_filecount to appendfile.
5230 36. Added control = no_multiline_responses to ACLs.
5232 37. There was a bug in the logic of the code that waits for the clock to tick
5233 in the case where the clock went backwards by a substantial amount such
5234 that the microsecond fraction of "now" was more than the microsecond
5235 fraction of "then" (but the whole seconds number was less).
5237 38. Added support for the libradius Radius client library this is found on
5238 FreeBSD (previously only the radiusclient library was supported).
5244 1. When certain lookups returned multiple values in the form name=value, the
5245 quoting of the values was not always being done properly. Specifically:
5246 (a) If the value started with a double quote, but contained no whitespace,
5248 (b) If the value contained whitespace other than a space character (i.e.
5249 tabs or newlines or carriage returns) it was not quoted.
5250 This fix has been applied to the mysql and pgsql lookups by writing a
5251 separate quoting function and calling it from the lookup code. The fix
5252 should probably also be applied to nisplus, ibase and oracle lookups, but
5253 since I cannot test any of those, I have not disturbed their existing code.
5255 2. A hit in the callout cache for a specific address caused a log line with no
5256 reason for rejecting RCPT. Now it says "Previous (cached) callout
5257 verification failure".
5259 3. There was an off-by-one bug in the queryprogram router. An over-long
5260 return line was truncated at 256 instead of 255 characters, thereby
5261 overflowing its buffer with the terminating zero. As well as fixing this, I
5262 have increased the buffer size to 1024 (and made a note to document this).
5264 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
5265 when Exim is waiting for an SMTP response from a remote server, Exim
5266 restarts its select() call on the socket, thereby resetting its timeout.
5267 This is not a problem when such interrupts are rare. Somebody set up a cron
5268 job to run exiwhat every 2 minutes, which is less than the normal select()
5269 timeout (5 or 10 minutes). This meant that the select() timeout never
5270 kicked in because it was always reset. I have fixed this by comparing the
5271 time when an interrupt arrives with the time at the start of the first call
5272 to select(). If more time than the timeout has elapsed, the interrupt is
5273 treated as a timeout.
5275 5. Some internal re-factoring in preparation for the addition of Sieve
5276 extensions (by MH). In particular, the "personal" test is moved to a
5277 separate function, and given an option for scanning Cc: and Bcc: (which is
5278 not set for Exim filters).
5280 6. When Exim created an email address using the login of the caller as the
5281 local part (e.g. when creating a From: or Sender: header line), it was not
5282 quoting the local part when it contained special characters such as @.
5284 7. Installed new OpenBSD configuration files.
5286 8. Reworded some messages for syntax errors in "and" and "or" conditions to
5287 try to make them clearer.
5289 9. Callout options, other than the timeout value, were being ignored when
5290 verifying sender addresses in header lines. For example, when using
5292 verify = header_sender/callout=no_cache
5294 the cache was (incorrectly) being used.
5296 10. Added a missing instance of ${EXE} to the exim_install script; this affects
5297 only the Cygwin environment.
5299 11. When return_path_on_delivery was set as a log selector, if different remote
5300 addresses in the same message used different return paths and parallel
5301 remote delivery occurred, the wrong values would sometimes be logged.
5302 (Whenever a remote delivery process finished, the return path value from
5303 the most recently started remote delivery process was logged.)
5305 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
5306 lines when AUTH and/or TLS are in use. This is the "received protocol"
5307 field. Exim used to use "asmtp" for authenticated SMTP, without any
5308 indication (in the protocol name) for TLS use. Now it follows the RFC and
5309 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
5310 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
5311 names appear in log lines as well as in Received: header lines.
5313 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
5314 extensions, and comparison tests, and to fix some bugs.
5316 14. Changes to the "personal" filter test:
5318 (1) The test was buggy in that it was just doing the equivalent of
5319 "contains" tests on header lines. For example, if a user's address was
5320 anne@some.where, the "personal" test would incorrectly be true for
5322 To: susanne@some.where
5324 This test is now done by extracting each address from the header in turn,
5325 and checking the entire address. Other tests that are part of "personal"
5326 are now done using regular expressions (for example, to check local parts
5327 of addresses in From: header lines).
5329 (2) The list of non-personal local parts in From: addresses has been
5330 extended to include "listserv", "majordomo", "*-request", and "owner-*",
5331 taken from the Sieve specification recommendations.
5333 (3) If the message contains any header line starting with "List-" it is
5334 treated as non-personal.
5336 (4) The test for "circular" in the Subject: header line has been removed
5337 because it now seems ill-conceived.
5339 15. Minor typos in src/EDITME comments corrected.
5341 16. Installed latest exipick from John Jetmore.
5343 17. If headers_add on a router specified a text string that was too long for
5344 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
5345 of string_sprintf() is now avoided.
5347 18. $message_body_size was not set (it was always zero) when running the DATA
5348 ACL and the local_scan() function.
5350 19. For the "mail" command in an Exim filter, no default was being set for
5351 the once_repeat time, causing a random time value to be used if "once" was
5352 specified. (If the value happened to be <= 0, no repeat happened.) The
5353 default is now 0s, meaning "never repeat". The "vacation" command was OK
5354 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
5355 (I found it when inspecting the code).
5357 20. There is now an overall timeout for performing a callout verification. It
5358 defaults to 4 times the callout timeout, which applies to individual SMTP
5359 commands during the callout. The overall timeout applies when there is more
5360 than one host that can be tried. The timeout is checked before trying the
5361 next host. This prevents very long delays if there are a large number of
5362 hosts and all are timing out (e.g. when the network connections are timing
5363 out). The value of the overall timeout can be changed by specifying an
5364 additional sub-option for "callout", called "maxwait". For example:
5366 verify = sender/callout=5s,maxwait=20s
5368 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
5369 to the end before writing, but this should make it even safer).
5371 22. Exim was forgetting that it had advertised PIPELINING for the second and
5372 subsequent messages on an SMTP connection. It was also not resetting its
5373 memory on STARTTLS and an internal HELO.
5375 23. When Exim logs an SMTP synchronization error within a session, it now
5376 records whether PIPELINING has been advertised or not.
5378 24. Added 3 instances of "(long int)" casts to time_t variables that were being
5379 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
5380 rather than long int.
5382 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
5384 26. Added the never_mail option to autoreply.
5390 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
5391 crash if the getsockname() call failed; this can happen if a connection is
5392 closed very soon after it is established. The problem was simply in the
5393 order in which certain operations were done, causing Exim to try to write
5394 to the SMTP stream before it had set up the file descriptor. The bug has
5395 been fixed by making things happen in the correct order.
5401 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
5402 before the connection was closed, thus losing the rejection response.
5404 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
5405 some early Solaris releases, but causes trouble in current releases where
5406 socklen_t is defined.
5408 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
5411 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
5414 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
5415 that releasing the top of it at the end releases what was used for sub-
5416 expansions (unless the block got too big). However, discard this block if
5417 the first thing is a variable or header, so that we can use its block when
5418 it is dynamic (useful for very large $message_headers, for example).
5420 6. Lookups now cache *every* query, not just the most recent. A new, separate
5421 store pool is used for this. It can be recovered when all lookup caches are
5422 flushed. Lookups now release memory at the end of their result strings.
5423 This has involved some general refactoring of the lookup sources.
5425 7. Some code has been added to the store_xxx() functions to reduce the amount
5426 of flapping under certain conditions.
5428 8. log_incoming_interface used to affect only the <= reception log lines. Now
5429 it causes the local interface and port to be added to several more SMTP log
5430 lines, for example "SMTP connection from", and rejection lines.
5432 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
5434 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
5436 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
5437 could be overwritten at the end of the current message (or the start of a
5438 new message if it was set in a HELO ACL). The value is now preserved for
5439 the duration of the SMTP connection.
5441 12. If a transport had a headers_rewrite setting, and a matching header line
5442 contained an unqualified address, that address was qualified, even if it
5443 did not match any rewriting rules. The underlying bug was that the values
5444 of the flags that permit the existence of unqualified sender and recipient
5445 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
5446 non-local messages, and by -bnq for local messages) were not being
5447 preserved with the message after it was received.
5449 13. When Exim was logging an SMTP synchronization error, it could sometimes log
5450 "next input=" as part of the text comprising the host identity instead of
5451 the correct text. The code was using the same buffer for two different
5452 strings. However, depending on which order the printing function evaluated
5453 its arguments, the bug did not always show up. Under Linux, for example, my
5454 test suite worked just fine.
5456 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
5457 doesn't work with some older versions of Perl. It has been changed to "my",
5458 which in any case is probably the better facility to use.
5460 15. A really picky compiler found some instances of statements for creating
5461 error messages that either had too many or two few arguments for the format
5464 16. The size of the buffer for calls to the DNS resolver has been increased
5465 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
5466 for addresses that have a lot of PTR records. This alleviates a problem; it
5467 does not fully solve it.
5469 17. A dnsdb lookup for PTR records that receives more data than will fit in the
5470 buffer now truncates the list and logs the incident, which is the same
5471 action as happens when Exim is looking up a host name and its aliases.
5472 Previously in this situation something unpredictable would happen;
5473 sometimes it was "internal error: store_reset failed".
5475 18. If a server dropped the connection unexpectedly when an Exim client was
5476 using GnuTLS and trying to read a response, the client delivery process
5477 crashed while trying to generate an error log message.
5479 19. If a "warn" verb in an ACL added multiple headers to a message in a single
5480 string, for example:
5482 warn message = H1: something\nH2: something
5484 the text was added as a single header line from Exim's point of view
5485 though it ended up OK in the delivered message. However, searching for the
5486 second and subsequent header lines using $h_h2: did not work. This has been
5487 fixed. Similarly, if a system filter added multiple headers in this way,
5488 the routers could not see them.
5490 20. Expanded the error message when iplsearch is called with an invalid key to
5491 suggest using net-iplsearch in a host list.
5493 21. When running tests using -bh, any delays imposed by "delay" modifiers in
5494 ACLs are no longer actually imposed (and a message to that effect is
5497 22. If a "gecos" field in a passwd entry contained escaped characters, in
5498 particular, if it contained a \" sequence, Exim got it wrong when building
5499 a From: or a Sender: header from that name. A second bug also caused
5500 incorrect handling when an unquoted " was present following a character
5501 that needed quoting.
5503 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
5504 was not being matched caselessly.
5506 24. Arranged for all hyphens in the exim.8 source to be escaped with
5509 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
5510 properly. Recipient callout cache records were still being keyed to include
5511 the sender, even when use_sender was set false. This led to far more
5512 callouts that were necessary. The sender is no longer included in the key
5513 when use_sender is false.
5515 26. Added "control = submission" modifier to ACLs.
5517 27. Added the ${base62d: operator to decode base 62 numbers.
5519 28. dnsdb lookups can now access SRV records.
5521 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
5522 the configuration file.
5524 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
5525 (-v) mode. This makes the output for a verbose queue run more intelligible.
5527 31. Added a use_postmaster feature to recipient callouts.
5529 32. Added the $body_zerocount variable, containing the number of binary zero
5530 bytes in the message body.
5532 33. The time of last modification of the "new" subdirectory is now used as the
5533 "mailbox time last read" when there is a quota error for a maildir
5536 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
5538 35. Added +ignore_unknown as a special item in host lists.
5540 36. Code for decoding IPv6 addresses in host lists is now included, even if
5541 IPv6 support is not being compiled. This fixes a bug in which an IPv6
5542 address was recognized as an IP address, but was then not correctly decoded
5543 into binary, causing unexpected and incorrect effects when compared with
5550 1. Very minor rewording of debugging text in manualroute to say "list of
5551 hosts" instead of "hostlist".
5553 2. If verify=header_syntax was set, and a header line with an unqualified
5554 address (no domain) and a large number of spaces between the end of the
5555 name and the colon was received, the reception process suffered a buffer
5556 overflow, and (when I tested it) crashed. This was caused by some obsolete
5557 code that should have been removed. The fix is to remove it!
5559 3. When running in the test harness, delay a bit after writing a bounce
5560 message to get a bit more predictability in the log output.
5562 4. Added a call to search_tidyup() just before forking a reception process. In
5563 theory, someone could use a lookup in the expansion of smtp_accept_max_
5564 per_host which, without the tidyup, could leave open a database connection.
5566 5. Added the variables $recipient_data and $sender_data which get set from a
5567 lookup success in an ACL "recipients" or "senders" condition, or a router
5568 "senders" option, similar to $domain_data and $local_part_data.
5570 6. Moved the writing of debug_print from before to after the "senders" test
5573 7. Change 4.31/66 (moving the time when the Received: is generated) caused
5574 problems for message scanning, either using a data ACL, or using
5575 local_scan() because the Received: header was not generated till after they
5576 were called (in order to set the time as the time of reception completion).
5577 I have revised the way this works. The header is now generated after the
5578 body is received, but before the ACL or local_scan() are called. After they
5579 are run, the timestamp in the header is updated.
5585 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
5586 before starting a queue runner without re-exec. This happened only when
5587 deliver_drop_privilege was set or when the Exim user was set to root. The
5588 effect of the bug was that timeouts during subsequent deliveries caused
5589 crashes instead of being properly handled. The handler is now left at its
5590 default (and expected) setting.
5592 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
5593 message, again when deliver_drop_privilege is set or Exim is run as root.
5594 The bug described in (1) was not present in this case, but the tidying up
5595 of the other signals was missing. I have made the two cases consistent.
5597 3. The ignore_target_hosts setting on a manualroute router was being ignored
5598 for hosts that were looked up using the /MX notation.
5600 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
5603 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
5604 operated on the sender address. After changing the $sender_address to <>
5605 for the sender address verify, Exim was re-instated it as the original
5606 (before rewriting) address, but remembering that it had rewritten it, so it
5607 wasn't rewriting it again. This bug also had the effect of breaking the
5608 sender address verification caching when the sender address was rewritten.
5610 6. The ignore_target_hosts option was being ignored by the ipliteral router.
5611 This has been changed so that if the ip literal address matches
5612 ignore_target_hosts, the router declines.
5614 7. Added expansion conditions match_domain, match_address, and match_local_
5615 part (NOT match_host).
5617 8. The placeholder for the Received: header didn't have a length field set.
5619 9. Added code to Exim itself and to exim_lock to test for a specific race
5620 condition that could lead to file corruption when using MBX delivery. The
5621 issue is with the lockfile that is created in /tmp. If this file is removed
5622 after a process has opened it but before that process has acquired a lock,
5623 there is the potential for a second process to recreate the file and also
5624 acquire a lock. This could lead to two Exim processes writing to the file
5625 at the same time. The added code performs the same test as UW imapd; it
5626 checks after acquiring the lock that its file descriptor still refers to
5627 the same named file.
5629 10. The buffer for building added header lines was of fixed size, 8192 bytes.
5630 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
5633 11. Added the smtp_active_hostname option. If used, this will typically be made
5634 to depend on the incoming interface address. Because $interface_address is
5635 not set up until the daemon has forked a reception process, error responses
5636 that can happen earlier (such as "too many connections") no longer contain
5639 12. If an expansion in a condition on a "warn" statement fails because a lookup
5640 defers, the "warn" statement is abandoned, and the next ACL statement is
5641 processed. Previously this caused the whole ACL to be aborted.
5643 13. Added the iplsearch lookup type.
5645 14. Added ident_timeout as a log selector.
5647 15. Added tls_certificate_verified as a log selector.
5649 16. Added a global option tls_require_ciphers (compare the smtp transport
5650 option of the same name). This controls incoming TLS connections.
5652 17. I finally figured out how to make tls_require_ciphers do a similar thing
5653 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
5654 before starting the TLS session.
5656 18. Tabs are now shown as \t in -bP output.
5658 19. If the log selector return_path_on_delivery was set, Exim crashed when
5659 bouncing a message because it had too many Received: header lines.
5661 20. If two routers both had headers_remove settings, and the first one included
5662 a superfluous trailing colon, the final name in the first list and the
5663 first name in the second list were incorrectly joined into one item (with a
5664 colon in the middle).
5670 1. Added -C and -D options to the exinext utility, mainly to make it easier
5671 to include in the automated testing, but these could be helpful when
5672 multiple configurations are in use.
5674 2. The exinext utility was not formatting the output nicely when there was
5675 an alternate port involved in the retry record key, nor when there was a
5676 message id as well (for retries that were specific to a specific message
5677 and a specific host). It was also confused by IPv6 addresses, because of
5678 the additional colons they contain. I have fixed the IPv4 problem, and
5679 patched it up to do a reasonable job for IPv6.
5681 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
5682 delivery, the log line now contains "pipelined" if PIPELINING was used.
5684 4. An SMTP transport process used to panic and die if the bind() call to set
5685 an explicit outgoing interface failed. This has been changed; it is now
5686 treated in the same way as a connect() failure.
5688 5. A reference to $sender_host_name in the part of a conditional expansion
5689 that was being skipped was still causing a DNS lookup. This no longer
5692 6. The def: expansion condition was not recognizing references to header lines
5693 that used bh_ and bheader_.
5695 7. Added the _cache feature to named lists.
5697 8. The code for checking quota_filecount in the appendfile transport was
5698 allowing one more file than it should have been.
5700 9. For compatibility with Sendmail, the command line option
5708 and sets the incoming protocol and host name (for trusted callers). The
5709 host name and its colon can be omitted when only the protocol is to be set.
5710 Note the Exim already has two private options, -pd and -ps, that refer to
5711 embedded Perl. It is therefore impossible to set a protocol value of "d" or
5712 "s", but I don't think that's a major issue.
5714 10. A number of refactoring changes to the code, none of which should affect
5717 (a) The number of logging options was getting close to filling up the
5718 32-bit word that was used as a bit map. I have split them into two classes:
5719 those that are passed in the argument to log_write(), and those that are
5720 only ever tested independently outside of that function. These are now in
5721 separate 32-bit words, so there is plenty of room for expansion again.
5722 There is no change in the user interface or the logging behaviour.
5724 (b) When building, for example, log lines, the code previously used a
5725 macro that called string_cat() twice, in order to add two strings. This is
5726 not really sufficiently general. Furthermore, there was one instance where
5727 it was actually wrong because one of the argument was used twice, and in
5728 one call a function was used. (As it happened, calling the function twice
5729 did not affect the overall behaviour.) The macro has been replaced by a
5730 function that can join an arbitrary number of extra strings onto a growing
5733 (c) The code for expansion conditions now uses a table and a binary chop
5734 instead of a serial search (which was left over from when there were very
5735 few conditions). Also, it now recognizes conditions like "pam" even when
5736 the relevant support is not compiled in: a suitably worded error message is
5737 given if an attempt is made to use such a condition.
5739 11. Added ${time_interval:xxxxx}.
5741 12. A bug was causing one of the ddress fields not to be passed back correctly
5742 from remote delivery subprocesses. The field in question was not being
5743 subsequently used, so this caused to problems in practice.
5745 13. Added new log selectors queue_time and deliver_time.
5747 14. Might have fixed a bug in maildirsizefile handling that threw up
5748 "unexpected character" debug warnings, and recalculated the data
5749 unnecessarily. In any case, I expanded the warning message to give more
5752 15. Added the message "Restricted characters in address" to the statements in
5753 the default ACL that block characters like @ and % in local parts.
5755 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
5756 Three changes have been made:
5758 (a) There was a serious bug; a negative response to MAIL caused the whole
5759 recipient domain to be cached as invalid, thereby blocking all messages
5760 to all local parts at the same domain, from all senders. This bug has
5761 been fixed. The domain is no longer cached after a negative response to
5762 MAIL if the sender used is not empty.
5764 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
5767 (c) A new callout option, "use_sender" has been added for people who want
5768 the modified behaviour.
5774 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
5777 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
5778 indeed breaks things for older releases.
5780 3. Added additional logging to the case where there is a problem reading data
5781 from a filter that is running in a subprocess using a pipe, in order to
5782 try to track down a specific problem.
5784 4. Testing facility fudge: when running in the test harness and attempting
5785 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
5786 getting "No route to host". Convert this to a timeout.
5788 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
5791 6. Some OS don't have socklen_t but use size_t instead. This affects the
5792 fifth argument of getsockopt() amongst other things. This is now
5793 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
5794 can be set for individual OS. I have set it for SunOS5, OSF1, and
5795 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
5796 some earlier ones do not.
5798 7. Change 4.30/15 was not doing the test caselessly.
5800 8. The standard form for an IPv6 address literal was being rejected by address
5801 parsing in, for example, MAIL and RCPT commands. An example of this kind of
5802 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5803 this, as well as the form without the "IPv6" on the front (but only when
5804 address literals are enabled, of course).
5806 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
5808 10. Exim crashed if a message with an empty sender address specified by -f
5809 encountered a router with an errors_to setting. This could be provoked only
5810 by a command such as
5814 where an empty string was supplied; "<>" did not hit this bug.
5816 11. Installed PCRE release 4.5.
5818 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
5819 remained set. It is now erased.
5821 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
5822 times from message ids (which are base 36 rather than the normal 62).
5824 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
5825 were being counted as actual protocol errors, and logged if the log
5826 selector +smtp_protocol_error was set. One cannot be perfect in this test,
5827 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
5828 and DATA following a set of rejected RCPTs do not count as protocol errors.
5829 In other words, Exim assumes they were pipelined, though this may not
5830 actually be the case. Of course, in all cases the client gets an
5831 appropriate error code.
5833 15. If a lookup fails in an ACL condition, a message about the failure may
5834 be available; it is used if testing the ACL cannot continue, because most
5835 such messages specify what the cause of the deferral is. However, some
5836 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
5837 that caused an old message to be retained and used if a later statement
5838 caused a defer, replacing the real cause of the deferral.
5840 16. If an IP address had so many PTR records that the DNS lookup buffer
5841 was not large enough to hold them, Exim could crash while trying to process
5842 the truncated data. It now detects and logs this case.
5844 17. Further to 4.21/58, another change has been made: if (and only if) the
5845 first line of a message (the first header line) ends with CRLF, a bare LF
5846 in a subsequent header line has a space inserted after it, so as not to
5847 terminate the header.
5849 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
5850 unnecessarily, used atoi() instead of strtol(), and didn't check the
5851 termination when getting file sizes from file names by regex.
5853 19. Completely re-implemented the support for maildirsize files, in the light
5854 of a number of problems with the previous contributed implementation
5855 (4.30/29). In particular:
5857 . If the quota is zero, the maildirsize file is maintained, but no quota is
5860 . If the maildir directory does not exist, it is created before any attempt
5861 to write a maildirsize file.
5863 . The quota value in the file is just a cache; if the quota is changed in
5864 the transport, the new value overrides.
5866 . A regular expression is available for excluding directories from the
5869 20. The autoreply transport checks the characters in options that define the
5870 message's headers; it allows continued headers, but it was checking with
5871 isspace() after an embedded newline instead of explicitly looking for a
5874 21. If all the "regular" hosts to which an address was routed had passed their
5875 expiry times, and had not reached their retry times, the address was
5876 bounced, even if fallback hosts were defined. Now Exim should go on to try
5879 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
5880 equivalent code in the SMTP transport. Some hosts send humungous responses
5881 to HELO/EHLO, more than 1024 it seems.
5883 23. Refactoring: code in filter.c used (void *) for "any old type" but this
5884 gives compiler warnings in some environments. I've now done it "properly",
5887 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
5888 (because of problems with the built-in one) was declared to return uschar *
5889 instead of char *, causing compiler failure.
5891 25. Fixed a file descriptor leak when processing alias/forward files.
5893 26. Fixed a minor format string issue in dbfn.c.
5895 27. Typo in exim.c: ("dmbnz" for "dbmnz").
5897 28. If a filter file refered to $h_xxx or $message_headers, and the headers
5898 contained RFC 2047 "words", Exim's memory could, under certain conditions,
5901 29. When a sender address is verified, it is cached, to save repeating the test
5902 when there is more than one recipient in a message. However, when the
5903 verification involves a callout, it is possible for different callout
5904 options to be set for different recipients. It is too complicated to keep
5905 track of this in the cache, so now Exim always runs a verification when a
5906 callout is required, relying on the callout cache for the optimization.
5907 The overhead is duplication of the address routing, but this should not be
5910 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
5911 to be verified with callout=postmaster, and the main callout worked but the
5912 postmaster check failed, the verification correctly failed. However, if a
5913 subsequent RCPT command asked for sender verification *without* the
5914 postmaster check, incorrect caching caused this verification also to fail,
5917 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
5918 it was not caching the DNS options (qualify_single, search_parents) that
5919 were used when the lookup failed. A subsequent lookup with different
5920 options therefore always gave the same answer, though there were cases
5921 where it should not have. (Example: a "domains = !$mx_any" option on a
5922 dnslookup router: the "domains" option is always processed without any
5923 widening, but the router might have qualify_single set.) Now Exim uses the
5924 cached value only when the same options are set.
5926 32. Added John Jetmore's "exipick" utility to the distribution.
5928 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
5929 than a timeout (e.g. a certificate is required, and is not provided), an
5930 Exim server now closes the connection immediately. Previously it waited for
5931 the client to close - but if the client is SSL, it seems that they each
5932 wait for each other, leading to a delay before one of them times out.
5934 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
5935 maintained 0.8.x compatibility because I don't think many are using it, and
5936 it is clearly obsolete.
5938 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
5941 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
5942 client certificate was expired. A simple patch fixes this, though I don't
5943 understand the full logic of why the verify callback is called multiple
5946 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
5947 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
5948 which causes problems with some clients (such as the Certicom SSL Plus
5949 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
5950 disables the coutermeasure allowing Eudora to connect."
5952 38. Exim was not checking that a write() to a log file succeeded. This could
5953 lead to Bad Things if a log got too big, in particular if it hit a file
5954 size limit. Exim now panics and dies if it cannot write to a log file, just
5955 as it does if it cannot open a log file.
5957 39. Modified OS/Makefile-Linux so that it now contains
5959 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
5961 The two -D definitions ensure that Exim is compiled with large file
5962 support, which makes it possible to handle log files that are bigger than
5965 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
5966 instance) a domain was checked against a named list that involved a lookup,
5967 causing $domain_data to be set, then another domain was checked against the
5968 same list, then the first domain was re-checked, the value of $domain_data
5969 after the final check could be wrong. In particular, if the second check
5970 failed, it could be set empty. This bug probably also applied to
5973 41. The strip_trailing_dot option was not being applied to the address given
5974 with the -f command-line option.
5976 42. The code for reading a message's header from the spool was incrementing
5977 $received_count, but never initializing it. This meant that the value was
5978 incorrect (doubled) while delivering a message in the same process in which
5979 it was received. In the most common configuration of Exim, this never
5980 happens - a fresh exec is done - but it can happen when
5981 deliver_drop_privilege is set.
5983 43. When Exim logs an SMTP synchronization error - client data sent too soon -
5984 it now includes up to 150 characters of the unexpected data in the log
5987 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
5988 and building data strings. The size of both of these buffers was 10 000
5989 bytes - far larger than anybody would *ever* want, thought I. Needless to
5990 say, somebody hit the limit. I have increased the maximum line length to
5991 20 000 and the maximum data length of concatenated lines to 100 000. I have
5992 also fixed two bugs, because there was no checking on these buffers. Tsk,
5993 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
5994 buffer is too small.
5996 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
5997 lsearch lookups. Now it does.
5999 46. When parsing a route_list item in a manualroute router, a fixed-length
6000 buffer was used for the list of hosts. I made this 1024 bytes long,
6001 thinking that nobody would ever have a list of hosts that long. Wrong.
6002 Somebody had a whole pile of complicated expansion conditions, and the
6003 string was silently truncated, leading to an expansion error. It turns out
6004 that it is easier to change to an unlimited length (owing to other changes
6005 that have happened since this code was originally written) than to build
6006 structure for giving a limitation error. The length of the item that
6007 expands into the list of hosts is now unlimited.
6009 47. The lsearch lookup could not handle data where the length of text line was
6010 more than 4095 characters. Such lines were truncated, leading to shortened
6011 data being returned. It should now handle lines of any length.
6013 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
6014 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
6017 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
6018 replacing strings such as BIN_DIRECTORY in a source file: the replacement
6019 no longer happens in comment lines. A list of replacements is now placed
6020 at the head of all of the source files, except those whose only change is
6021 to replace PERL_COMMAND in the very first #! line.
6023 50. Replaced the slow insertion sort in queue.c, for sorting the list of
6024 messages on the queue, with a bottom-up merge sort, using code contributed
6025 by Michael Haardt. This should make operations like -bp somewhat faster on
6026 large queues. It won't affect queue runners, except when queue_run_in_order
6029 51. Installed eximstats 1.31 in the distribution.
6031 52. Added support for SRV lookups to the dnslookup router.
6033 53. If an ACL referred to $message_body or $message_body_end, the value was not
6034 reset for any messages that followed in the same SMTP session.
6036 54. The store-handling optimization for building very long strings was not
6037 differentiating between the different store pools. I don't think this
6038 actually made any difference in practice, but I've tidied it.
6040 55. While running the routers to verify a sender address, $sender_address
6041 was still set to the sender address. This is wrong, because when routing to
6042 send a bounce to the sender, it would be empty. Therefore, I have changed
6043 it so that, while verifying a sender address, $sender_address is set to <>.
6044 (There is no change to what happens when verifying a recipient address.)
6046 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
6047 target A or AAAA records (if not already returned) without resetting the
6048 qualify_single or search_parents options of the DNS resolver. These are
6049 inappropriate in this case because the targets of MX and SRV records must
6050 be FQDNs. A broken DNS record could cause trouble if it happened to have a
6051 target that, when qualified, matched something in the local domain. These
6052 two options are now turned off when doing these lookups.
6054 57. It seems that at least some releases of Reiserfs (which does not have the
6055 concept of a fixed number of inodes) returns zero and not -1 for the
6056 number of available inodes. This interacted badly with check_spool_inodes,
6057 which assumed that -1 was the "no such thing" setting. What I have done is
6058 to check that the total number of inodes is greater than zero before doing
6059 the test of how many are available.
6061 58. When a "warn" ACL statement has a log_message modifier, the message is
6062 remembered, and not repeated. This is to avoid a lot of repetition when a
6063 message has many recipients that cause the same warning to be written.
6064 However, Exim was preserving the list of already written lines for an
6065 entire SMTP session, which doesn't seem right. The memory is now reset if a
6066 new message is started.
6068 59. The "rewrite" debugging flag was not showing the result of rewriting in the
6069 debugging output unless log_rewrite was also set.
6071 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
6072 of (int)(handle) when we know that handle contains (void *)(-1).
6074 61. The Exim daemon panic-logs an error return when it closes the incoming
6075 connection. However "connection reset by peer" seems to be common, and
6076 isn't really an error worthy of noting specially, so that particular error
6079 62. When Exim is trying to find all the local interfaces, it used to panic and
6080 die if the ioctl to get the interface flags failed. However, it seems that
6081 on at least one OS (Solaris 9) it is possible to have an interface that is
6082 included in the list of interfaces, but for which you get a failure error
6083 for this call. This happens when the interface is not "plumbed" into a
6084 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
6085 failure of the "get flags" call assumes that the interface is down.
6087 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
6088 makes life easier for people who are doing arithmetic on fields extracted
6089 from dates, where you often get leading zeros that should not be
6090 interpreted as octal.
6092 64. Added qualify_domain to the redirect router, to override the global
6095 65. If a pathologically long header line contained very many addresses (the
6096 report of this problem mentioned 10 000) and each of them was rewritten,
6097 Exim could use up a very large amount of memory. (It kept on making new
6098 copies of the header line as it rewrote, and never released the old ones.)
6099 At the expense of a bit more processing, the header rewriting function has
6100 been changed so that it no longer eats memory in this way.
6102 66. The generation of the Received: header has been moved from the time that a
6103 message starts to be received, to the time that it finishes. The timestamp
6104 in the Received: header should now be very close to that of the <= log
6105 line. There are two side-effects of this change:
6107 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
6108 logged header lines no longer include the local Received: line, because
6109 it has not yet been created. The same applies to a copy of the message
6110 that is returned to a non-SMTP sender when a message is rejected.
6112 (b) When a filter file is tested using -bf, no additional Received: header
6113 is added to the test message. After some thought, I decided that this
6116 This change does not affect the value of $received_for. It is still set
6117 after address rewriting, but before local_scan() is called.
6119 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
6121 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
6122 gave an unhelpful panic error message, and a defer error. I have managed to
6123 change this behaviour so that it now rejects any supplied certificate,
6124 which seems right, as the list of acceptable certificates is empty.
6126 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
6127 gave an unhelpful defer error. I have not managed to make this reject any
6128 supplied certificates, but the error message it gives is "no certificate
6129 supplied", which is not helpful.
6131 70. exigrep's output now also includes lines that are not associated with any
6132 message, but which match the given pattern. Implemented by a patch from
6133 Martin Sluka, which also tidied up the Perl a bit.
6135 71. Recipient callout verification, like sender verification, was using <> in
6136 the MAIL FROM command. This isn't really the right thing, since the actual
6137 sender may affect whether the remote host accepts the recipient or not. I
6138 have changed it to use the actual sender in the callout; this means that
6139 the cache record is now keyed on a recipient/sender pair, not just the
6140 recipient address. There doesn't seem to be a real danger of callout loops,
6141 since a callout by the remote host to check the sender would use <>.
6142 [SEE ABOVE: changed after hitting problems.]
6144 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
6145 temporary errors. However, in the case of such a code being given after
6146 the end of a data transmission (i.e. after ".") Exim was failing to write
6147 a retry record for the message. (Yes, there was some broken host that was
6148 actually sending 8xx at this point.)
6150 73. An unknown lookup type in a host list could cause Exim to panic-die when
6151 the list was checked. (An example that provoked this was putting <; in the
6152 middle of a list instead of at the start.) If this happened during a DATA
6153 ACL check, a -D file could be left lying around. This kind of configuration
6154 error no longer causes Exim to die; instead it causes a defer error. The
6155 incident is still logged to the main and panic logs.
6157 74. Buglet left over from Exim 3 conversion. The message "too many messages
6158 in one connection" was written to the rejectlog but not the mainlog, except
6159 when address rewriting (yes!) was being logged.
6161 75. Added write_rejectlog option.
6163 76. When a system filter was run not as root (that is, when system_filter_user
6164 was set), the values of the $n variables were not being returned to the
6165 main process; thus, they were not subsequently available in the $sn
6168 77. Added +return_path_on_delivery log selector.
6170 78. A connection timeout was being treated differently from recipients deferred
6171 when testing hosts_max_try with a message that was older than the host's
6172 retry timeout. (The host should not be counted, thus allowing all hosts to
6173 be tried at least once before bouncing.) This may have been the cause of an
6174 occasionally reported bug whereby a message would remain on the queue
6175 longer than the retry timeout, but would be bounced if a delivery was
6176 forced. I say "may" because I never totally pinned down the problem;
6177 setting up timeout/retry tests is difficult. See also the next item.
6179 79. The ultimate address timeout was not being applied to errors that involved
6180 a combination of host plus message (for example, a timeout on a MAIL
6181 command). When an address resolved to a number of possible hosts, and they
6182 were not all tried for each delivery (e.g. because of hosts_max_try), a
6183 message could remain on the queue longer than the retry timeout.
6185 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
6188 81. Fixed an obscure SMTP outgoing bug which required at least the following
6189 conditions: (a) there was another message waiting for the same server;
6190 (b) the server returned 5xx to all RCPT commands in the first message so
6191 that the message was not completed; (c) the server dropped the connection
6192 or gave a negative response to the RSET that Exim sends to abort the
6193 transaction. The observed case was a dropped connection after DATA that had
6194 been sent in pipelining mode. That is, the server had advertised PIPELINING
6195 but was not implementing it correctly. The effect of the bug was incorrect
6196 behaviour, such as trying another host, and this could lead to a crash.
6202 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
6203 and daemon.c were passed as pointers to ints; they should have been
6204 pointers to socklen_t variables (which are typically unsigned ints).
6206 2. Some signed/unsigned type warnings in the os.c file for Linux have been
6209 3. Fixed a really odd bug that affected only the testing scheme; patching a
6210 certain fixed string in the binary changed the value of another string that
6211 happened to be identical to the end of the original first string.
6213 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
6214 name", it returns that address as the IP address. On some operating
6215 systems (e.g. Solaris), it also passes back the IP address string as the
6216 "host name". However, on others (e.g. Linux), it passes back an empty
6217 string. Exim wasn't checking for this, and was changing the host name to an
6218 empty string, assuming it had been canonicalized.
6220 5. Although rare, it is permitted to have more than one PTR record for a given
6221 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
6222 all the names associated with an address, because they do in Solaris.
6223 However, it seems that they do not in Linux for data that comes from the
6224 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
6225 I found this out when I moved to a new Linux workstation and tried to run
6226 the Exim test suite.
6228 To get round this problem I have changed the code so that it now does its
6229 own call to the DNS to look up PTR records when searching for a host name.
6230 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
6231 addresses that are only in /etc/hosts are still found.
6233 This behaviour is, however, controlled by an option called host_lookup_
6234 order, which defaults to "bydns:byaddr". If people want to use the other
6235 order, or indeed, just use one or the other means of lookup, they can
6236 specify it in this variable.
6238 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
6239 some operating systems, this comes back from gethostbyaddr() as an empty
6240 string, and this is what Exim used to test for. However, it seems that in
6241 other systems, "." is yielded. Exim now tests for this case too.
6243 7. The values of check_spool_space and check_log_space are now held internally
6244 as a number of kilobytes instead of an absolute number of bytes. If a
6245 numbers is specified without 'K' or 'M', it is rounded up to the nearest
6246 kilobyte. This means that much larger values can be stored.
6248 8. Exim monitor: an attempt to get the action menu when not actually pointing
6249 at a message produces an empty menu entitled "No message selected". This
6250 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
6251 no entries in it ("Shell widget menu has zero width and/or height"). So I
6252 have added a single, blank menu entry in this case.
6254 9. Added ${quote_local_part.
6256 10. MIME decoding is now applied to the contents of Subject: header lines when
6259 11. Now that a reference to $sender_host_address automatically causes a reverse
6260 lookup to occur if necessary (4.13/18), there is no need to arrange for a
6261 host lookup before query-style lookups in lists that might use this
6262 variable. This has therefore been abolished, and the "net-" prefix is no
6263 longer necessary for query-style lookups.
6265 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
6266 have been a typo for LFLAGS, so it has been changed.
6268 13. The install script calls Exim with "-C /dev/null" in order to find the
6269 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
6270 to be output. However, since Exim outputs its version number before the
6271 error, it didn't break the script. It just looked ugly. I fixed this by
6272 always allowing "-C /dev/null" if the caller is root.
6274 14. Ignore overlarge ACL variable number when reading spool file - insurance
6275 against a later release with more variables having written the file.
6277 15. The standard form for an IPv6 address literal was being rejected by EHLO.
6278 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
6279 this, as well as the form without the "IPv6" on the front.
6281 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
6282 OS/Makefile-Darwin file.
6284 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
6285 only to LDAP libraries that do not have LDAP_OPT_DEREF.
6287 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
6288 variable. However, some OS use ints for inodes. I've added cast to long int
6289 to get rid of the compiler warning.
6291 19. I had forgotten to lock out "/../" in configuration file names when
6292 ALT_CONFIG_PREFIX was set.
6294 20. Routers used for verification do not need to specify transports. However,
6295 if such a router generated a host list, and callout was configured, Exim
6296 crashed, because it could not find a port number from the (non-existent)
6297 transport. It now assumes port 25 in this circumstance.
6299 21. Added the -t option to exigrep.
6301 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
6302 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
6303 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
6304 them is set, LOOKUP_LSEARCH is forced.
6306 23. "exim -bV" now outputs a list of lookups that are included in the binary.
6308 24. Added sender and host information to the "rejected by local_scan()" log
6309 line; previously there was no indication of these.
6311 25. Added .include_if_exists.
6313 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
6314 Linux. It turns out that not all file systems support this. Apparently some
6315 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
6316 do it.) To cope with this, the error EINVAL, which means that sync-ing is
6317 not supported on the file descriptor, is now ignored when Exim is trying to
6318 sync a directory. This applies only to Linux.
6320 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
6322 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
6323 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
6326 29. Added support for maildirsize files from supplied patch (modified a bit).
6328 30. The use of :fail: followed by an empty string could lead Exim to respond to
6329 sender verification failures with (e.g.):
6331 550 Verification failed for <xxx>
6332 550 Sender verify failed
6334 where the first response line was missing the '-' that indicates it is not
6335 the final line of the response.
6337 31. The loop for finding the name of the user that called Exim had a hardwired
6338 limit of 10; it now uses the value of finduser_retries, which is used for
6339 all other user lookups.
6341 32. Added $received_count variable, available in data and not_smtp ACLs, and at
6344 33. Exim was neglecting to zero errno before one call of strtol() when
6345 expanding a string and expecting an integer value. On some systems this
6346 resulted in spurious "integer overflow" errors. Also, it was casting the
6347 result into an int without checking.
6349 34. Testing for a connection timeout using "timeout_connect" in the retry rules
6350 did not work. The code looks as if it has *never* worked, though it appears
6351 to have been documented since at least release 1.62. I have made it work.
6353 35. The "timeout_DNS" error in retry rules, also documented since at least
6354 1.62, also never worked. As it isn't clear exactly what this means, and
6355 clearly it isn't a major issue, I have abolished the feature by treating it
6356 as "timeout", and writing a warning to the main and panic logs.
6358 36. The display of retry rules for -brt wasn't always showing the error code
6361 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
6362 timeout_connect_A, timeout_connect_MX.
6364 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
6365 to the empty sender.
6367 39. The daemon was not analysing the content of -oX till after it had closed
6368 stderr and disconnected from the controlling terminal. This meant that any
6369 syntax errors were only noted on the panic log, and the return code from
6370 the command was 0. By re-arranging the code a little, I've made the
6371 decoding happen first, so such errors now appear on stderr, and the return
6372 code is 1. However, the actual setting up of the sockets still happens in
6373 the disconnected process, so errors there are still only recorded on the
6376 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
6377 connections (as happens on some IP stacks) was logged at start up time as
6378 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
6379 it from "IPv6 and IPv4", which means that two separate sockets are being
6382 41. The debug output for gethostbyname2() or getipnodebyname() failures now
6383 says whether AF_INET or AF_INET6 was passed as an argument.
6385 42. Exiwhat output was messed up when time zones were included in log
6388 43. Exiwhat now gives more information about the daemon's listening ports,
6389 and whether -tls-on-connect was used.
6391 44. The "port" option of the smtp transport is now expanded.
6393 45. A "message" modifier in a "warn" statement in a non-message ACL was being
6394 silently ignored. Now an error message is written to the main and panic
6397 46. There's a new ACL modifier called "logwrite" which writes to a log file
6398 as soon as it is encountered.
6400 47. Added $local_user_uid and $local_user_gid at routing time.
6402 48. Exim crashed when trying to verify a sender address that was being
6405 49. Exim was recognizing only a space character after ".include". It now also
6406 recognizes a tab character.
6408 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
6409 extracting the relevant information from the specification. The man page no
6410 longer contains scrambled data for the -d option, and I've added a section
6411 at the front about calling Exim under different names.
6413 51. Added "extra_headers" argument to the "mail" command in filter files.
6415 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
6418 53. Installed eximstats 1.29.
6420 54. Added transport_filter_timeout as a generic transport option.
6422 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
6423 Cc: header lines. This was required by RFC 822, but it not required by RFC
6426 56. Exim used to add From:, Date:, and Message-Id: header lines to any
6427 incoming messages that did not have them. Now it does so only if the
6428 message originates locally, that is, if there is no associated remote host
6429 address. When Resent- header lines are present, this applies to the Resent-
6430 lines rather than the non-Resent- lines.
6432 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
6433 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
6435 58. Messages for configuration errors now include the name of the main
6436 configuration file - useful now that there may be more than one file in a
6437 list (.included file names were always shown).
6439 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
6440 for those rare installations that do not start the daemon as root or run it
6441 setuid root. I've cut out the call to initgroups() if the daemon is not
6444 60. The Exim user and group can now be bound into the binary as text strings
6445 that are looked up at the start of Exim's processing.
6447 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
6449 62. Added $mailstore_basename variable.
6451 63. Installed patch to sieve.c from Michael Haardt.
6453 64. When Exim failed to open the panic log after failing to open the main log,
6454 the original message it was trying to log was written to stderr and debug
6455 output, but if they were not available (the usual case in production), it
6456 was lost. Now it is written to syslog before the two lines that record the
6457 failures to open the logs.
6459 65. Users' Exim filters run in subprocesses under the user's uid. It is
6460 possible for a "deliver" command or an alias in a "personal" command to
6461 provoke an address rewrite. If logging of address rewriting is configured,
6462 this fails because the process is not running as root or exim. There may be
6463 a better way of dealing with this, but for the moment (because 4.30 needs
6464 to be released), I have disabled address rewrite logging when running a
6465 filter in a non-root, non-exim process.
6471 1. The buildconfig auxiliary program wasn't quoting the value set for
6472 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
6473 not defined. This bug was masked in 4.22 by the effect that was fixed in
6476 2. Some messages that were rejected after a message id was allocated were
6477 shown as "incomplete" by exigrep. It no longer does this for messages that
6478 are rejected by local_scan() or the DATA or non-SMTP ACLs.
6480 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
6481 have allow_domain_literals set, the ID did not get logged in the <= line.
6482 Domain literals are now always recognized in Message-ID: header lines.
6484 4. The first argument for a ${extract expansion item is the key name or field
6485 number. Leading and trailing spaces in this item were not being ignored,
6486 causing some misleading effects.
6488 5. When deliver_drop_privilege was set, single queue runner processes started
6489 manually (i.e. by the command "exim -q") or by the daemon (which uses the
6490 same command in the process it spins off) were not dropping privilege.
6492 6. When the daemon running as "exim" started a queue runner, it always
6493 re-executed Exim in the spun-off process. This is a waste of effort when
6494 deliver_drop_privilege is set. The new process now just calls the
6495 queue-runner function directly.
6501 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
6504 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
6505 ignored. Though the use of -oP was forcing the writing of a pid file, it
6506 was always written to the default place.
6508 3. If the message "no IP address found for host xxxx" is generated during
6509 incoming verification, it is now followed by identification of the incoming
6510 connection (so you can more easily find what provoked it).
6512 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
6514 5. Added some features to "harden" Exim a bit more against certain attacks:
6516 (a) There is now a build-time option called FIXED_NEVER_USERS that can
6517 be put in Local/Makefile. This is like the never_users runtime option,
6518 but it cannot be overridden. The default setting is "root".
6520 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
6521 prefix string with which any file named in a -C command line option
6524 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
6525 is retained for -C and -D only if the caller of Exim is root. Without
6526 it, the exim user may also use -C and -D and retain privilege.
6528 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
6529 command line option is disabled.
6531 6. Macro names set by the -D option must start with an upper case letter, just
6532 like macro names defined in the configuration file.
6534 7. Added "dereference=" facility to LDAP.
6536 8. Two instances of the typo "uknown" in the source files are fixed.
6538 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
6539 the Configure-Makefile script screwed up while processing it.
6541 10. Incorporated PCRE 4.4.
6543 11. The SMTP synchronization check was not operating right at the start of an
6544 SMTP session. For example, it could not catch a HELO sent before the client
6545 waited for the greeting. There is now a check for outstanding input at the
6546 point when the greeting is written. Because of the duplex, asynchronous
6547 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
6548 way, but not yet received, when the check is performed.
6550 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
6551 on TCP/IP sockets, because this apparently causes some broken clients to
6554 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
6555 unchanged) from the Cygwin maintainer.
6557 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
6558 was supported instead of testing for mbx. Effectively a typo.
6560 15. The spa authenticator server code was not checking that the input it
6561 received was valid base64.
6563 16. The debug output line for the "set" modifier in ACLs was not showing the
6564 name of the variable that was being set.
6566 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
6568 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
6569 lookup on its own. Something else was needed to trigger the lookup. For
6570 example, a match in host_lookup or the need for a host name in a host list.
6571 Now, if $sender_host_name is referenced and the host name has not yet been
6572 looked up, a lookup is performed. If the lookup fails, the variable remains
6573 empty, and $host_lookup_failed is set to "1".
6575 19. Added "eqi" as a case-independent comparison operator.
6577 20. The saslauthd authentication condition could segfault if neither service
6578 nor realm was specified.
6580 21. If an overflowing value such as "2048M" was set for message_size_limit, the
6581 error message that was logged was misleading, and incoming SMTP
6582 connections were dropped. The message is now more accurate, and temporary
6583 errors are given to SMTP connections.
6585 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
6586 (except RSET) with a 421 error, until QUIT is received. However, it was
6587 failing to send a response to QUIT.
6589 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
6590 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
6591 ACL is now run after the helo_try_verify_hosts code.
6593 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
6594 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
6595 case-independent, so other case variants are also recognized). Apparently
6596 some systems use these upper case variants.
6598 25. If more than two messages were waiting for the same host, and a transport
6599 filter was specified for the transport, Exim sent two messages over the
6600 same TCP/IP connection, and then failed with "socket operation on non-
6601 socket" when it tried to send the third.
6603 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
6605 27. The extern definition of crypt16() in expand.c was not being excluded when
6606 the OS had its own crypt16() function.
6608 28. Added bounce_return_body as a new option, and bounce_return_size_limit
6609 as a preferred synonym for return_size_limit, both as an option and as an
6612 29. Added LIBS=-liconv to OS/Makefile-OSF1.
6614 30. Changed the default configuration ACL to relax the local part checking rule
6615 for addresses that are not in any local domains. For these addresses,
6616 slashes and pipe symbols are allowed within local parts, but the sequence
6617 /../ is explicitly forbidden.
6619 31. SPA server authentication was not clearing the challenge buffer before
6622 32. log_message in a "warn" ACL statement was writing to the reject log as
6623 well as to the main log, which contradicts the documentation and doesn't
6624 seem right (because no rejection is happening). So I have stopped it.
6626 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
6627 However, I am unable to do any testing of this.
6629 34. Fixed an infelicity in the appendfile transport. When checking directories
6630 for a mailbox, to see if any needed to be created, it was accidentally
6631 using path names with one or more superfluous leading slashes; tracing
6632 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
6634 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
6635 contents are added to the log line that is written for every discarded
6636 recipient. (Previously a log_message setting was ignored.)
6638 36. The ${quote: operator now quotes the string if it is empty.
6640 37. The install script runs exim in order to find its version number. If for
6641 some reason other than non-existence or emptiness, which it checks, it
6642 could not run './exim', it was installing it with an empty version number,
6643 i.e. as "exim-". This error state is now caught, and the installation is
6646 38. An argument was missing from the function that creates an error message
6647 when Exim fails to connect to the socket for saslauthd authentication.
6648 This could cause Exim to crash, or give a corrupted message.
6650 39. Added isip, isip4, and isip6 to ${if conditions.
6652 40. The ACL variables $acl_xx are now saved with the message, and can be
6653 accessed later in routers, transports, and filters.
6655 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
6656 strings in the file are not string-expanded.
6658 42. If a MAIL command specified a SIZE value that was too large to fit into an
6659 int variable, the check against message_size_limit failed. Such values are
6660 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
6661 day this will have to be increased, but I don't think I want to be around
6662 when emails are that large.
6669 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
6670 iconv() is not standard in FreeBSD.
6672 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
6673 IPv6 enabled. The observed symptom was a segmentation fault on return from
6674 the function os_common_find_running_interfaces() in src/os.c.
6676 3. In the check_special_case() function in daemon.c I had used "errno" as an
6677 argument name, which causes warnings on some systems. This was basically a
6678 typo, since it was named "eno" in the comments!
6680 4. The code that waits for the clock to tick (at a resolution of some fraction
6681 of a second) so as to ensure message-id uniqueness was always waiting for
6682 at least one whole tick, when it could have waited for less. [This is
6683 almost certainly not relevant at current processor speeds, where it is
6684 unlikely to ever wait at all. But we try to future-proof.]
6686 5. The function that sleeps for a time interval that includes fractions of a
6687 second contained a race. It did not block SIGALRM between setting the
6688 timer, and suspending (a couple of lines later). If the interval was short
6689 and the sigsuspend() was delayed until after it had expired, the suspension
6690 never ended. On busy systems this could lead to processes getting stuck for
6693 6. Some uncommon configurations may cause a lookup to happen in a queue runner
6694 process, before it forks any delivery processes. The open lookup caching
6695 mechanism meant that the open file or database connection was passed into
6696 the delivery process. The problem was that delivery processes always tidy
6697 up cached lookup data. This could cause a problem for the next delivery
6698 process started by the queue runner, because the external queue runner
6699 process does not know about the closure. So the next delivery process
6700 still has data in the lookup cache. In the case of a file lookup, there was
6701 no problem because closing a file descriptor in a subprocess doesn't affect
6702 the parent. However, if the lookup was caching a connection to a database,
6703 the connection was closed, and the second delivery process was likely to
6704 see errors such as "PGSQL: query failed: server closed the connection
6705 unexpectedly". The problem has been fixed by closing all cached lookups
6706 in a queue runner before running a delivery process.
6708 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
6709 seem to have the "const" qualifier which it has on other OS. I've
6712 8. Change 4.21/2 was too strict. It is only if there are two authenticators
6713 *of the same type* (client or server) with the same public name that an
6714 error should be diagnosed.
6716 9. When Exim looked up a host name for an IP address, but failed to find the
6717 original IP address when looking up the host name (a safety check), it
6718 output the message "<ip address> does not match any IP for NULL", which was
6719 confusing, to say the least. The bug was that the host name should have
6720 appeared instead of "NULL".
6722 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
6723 user that is built into the binary, and the -C or -D options is used, root
6724 privilege is dropped before the configuration file is read. In addition,
6725 logging is switched to stderr instead of the normal log files. If the
6726 configuration then re-defines the Exim user, the unprivileged environment
6727 is probably not what is expected, so Exim logs a panic warning message (but
6730 However, if deliver_drop_privilege is set, the unprivileged state may well
6731 be exactly what is intended, so the warning has been cut out in that case,
6732 and Exim is allowed to try to write to its normal log files.
6738 1. smtp_return_error_details was not giving details for temporary sender
6739 or receiver verification errors.
6741 2. Diagnose a configuration error if two authenticators have the same public
6744 3. Exim used not to create the message log file for a message until the first
6745 delivery attempt. This could be confusing when incoming messages were held
6746 for policy or load reasons. The message log file is now created at the time
6747 the message is received, and an initial "Received" line is written to it.
6749 4. The automatically generated man page for command line options had a minor
6750 bug that caused no ill effects; however, a more serious problem was that
6751 the procedure for building the man page automatically didn't always
6752 operate. Consequently, release 4.20 contains an out-of-date version. This
6753 shouldn't happen again.
6755 5. When building Exim with embedded Perl support, the script that builds the
6756 Makefile was calling 'perl' to find its compile-time parameters, ignoring
6757 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
6759 6. The freeze_tell option was not being used for messages that were frozen on
6760 arrival, either by an ACL or by local_scan().
6762 7. Added the smtp_incomplete_transaction log selector.
6764 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
6765 was accepting AUTH without a new EHLO.
6767 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
6768 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
6771 10. Logging of TCP/IP connections (when configured) now happens in the main
6772 daemon process instead of the child process, so that the TCP/IP connection
6773 count is more accurate (but it can never be perfect).
6775 11. The use of "drop" in a nested ACL was not being handled correctly in the
6776 outer ACL. Now, if condition failure induced by the nested "drop" causes
6777 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
6778 or "require"), the connection is dropped.
6780 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
6781 that yield "discard" can now be used with an "accept" or a "discard" verb,
6782 but an error is generated for any others (because I can't see a useful way
6783 to define what should happen).
6785 13. When an ACL is read dynamically from a file (or anywhere else), the lines
6786 are now processed in the same way as lines in the Exim configuration file.
6787 In particular, continuation lines are supported.
6789 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
6791 15. Added -ti meaning -t -i.
6793 16. Check for letters, digits, hyphens, and dots in the names of dnslist
6794 domains, and warn by logging if others are found.
6796 17. At least on BSD, alignment is not guaranteed for the array of ifreq's
6797 returned from GIFCONF when Exim is trying to find the list of interfaces on
6798 a host. The code in os.c has been modified to copy each ifreq to an aligned
6799 structure in all cases.
6801 Also, in some cases, the returned ifreq's were being copied to a 'struct
6802 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
6803 means the last couple of bytes of an IPv6 address could be chopped if the
6804 ifreq contained only a normal sockaddr (14 bytes storage).
6806 18. Named domain lists were not supported in the hosts_treat_as_local option.
6807 An entry such as +xxxx was not recognized, and was treated as a literal
6810 19. Ensure that header lines added by a DATA ACL are included in the reject log
6811 if the ACL subsequently rejects the message.
6813 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
6814 MD5 (which is deprecated).
6816 21. When testing a filter file using -bf, Exim was writing a message when it
6817 took the sender from a "From " line in the message, but it was not doing so
6818 when it took $return_path from a Return-Path: header line. It now does.
6820 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
6821 with a valid header line field name (a series of printing characters
6822 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
6824 23. Changed "disc" in the source to "disk" to conform to the documentation and
6825 the book and for uniformity.
6827 24. Ignore Sendmail's -Ooption=value command line item.
6829 25. When execve() failed while trying to run a command in a pipe transport,
6830 Exim was returning EX_UNAVAILABLE (69) from the subprocess. However, this
6831 could be confused with a return value of 69 from the command itself. This
6832 has been changed to 127, the value the shell returns if it is asked to run
6833 a non-existent command. The wording for the related log line suggests a
6834 non-existent command as the problem.
6836 26. If received_header_text expands to an empty string, do not add a Received:
6837 header line to the message. (Well, it adds a token one on the spool, but
6838 marks it "old" so that it doesn't get used or transmitted.)
6840 27. Installed eximstats 1.28 (addition of -nt option).
6842 28. There was no check for failure on the call to getsockname() in the daemon
6843 code. This can fail if there is a shortage of resources on the system, with
6844 ENOMEM, for example. A temporary error is now given on failure.
6846 29. Contrary to the C standard, it seems that in some environments, the
6847 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
6848 program. Exim now does this explicitly; it affects the formatting of
6849 timestamps using strftime().
6851 30. If exiqsumm was given junk data, it threw up some uninitialized variable
6852 complaints. I've now initialized all the variables, to avoid this.
6854 32. Header lines added by a system filter were not being "seen" during
6855 transport-time rewrites.
6857 33. The info_callback() function passed to OpenSSL is set up with type void
6858 (*)(SSL *, int, int), as described somewhere. However, when calling the
6859 function (actually a macro) that sets it up, the type void(*)() is
6860 expected. I've put in a cast to prevent warnings from picky compilers.
6862 34. If a DNS black list lookup found a CNAME record, but there were no A
6863 records associated with the domain it pointed at, Exim crashed.
6865 35. If a DNS black list lookup returned more than one A record, Exim ignored
6866 all but the first. It now scans all returned addresses if a particular IP
6867 value is being sought. In this situation, the contents of the
6868 $dnslist_value variable are a list of all the addresses, separated by a
6871 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
6872 to accept a host name and all its aliases if the forward lookup for any of
6873 them yielded the IP address of the incoming connection. Now it accepts only
6874 those names whose forward lookup yields the correct IP address. Any other
6875 names are discarded. This closes a loophole whereby a rogue DNS
6876 administrator could create reverse DNS records to break through a
6877 wildcarded host restriction in an ACL.
6879 37. If a user filter or a system filter that ran in a subprocess used any of
6880 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
6881 the wrong values were passed to the pipe command ($thisaddress had the
6882 value of $0, $0 had the value of $1, etc). This bug was introduced by
6883 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
6885 38. Improved the line breaking for long SMTP error messages from ACLs.
6886 Previously, if there was no break point between 40 and 75 characters, Exim
6887 left the rest of the message alone. Two changes have been made: (a) I've
6888 reduced the minimum length to 35 characters; (b) if it can't find a break
6889 point between 35 and 75 characters, it looks ahead and uses the first one
6890 that it finds. This may give the occasional overlong line, but at least the
6891 remaining text gets split now.
6893 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
6894 file descriptors might be low, and that setting 1000 would always raise it.
6895 It turns out that in some environments, the limit is already over 1000 and
6896 that lowering it causes trouble. So now Exim takes care not to decrease it.
6898 40. When delivering a message, the value of $return_path is set to $sender_
6899 address at the start of routing (routers may change the value). By an
6900 oversight, this default was not being set up when an address was tested by
6901 -bt or -bv, which affected the outcome if any router or filter referred to
6904 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
6905 log only when "message" or "log_message" are set. However, if one of the
6906 conditions was an address verification, or a call to a nested ACL, the
6907 messages generated by the underlying test were being passed through. This
6908 no longer happens. The underlying message is available in $acl_verify_
6909 message for both "message" and "log_message" expansions, so it can be
6910 passed through if needed.
6912 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
6913 new expansion $bh_ to give the encoded byte string without charset
6914 translation. Translation happens only if iconv() is available; HAVE_ICONV
6915 indicates this at build time. HEADERS_CHARSET gives the charset to
6916 translate to; headers_charset can change it in the configuration, and
6917 "headers charset" can change it in an individual filter file.
6919 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
6920 that creates RFC 2047 encoded "words" labels them as that charset instead
6921 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
6922 expansion operator; (ii) when Exim creates a From: line for a local
6923 message; (iii) when a header line is rewritten to include a "phrase" part.
6925 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
6926 buggy, causing it to skip the first lines of messages whose message ID
6927 ended in 'D'. This would not have bitten before Exim release 4.14, because
6928 message IDs were unlikely to end in 'D' before then. The effect was to have
6929 incorrect size information for certain domains.
6931 45. #include "config.h" was missing at the start of the crypt16.c module. This
6932 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
6935 46. If there was a timeout during a "random" callout check, Exim treated it as
6936 a failure of the random address, and carried on sending RSET and the real
6937 address. If the delay was just some slowness somewhere, the response to the
6938 original RCPT would be taken as a response to RSET and so on, causing
6939 mayhem of various kinds.
6941 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
6942 when I implemented it. It didn't allow for the fact that some option values
6943 may legitimately be negative (e.g. size_addition), and it didn't even do
6944 the right test for positive values.
6946 48. Domain names in DNS records are case-independent. Exim always looks them up
6947 in lower case. Some resolvers return domain names in exactly the case they
6948 appear in the zone file, that is, they may contain uppercase letters. Not
6949 all resolvers do this - some return always lower case. Exim was treating a
6950 change of case by a resolver as a change of domain, similar to a widening
6951 of a domain abbreviation. This triggered its re-routing code and so it was
6952 trying to route what was effectively the same domain again. This normally
6953 caused routing to fail (because the router wouldn't handle the domain
6954 twice). Now Exim checks for this case specially, and just changes the
6955 casing of the domain that it ultimately uses when it transmits the message
6958 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
6961 50. If a filter generated a file delivery with a non-absolute name (possible if
6962 no home directory exists for the router), the forbid_file option was not
6965 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
6966 the existing equality matching.
6968 52. Exim was using ints instead of ino_t variables in some places where it was
6969 dealing with inode numbers.
6971 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
6972 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
6973 called TMPDIR, and if it finds it is different, it changes its value.
6975 54. The smtp_printf() function is now made available to local_scan() so
6976 additional output lines can be written before returning. There is also an
6977 smtp_fflush() function to enable the detection of a dropped connection.
6978 The variables smtp_input and smtp_batched_input are exported to
6981 55. Changed the default runtime configuration: the message "Unknown user"
6982 has been removed from the ACL, and instead placed on the localuser router,
6983 using the cannot_route_message feature. This means that any verification
6984 failures that generate their own messages won't get overridden. Similarly,
6985 the "Unrouteable address" message that was in the ACL for unverifiable
6986 relay addresses has also been removed.
6988 56. Added hosts_avoid_esmtp to the smtp transport.
6990 57. The exicyclog script was not checking for the esoteric option
6991 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
6992 will work only if exicyclog is run under the appropriate euid.
6994 58. Following a discussion on the list, the rules by which Exim recognises line
6995 endings on incoming messages have been changed. The -dropcr and drop_cr
6996 options are now no-ops, retained only for backwards compatibility. The
6997 following line terminators are recognized: LF CRLF CR. However, special
6998 processing applies to CR:
7000 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
7001 nor a local message in the state where . is a terminator.
7003 (ii) If a bare CR is encountered in a header line, an extra space is added
7004 after the line terminator so as not to end the header. The reasoning
7005 behind this is that bare CRs in header lines are most likely either
7006 to be mistakes, or people trying to play silly games.
7008 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
7009 was being incorrectly given as 18 bytes larger than it should have been.
7010 This is a VOB (very old bug).
7012 60. This may never have affected anything current, but just in case it has:
7013 When the local host is found other than at the start of a list of hosts,
7014 the local host, those with the same MX, and any that follow, are discarded.
7015 When the list in question was part of a longer list of hosts, the following
7016 hosts (not currently being processed) were also being discarded. This no
7017 longer happens. I'm not sure if this situation could ever has previously
7020 61. Added the "/MX" feature to lists of hosts in the manualroute and query
7023 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
7024 header. This is something that is recommended in a new Internet Draft, and
7025 is something that is documented as being done by Sendmail. There are two
7026 possible values. For messages generated by the autoreply transport, Exim
7029 Auto-Submitted: auto-replied
7031 whereas for all other generated messages (e.g. bounces) it adds
7033 Auto-Submitted: auto-generated
7035 63. The "personal" condition in filters now includes a test for the
7036 Auto-Submitted: header. If it contains the string "auto-" the message it
7037 not considered personal.
7039 64. Added rcpt_include_affixes as a generic transport option.
7041 65. Added queue_only_override (default true).
7043 66. Added the syslog_duplication option.
7045 67. If what should have been the first header line of a message consisted of
7046 a space followed by a colon, Exim was mis-interpreting it as a header line.
7047 It isn't of course - it is syntactically invalid and should therefore be
7048 treated as the start of the message body. The misbehaviour could have
7049 caused a number of strange effects, including loss of data in subsequent
7050 header lines, and spool format errors.
7052 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
7053 client host had authenticated. This control can now be exercised by an ACL
7054 for more flexibility.
7056 69. By default, callouts do not happen when testing with -bh. There is now a
7057 variant, -bhc, which does actually run the callout code, including
7058 consulting and updating the callout cache.
7060 70. Added support for saslauthd authentication, courtesy of Alexander
7063 71. If statvfs() failed on the spool or log directories while checking their
7064 size for availability, Exim confusingly gave the error "space shortage".
7065 Furthermore, in debugging mode it crashed with a floating point exception.
7066 These checks are done if check_{spool,log}_{space,inodes} are set, and when
7067 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
7068 serious problem, Exim now writes to the main and panic logs when this
7069 happens, with details of the failure. It then refuses to accept the
7070 incoming message, giving the message "spool directory problem" or "log
7071 directory problem" with a 421 code for SMTP messages.
7073 72. When Exim is about to re-exec itself, it ensures that the file descriptors
7074 0, 1, and 2 exist, because some OS complain for execs without them (see
7075 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
7076 descriptors. However, the code omitted to check that the open succeeded,
7077 causing mysterious errors if for some reason the permissions on /dev/null
7078 got screwed. Now Exim writes a message to the main and panic logs, and
7079 bombs out if it can't open /dev/null.
7081 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
7082 interact so that it is all more flexible. It is supposed to remain
7083 backwards compatible. Also added extra_local_interfaces.
7085 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
7086 to bomb out with an assertion failure - to the client this appears as a
7087 connection drop. This problem occurs in the part of the code that was taken
7088 from the Samba project. Fortunately, the assertion is in a very simple
7089 function, so I have fixed this by reproducing the function inline in the
7090 one place where it is called, and arranging for authentication to fail
7091 instead of killing the process with assert().
7093 75. The SPA client code was not working when the server requested OEM rather
7094 than Unicode encoding.
7096 76. Added code to make require_files with a specific uid setting more usable in
7097 the case where statting the file as root fails - usually a non-root-mounted
7098 NFS file system. When this happens and the failure is EACCES, Exim now
7099 forks a subprocess and does the per-uid checking as the relevant uid.
7101 77. Added process_log_path.
7103 78. If log_file_path was not explicitly set, a setting of check_log_space or
7104 check_log_inodes was ignored.
7106 79. If a space check for the spool or log partitions fails, the incident is now
7107 logged. Of course, in the latter case the data may get lost...
7109 80. Added the %p formatting code to string_format() so that it can be used to
7110 print addresses in debug_print(). Adjusted all the address printing in the
7111 debugging in store.c to use %p rather than %d.
7113 81. There was a concern that a line of code in smtp_in.c could overflow a
7114 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
7115 initially expressed, the concern was not well-founded, because trailing
7116 spaces are removed early. However, if the trailing spaces were followed by
7117 a NULL, they did not get removed, so the overflow was possible. Two fixes
7120 (a) I re-wrote the offending code in a cleaner fashion.
7121 (b) If an incoming SMTP command contains a NULL character, it is rejected
7124 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
7125 runs initgroups(), so that if the Exim user is in any additional groups,
7126 they will be used during message reception.
7132 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / users / jgh / exim.git / blob