1 Change log file for Exim from version 4.21
2 ------------------------------------------
3 This document describes *changes* to previous versions, that might
4 affect Exim's operation, with an unchanged configuration file. For new
5 options, and new features, see the NewStuff file next to this ChangeLog.
8 Since Exim version 4.90
9 -----------------------
11 JH/01 Replace the store_release() internal interface with store_newblock(),
12 which internalises the check required to safely use the old one, plus
13 the allocate and data copy operations duplicated in both (!) of the
16 JH/02 Disallow '/' characters in queue names specified for the "queue=" ACL
17 modifier. This matches the restriction on the commandline.
19 JH/03 Fix pgsql lookup for multiple result-tuples with a single column.
20 Previously only the last row was returned.
22 JH/04 Bug 2217: Tighten up the parsing of DKIM signature headers. Previously
23 we assumed that tags in the header were well-formed, and parsed the
24 element content after inspecting only the first char of the tag.
25 Assumptions at that stage could crash the receive process on malformed
28 JH/05 Bug 2215: Fix crash associated with dnsdb lookup done from DKIM ACL.
29 While running the DKIM ACL we operate on the Permanent memory pool so that
30 variables created with "set" persist to the DATA ACL. Also (at any time)
31 DNS lookups that fail create cache records using the Permanent pool. But
32 expansions release any allocations made on the current pool - so a dnsdb
33 lookup expansion done in the DKIM ACL releases the memory used for the
34 DNS negative-cache, and bad things result. Solution is to switch to the
35 Main pool for expansions.
36 While we're in that code, add checks on the DNS cache during store_reset,
37 active in the testsuite.
38 Problem spotted, and debugging aided, by Wolfgang Breyha.
40 JH/06 Fix issue with continued-connections when the DNS shifts unreliably.
41 When none of the hosts presented to a transport match an already-open
42 connection, close it and proceed with the list. Previously we would
43 queue the message. Spotted by Lena with Yahoo, probably involving
46 JH/07 Bug 2214: Fix SMTP responses resulting from non-accept result of MIME ACL.
47 Previously a spurious "250 OK id=" response was appended to the proper
50 JH/08 The "support for" informational output now, which built with Content
51 Scanning support, has a line for the malware scanner interfaces compiled
52 in. Interface can be individually included or not at build time.
54 JH/09 The "aveserver", "kavdaemon" and "mksd" interfaces are now not included
55 by the template makefile "src/EDITME". The "STREAM" support for an older
56 ClamAV interface method is removed.
58 JH/10 Bug 2223: Fix mysql lookup returns for the no-data case (when the number of
59 rows affected is given instead).
61 JH/11 The runtime Berkeley DB library version is now additionally output by
62 "exim -d -bV". Previously only the compile-time version was shown.
64 JH/12 Bug 2230: Fix cutthrough routing for nonfirst messages in an initiating
65 SMTP connection. Previously, when one had more receipients than the
66 first, an abortive onward connection was made. Move to full support for
67 multiple onward connections in sequence, handling cutthrough connection
68 for all multi-message initiating connections.
70 JH/13 Bug 2229: Fix cutthrough routing for nonstandard port numbers defined by
71 routers. Previously, a multi-recipient message would fail to match the
72 onward-connection opened for the first recipient, and cause its closure.
74 JH/14 Bug 2174: A timeout on connect for a callout was also erroneously seen as
75 a timeout on read on a GnuTLS initiating connection, resulting in the
76 initiating connection being dropped. This mattered most when the callout
77 was marked defer_ok. Fix to keep the two timeout-detection methods
80 JH/15 Relax results from ACL control request to enable cutthrough, in
81 unsupported situations, from error to silently (except under debug)
82 ignoring. This covers use with PRDR, frozen messages, queue-only and
85 HS/01 Fix Buffer overflow in base64d() (CVE-2018-6789)
87 JH/16 Fix bug in DKIM verify: a buffer overflow could corrupt the malloc
88 metadata, resulting in a crash in free().
90 PP/01 Fix broken Heimdal GSSAPI authenticator integration.
91 Broken in f2ed27cf5, missing an equals sign for specified-initialisers.
92 Broken also in d185889f4, with init system revamp.
94 JH/17 Bug 2113: Fix conversation closedown with the Avast malware scanner.
95 Previously we abruptly closed the connection after reading a malware-
96 found indication; now we go on to read the "scan ok" response line,
99 JH/18 Convert macro handling to be tree-based, from the previous linear list.
100 With the number of builtin macros we now have this is worthwhile,
101 dropping the config-file read time (during which new macros are checked
102 and registered, and macros are expanded) from about 500 usec to about 180.
108 JH/01 Rework error string handling in TLS interface so that the caller in
109 more cases is responsible for logging. This permits library-sourced
110 string to be attached to addresses during delivery, and collapses
111 pairs of long lines into single ones.
113 PP/01 Allow PKG_CONFIG_PATH to be set in Local/Makefile and use it correctly
114 during configuration. Wildcards are allowed and expanded.
116 JH/02 Rework error string handling in DKIM to pass more info back to callers.
117 This permits better logging.
119 JH/03 Rework the transport continued-connection mechanism: when TLS is active,
120 do not close it down and have the child transport start it up again on
121 the passed-on TCP connection. Instead, proxy the child (and any
122 subsequent ones) for TLS via a unix-domain socket channel. Logging is
123 affected: the continued delivery log lines do not have any DNSSEC, TLS
124 Certificate or OCSP information. TLS cipher information is still logged.
126 JH/04 Shorten the log line for daemon startup by collapsing adjacent sets of
127 identical IP addresses on different listening ports. Will also affect
130 PP/02 Bug 2070: uClibc defines __GLIBC__ without providing glibc headers;
131 add noisy ifdef guards to special-case this sillyness.
132 Patch from Bernd Kuhls.
134 JH/05 Tighten up the checking in isip4 (et al): dotted-quad components larger
135 than 255 are no longer allowed.
137 JH/06 Default openssl_options to include +no_ticket, to reduce load on peers.
138 Disable the session-cache too, which might reduce our load. Since we
139 currrectly use a new context for every connection, both as server and
140 client, there is no benefit for these.
141 GnuTLS appears to not support tickets server-side by default (we don't
142 call gnutls_session_ticket_enable_server()) but client side is enabled
143 by default on recent versions (3.1.3 +) unless the PFS priority string
146 PP/03 Add $SOURCE_DATE_EPOCH support for reproducible builds, per spec at
147 <https://reproducible-builds.org/specs/source-date-epoch/>.
149 JH/07 Fix smtp transport use of limited max_rcpt under mua_wrapper. Previously
150 the check for any unsuccessful recipients did not notice the limit, and
151 erroneously found still-pending ones.
153 JH/08 Pipeline CHUNKING command and data together, on kernels that support
154 MSG_MORE. Only in-clear (not on TLS connections).
156 JH/09 Avoid using a temporary file during transport using dkim. Unless a
157 transport-filter is involved we can buffer the headers in memory for
158 creating the signature, and read the spool data file once for the
159 signature and again for transmission.
161 JH/10 Enable use of sendfile in Linux builds as default. It was disabled in
162 4.77 as the kernel support then wasn't solid, having issues in 64bit
163 mode. Now, it's been long enough. Add support for FreeBSD also.
165 JH/11 Bug 2104: Fix continued use of a transport connection with TLS. In the
166 case where the routing stage had gathered several addresses to send to
167 a host before calling the transport for the first, we previously failed
168 to close down TLS in the old transport process before passing the TCP
169 connection to the new process. The new one sent a STARTTLS command
170 which naturally failed, giving a failed delivery and bloating the retry
171 database. Investigation and fix prototype from Wolfgang Breyha.
173 JH/12 Fix check on SMTP command input synchronisation. Previously there were
174 false-negatives in the check that the sender had not preempted a response
175 or prompt from Exim (running as a server), due to that code's lack of
176 awareness of the SMTP input buffering.
178 PP/04 Add commandline_checks_require_admin option.
179 Exim drops privileges sanely, various checks such as -be aren't a
180 security problem, as long as you trust local users with access to their
181 own account. When invoked by services which pass untrusted data to
182 Exim, this might be an issue. Set this option in main configuration
183 AND make fixes to the calling application, such as using `--` to stop
186 JH/13 Do pipelining under TLS. Previously, although safe, no advantage was
187 taken. Now take care to pack both (client) MAIL,RCPT,DATA, and (server)
188 responses to those, into a single TLS record each way (this usually means
189 a single packet). As a side issue, smtp_enforce_sync now works on TLS
192 PP/05 OpenSSL/1.1: use DH_bits() for more accurate DH param sizes. This
193 affects you only if you're dancing at the edge of the param size limits.
194 If you are, and this message makes sense to you, then: raise the
195 configured limit or use OpenSSL 1.1. Nothing we can do for older
198 JH/14 For the "sock" variant of the malware scanner interface, accept an empty
199 cmdline element to get the documented default one. Previously it was
202 JH/15 Fix a crash in the smtp transport caused when two hosts in succession
203 are unsuable for non-message-specific reasons - eg. connection timeout,
204 banner-time rejection.
206 JH/16 Fix logging of delivery remote port, when specified by router, under
209 PP/06 Repair manualroute's ability to take options in any order, even if one
210 is the name of a transport.
213 HS/01 Cleanup, prevent repeated use of -p/-oMr (CVE-2017-1000369)
215 JH/17 Change the list-building routines interface to use the expanding-string
216 triplet model, for better allocation and copying behaviour.
218 JH/18 Prebuild the data-structure for "builtin" macros, for faster startup.
219 Previously it was constructed the first time a possibly-matching string
220 was met in the configuration file input during startup; now it is done
223 JH/19 Bug 2141: Use the full-complex API for Berkeley DB rather than the legacy-
224 compatible one, to avoid the (poorly documented) possibility of a config
225 file in the working directory redirecting the DB files, possibly correpting
226 some existing file. CVE-2017-10140 assigned for BDB.
228 JH/20 Bug 2147: Do not defer for a verify-with-callout-and-random which is not
229 cache-hot. Previously, although the result was properly cached, the
230 initial verify call returned a defer.
232 JH/21 Bug 2151: Avoid using SIZE on the MAIL for a callout verify, on any but
233 the main verify for receipient in uncached-mode.
235 JH/22 Retire historical build files to an "unsupported" subdir. These are
236 defined as "ones for which we have no current evidence of testing".
238 JH/23 DKIM: enforce the DNS pubkey record "h" permitted-hashes optional field,
239 if present. Previously it was ignored.
241 JH/24 Start using specified-initialisers in C structure init coding. This is
242 a C99 feature (it's 2017, so now considered safe).
244 JH/25 Use one-bit bitfields for flags in the "addr" data structure. Previously
245 if was a fixed-sized field and bitmask ops via macros; it is now more
248 PP/07 GitHub PR 56: Apply MariaDB build fix.
249 Patch provided by Jaroslav Škarvada.
251 PP/08 Bug 2161: Fix regression in sieve quoted-printable handling introduced
252 during Coverity cleanups [4.87 JH/47]
253 Diagnosis and fix provided by Michael Fischer v. Mollard.
255 JH/26 Fix DKIM bug: when the pseudoheader generated for signing was exactly
256 the right size to place the terminating semicolon on its own folded
257 line, the header hash was calculated to an incorrect value thanks to
258 the (relaxed) space the fold became.
260 HS/02 Fix Bug 2130: large writes from the transport subprocess where chunked
261 and confused the parent.
263 JH/27 Fix SOCKS bug: an unitialized pointer was deref'd by the transport process
264 which could crash as a result. This could lead to undeliverable messages.
266 JH/28 Logging: "next input sent too soon" now shows where input was truncated
269 JH/29 Fix queue_run_in_order to ignore the PID portion of the message ID. This
270 matters on fast-turnover and PID-randomising systems, which were getting
271 out-of-order delivery.
273 JH/30 Fix a logging bug on aarch64: an unsafe routine was previously used for
274 a possibly-overlapping copy. The symptom was that "Remote host closed
275 connection in response to HELO" was logged instead of the actual 4xx
278 JH/31 Fix CHUNKING code to properly flush the unwanted chunk after an error.
279 Previously only that bufferd was discarded, resulting in SYMTP command
282 JH/32 DKIM: when a message has multiple signatures matching an identity given
283 in dkim_verify_signers, run the dkim acl once for each. Previously only
284 one run was done. Bug 2189.
286 JH/33 Downgrade an unfound-list name (usually a typo in the config file) from
287 "panic the current process" to "deliberately defer". The panic log is
288 still written with the problem list name; the mail and reject logs now
289 get a temp-reject line for the message that was being handled, saying
290 something like "domains check lookup or other defer". The SMTP 451
291 message is still "Temporary local problem".
293 JH/34 Bug 2199: Fix a use-after-free while reading smtp input for header lines.
294 A crafted sequence of BDAT commands could result in in-use memory beeing
295 freed. CVE-2017-16943.
297 HS/03 Bug 2201: Fix checking for leading-dot on a line during headers reading
298 from SMTP input. Previously it was always done; now only done for DATA
299 and not BDAT commands. CVE-2017-16944.
301 JH/35 Bug 2201: Flush received data in BDAT mode after detecting an error fatal
302 to the message (such as an overlong header line). Previously this was
303 not done and we did not exit BDAT mode. Followon from the previous item
304 though a different problem.
310 JH/01 Bug 1922: Support IDNA2008. This has slightly different conversion rules
311 than -2003 did; needs libidn2 in addition to libidn.
313 JH/02 The path option on a pipe transport is now expanded before use.
315 PP/01 GitHub PR 50: Do not call ldap_start_tls_s on ldapi:// connections.
316 Patch provided by "Björn", documentation fix added too.
318 JH/03 Bug 2003: fix Proxy Protocol v2 handling: the address size field was
319 missing a wire-to-host endian conversion.
321 JH/04 Bug 2004: fix CHUNKING in non-PIPELINEING mode. Chunk data following
322 close after a BDAT command line could be taken as a following command,
323 giving a synch failure. Fix by only checking for synch immediately
324 before acknowledging the chunk.
326 PP/02 GitHub PR 52: many spelling fixes, which include fixing parsing of
327 no_require_dnssec option and creation of _HAVE_TRANSPORT_APPEND_MAILDIR
328 macro. Patches provided by Josh Soref.
330 JH/05 Have the EHLO response advertise VRFY, if there is a vrfy ACL defined.
331 Previously we did not; the RFC seems ambiguous and VRFY is not listed
332 by IANA as a service extension. However, John Klensin suggests that we
335 JH/06 Bug 2017: Fix DKIM verification in -bh test mode. The data feed into
336 the dkim code may be unix-mode line endings rather than smtp wire-format
337 CRLF, so prepend a CR to any bare LF.
339 JH/07 Rationalise the coding for callout smtp conversations and transport ones.
340 As a side-benfit, callouts can now use PIPELINING hence fewer round-trips.
342 JH/08 Bug 2016: Fix DKIM verification vs. CHUNKING. Any BDAT commands after
343 the first were themselves being wrongly included in the feed into dkim
344 processing; with most chunk sizes in use this resulted in an incorrect
345 body hash calculated value.
347 JH/09 Bug 2014: permit inclusion of a DKIM-Signature header in a received
348 DKIM signature block, for verification. Although advised against by
349 standards it is specifically not ruled illegal.
351 JH/10 Bug 2025: Fix reception of (quoted) local-parts with embedded spaces.
353 JH/11 Bug 2029: Fix crash in DKIM verification when a message signature block is
354 missing a body hash (the bh= tag).
356 JH/12 Bug 2018: Re-order Proxy Protocol startup versus TLS-on-connect startup.
357 It seems that HAProxy sends the Proxy Protocol information in clear and
358 only then does a TLS startup, so do the same.
360 JH/13 Bug 2027: Avoid attempting to use TCP Fast Open for non-transport client
361 TCP connections (such as for Spamd) unless the daemon successfully set
362 Fast Open mode on its listening sockets. This fixes breakage seen on
363 too-old kernels or those not configured for Fast Open, at the cost of
364 requiring both directions being enabled for TFO, and TFO never being used
365 by non-daemon-related Exim processes.
367 JH/14 Bug 2000: Reject messages recieved with CHUNKING but with malformed line
368 endings, at least on the first header line. Try to canonify any that get
369 past that check, despite the cost.
371 JH/15 Angle-bracket nesting (an error inserted by broken sendmails) levels are
372 now limited to an arbitrary five deep, while parsing addresses with the
373 strip_excess_angle_brackets option enabled.
375 PP/03 Bug 2018: For Proxy Protocol and TLS-on-connect, do not over-read and
376 instead leave the unprompted TLS handshake in socket buffer for the
377 TLS library to consume.
379 PP/04 Bug 2018: Also handle Proxy Protocol v2 safely.
381 PP/05 FreeBSD compat: handle that Ports no longer create /usr/bin/perl
383 JH/16 Drop variables when they go out of scope. Memory management drops a whole
384 region in one operation, for speed, and this leaves assigned pointers
385 dangling. Add checks run only under the testsuite which checks all
386 variables at a store-reset and panics on a dangling pointer; add code
387 explicitly nulling out all the variables discovered. Fixes one known
388 bug: a transport crash, where a dangling pointer for $sending_ip_address
389 originally assigned in a verify callout, is re-used.
391 PP/06 Drop '.' from @INC in various Perl scripts.
393 PP/07 Switch FreeBSD iconv to always use the base-system libc functions.
395 PP/08 Reduce a number of compilation warnings under clang; building with
396 CC=clang CFLAGS+=-Wno-dangling-else -Wno-logical-op-parentheses
397 should be warning-free.
399 JH/17 Fix inbound CHUNKING when DKIM disabled at runtime.
401 HS/01 Fix portability problems introduced by PP/08 for platforms where
402 realloc(NULL) is not equivalent to malloc() [SunOS et al].
404 HS/02 Bug 1974: Fix missing line terminator on the last received BDAT
405 chunk. This allows us to accept broken chunked messages. We need a more
406 general solution here.
408 PP/09 Wrote util/chunking_fixqueue_finalnewlines.pl to help recover
409 already-broken messages in the queue.
411 JH/18 Bug 2061: Fix ${extract } corrupting an enclosing ${reduce } $value.
413 JH/19 Fix reference counting bug in routing-generated-address tracking.
419 JH/01 Use SIZE on MAIL FROM in a cutthrough connection, if the destination
420 supports it and a size is available (ie. the sending peer gave us one).
422 JH/02 The obsolete acl condition "demime" is removed (finally, after ten
423 years of being deprecated). The replacements are the ACLs
424 acl_smtp_mime and acl_not_smtp_mime.
426 JH/03 Upgrade security requirements imposed for hosts_try_dane: previously
427 a downgraded non-dane trust-anchor for the TLS connection (CA-style)
428 or even an in-clear connection were permitted. Now, if the host lookup
429 was dnssec and dane was requested then the host is only used if the
430 TLSA lookup succeeds and is dnssec. Further hosts (eg. lower priority
431 MXs) will be tried (for hosts_try_dane though not for hosts_require_dane)
432 if one fails this test.
433 This means that a poorly-configured remote DNS will make it incommunicado;
434 but it protects against a DNS-interception attack on it.
436 JH/04 Bug 1810: make continued-use of an open smtp transport connection
437 non-noisy when a race steals the message being considered.
439 JH/05 If main configuration option tls_certificate is unset, generate a
440 self-signed certificate for inbound TLS connections.
442 JH/06 Bug 165: hide more cases of password exposure - this time in expansions
443 in rewrites and routers.
445 JH/07 Retire gnutls_require_mac et.al. These were nonfunctional since 4.80
446 and logged a warning sing 4.83; now they are a configuration file error.
448 JH/08 Bug 1836: Fix crash in VRFY handling when handed an unqualified name
449 (lacking @domain). Apply the same qualification processing as RCPT.
451 JH/09 Bug 1804: Avoid writing msglog files when in -bh or -bhc mode.
453 JH/10 Support ${sha256:} applied to a string (as well as the previous
456 JH/11 Cutthrough: avoid using the callout hints db on a verify callout when
457 a cutthrough deliver is pending, as we always want to make a connection.
458 This also avoids re-routing the message when later placing the cutthrough
459 connection after a verify cache hit.
460 Do not update it with the verify result either.
462 JH/12 Cutthrough: disable when verify option success_on_redirect is used, and
463 when routing results in more than one destination address.
465 JH/13 Cutthrough: expand transport dkim_domain option when testing for dkim
466 signing (which inhibits the cutthrough capability). Previously only
467 the presence of an option was tested; now an expansion evaluating as
468 empty is permissible (obviously it should depend only on data available
469 when the cutthrough connection is made).
471 JH/14 Fix logging of errors under PIPELINING. Previously the log line giving
472 the relevant preceding SMTP command did not note the pipelining mode.
474 JH/15 Fix counting of empty lines in $body_linecount and $message_linecount.
475 Previously they were not counted.
477 JH/16 DANE: treat a TLSA lookup response having all non-TLSA RRs, the same
478 as one having no matching records. Previously we deferred the message
479 that needed the lookup.
481 JH/17 Fakereject: previously logged as a normal message arrival "<="; now
482 distinguished as "(=".
484 JH/18 Bug 1867: make the fail_defer_domains option on a dnslookup router work
485 for missing MX records. Previously it only worked for missing A records.
487 JH/19 Bug 1850: support Radius libraries that return REJECT_RC.
489 JH/20 Bug 1872: Ensure that acl_smtp_notquit is run when the connection drops
490 after the data-go-ahead and data-ack. Patch from Jason Betts.
492 JH/21 Bug 1846: Send DMARC forensic reports for reject and quarantine results,
493 even for a "none" policy. Patch from Tony Meyer.
495 JH/22 Fix continued use of a connection for further deliveries. If a port was
496 specified by a router, it must also match for the delivery to be
499 JH/23 Bug 1874: fix continued use of a connection for further deliveries.
500 When one of the recipients of a message was unsuitable for the connection
501 (has no matching addresses), we lost track of needing to mark it
502 deferred. As a result mail would be lost.
504 JH/24 Bug 1832: Log EHLO response on getting conn-close response for HELO.
506 JH/25 Decoding ACL controls is now done using a binary search; the source code
507 takes up less space and should be simpler to maintain. Merge the ACL
508 condition decode tables also, with similar effect.
510 JH/26 Fix problem with one_time used on a redirect router which returned the
511 parent address unchanged. A retry would see the parent address marked as
512 delivered, so not attempt the (identical) child. As a result mail would
515 JH/27 Fix a possible security hole, wherein a process operating with the Exim
516 UID can gain a root shell. Credit to http://www.halfdog.net/ for
517 discovery and writeup. Ubuntu bug 1580454; no bug raised against Exim
520 JH/28 Enable {spool,log} filesystem space and inode checks as default.
521 Main config options check_{log,spool}_{inodes,space} are now
522 100 inodes, 10MB unless set otherwise in the configuration.
524 JH/29 Fix the connection_reject log selector to apply to the connect ACL.
525 Previously it only applied to the main-section connection policy
528 JH/30 Bug 1897: fix callouts connection fallback from TLS to cleartext.
530 PP/01 Changed default Diffie-Hellman parameters to be Exim-specific, created
531 by me. Added RFC7919 DH primes as an alternative.
533 PP/02 Unbreak build via pkg-config with new hash support when crypto headers
534 are not in the system include path.
536 JH/31 Fix longstanding bug with aborted TLS server connection handling. Under
537 GnuTLS, when a session startup failed (eg because the client disconnected)
538 Exim did stdio operations after fclose. This was exposed by a recent
539 change which nulled out the file handle after the fclose.
541 JH/32 Bug 1909: Fix OCSP proof verification for cases where the proof is
542 signed directly by the cert-signing cert, rather than an intermediate
543 OCSP-signing cert. This is the model used by LetsEncrypt.
545 JH/33 Bug 1914: Ensure socket is nonblocking before draining after SMTP QUIT.
547 HS/01 Fix leak in verify callout under GnuTLS, about 3MB per recipient on
548 an incoming connection.
550 HS/02 Bug 1802: Do not half-close the connection after sending a request
553 HS/03 Use "auto" as the default EC curve parameter. For OpenSSL < 1.0.2
554 fallback to "prime256v1".
556 JH/34 SECURITY: Use proper copy of DATA command in error message.
557 Could leak key material. Remotely exploitable. CVE-2016-9963.
563 JH/01 Bug 1664: Disable OCSP for GnuTLS library versions at/before 3.3.16
564 and 3.4.4 - once the server is enabled to respond to an OCSP request
565 it does even when not requested, resulting in a stapling non-aware
566 client dropping the TLS connection.
568 TF/01 Code cleanup: Overhaul the debug_selector and log_selector machinery to
569 support variable-length bit vectors. No functional change.
571 TF/02 Improve the consistency of logging incoming and outgoing interfaces.
572 The I= interface field on outgoing lines is now after the H= remote
573 host field, same as incoming lines. There is a separate
574 outgoing_interface log selector which allows you to disable the
577 JH/02 Bug 728: Close logfiles after a daemon-process "exceptional" log write.
578 If not running log_selector +smtp_connection the mainlog would be held
579 open indefinitely after a "too many connections" event, including to a
580 deleted file after a log rotate. Leave the per net connection logging
581 leaving it open for efficiency as that will be quickly detected by the
582 check on the next write.
584 HS/01 Bug 1671: Fix post transport crash.
585 Processing the wait-<transport> messages could crash the delivery
586 process if the message IDs didn't exist for some reason. When
587 using 'split_spool_directory=yes' the construction of the spool
588 file name failed already, exposing the same netto behaviour.
590 JH/03 Bug 425: Capture substrings in $regex1, $regex2 etc from regex &
591 mime_regex ACL conditions.
593 JH/04 Bug 1686: When compiled with EXPERIMENTAL_DSN_INFO: Add extra information
594 to DSN fail messages (bounces): remote IP, remote greeting, remote response
595 to HELO, local diagnostic string.
597 JH/05 Downgrade message for a TLS-certificate-based authentication fail from
598 log line to debug. Even when configured with a tls authenticator many
599 client connections are expected to not authenticate in this way, so
600 an authenticate fail is not an error.
602 HS/02 Add the Exim version string to the process info. This way exiwhat
603 gives some more detail about the running daemon.
605 JH/06 Bug 1395: time-limit caching of DNS lookups, to the TTL value. This may
606 matter for fast-change records such as DNSBLs.
608 JH/07 Bug 1678: Always record an interface option value, if set, as part of a
609 retry record, even if constant. There may be multiple transports with
610 different interface settings and the retry behaviour needs to be kept
613 JH/08 Bug 1586: exiqgrep now refuses to run if there are unexpected arguments.
615 JH/09 Bug 1700: ignore space & tab embedded in base64 during decode.
617 JH/10 Bug 840: fix log_defer_output option of pipe transport
619 JH/11 Bug 830: use same host for all RCPTS of a message, even under
620 hosts_randomize. This matters a lot when combined with mua_wrapper.
622 JH/12 Bug 1706: percent and underbar characters are no longer escaped by the
623 ${quote_pgsql:<string>} operator.
625 JH/13 Bug 1708: avoid misaligned access in cached lookup.
627 JH/14 Change header file name for freeradius-client. Relevant if compiling
628 with Radius support; from the Gentoo tree and checked under Fedora.
630 JH/15 Bug 1712: Introduce $prdr_requested flag variable
632 JH/16 Bug 1714: Permit an empty string as expansion result for transport
633 option transport_filter, meaning no filtering.
635 JH/17 Bug 1713: Fix non-PDKIM_DEBUG build. Patch from Jasen Betts.
637 JH/18 Bug 1709: When built with TLS support, the tls_advertise_hosts option now
638 defaults to "*" (all hosts). The variable is now available when not built
639 with TLS, default unset, mainly to enable keeping the testsuite sane.
640 If a server certificate is not supplied (via tls_certificate) an error is
641 logged, and clients will find TLS connections fail on startup. Presumably
642 they will retry in-clear.
643 Packagers of Exim are strongly encouraged to create a server certificate
644 at installation time.
646 HS/03 Add -bP config_file as a synonym for -bP configure_file, for consistency
647 with the $config_file variable.
649 JH/19 Two additional event types: msg:rcpt:defer and msg:rcpt:host:defer. Both
650 in transport context, after the attempt, and per-recipient. The latter type
651 is per host attempted. The event data is the error message, and the errno
652 information encodes the lookup type (A vs. MX) used for the (first) host,
653 and the trailing two digits of the smtp 4xx response.
655 GF/01 Bug 1715: Fix for race condition in exicyclog, where exim could attempt
656 to write to mainlog (or rejectlog, paniclog) in the window between file
657 creation and permissions/ownership being changed. Particularly affects
658 installations where exicyclog is run as root, rather than exim user;
659 result is that the running daemon panics and dies.
661 JH/20 Bug 1701: For MySQL lookups, support MySQL config file option group names.
663 JH/21 Bug 1720: Add support for priority groups and weighted-random proxy
664 selection for the EXPERIMENTAL_SOCKS feature, via new per-proxy options
665 "pri" and "weight". Note that the previous implicit priority given by the
666 list order is no longer honoured.
668 JH/22 Bugs 963, 1721: Fix some corner cases in message body canonicalization
671 JH/23 Move SOCKS5 support from Experimental to mainline, enabled for a build
672 by defining SUPPORT_SOCKS.
674 JH/26 Move PROXY support from Experimental to mainline, enabled for a build
675 by defining SUPPORT_PROXY. Note that the proxy_required_hosts option
676 is renamed to hosts_proxy, and the proxy_{host,target}_{address,port}.
677 variables are renamed to proxy_{local,external}_{address,port}.
679 JH/27 Move Internationalisation support from Experimental to mainline, enabled
680 for a build by defining SUPPORT_I18N
682 JH/28 Bug 1745: Fix redis lookups to handle (quoted) spaces embedded in parts
683 of the query string, and make ${quote_redis:} do that quoting.
685 JH/29 Move Events support from Experimental to mainline, enabled by default
686 and removable for a build by defining DISABLE_EVENT.
688 JH/30 Updated DANE implementation code to current from Viktor Dukhovni.
690 JH/31 Fix bug with hosts_connection_nolog and named-lists which were wrongly
691 cached by the daemon.
693 JH/32 Move Redis support from Experimental to mainline, enabled for a build
694 by defining LOOKUP_REDIS. The libhiredis library is required.
696 JH/33 Bug 1748: Permit ACL dnslists= condition in non-smtp ACLs if explicit
697 keys are given for lookup.
699 JH/34 Bug 1192: replace the embedded copy of PolarSSL RSA routines in the DKIM
700 support, by using OpenSSL or GnuTLS library ones. This means DKIM is
701 only supported when built with TLS support. The PolarSSL SHA routines
702 are still used when the TLS library is too old for convenient support.
704 JH/35 Require SINGLE_DH_USE by default in OpenSSL (main config option
705 openssl_options), for security. OpenSSL forces this from version 1.1.0
706 server-side so match that on older versions.
708 JH/36 Bug 1778: longstanding bug in memory use by the ${run } expansion: A fresh
709 allocation for $value could be released as the expansion processing
710 concluded, but leaving the global pointer active for it.
712 JH/37 Bug 1769: Permit a VRFY ACL to override the default 252 response,
713 and to use the domains and local_parts ACL conditions.
715 JH/38 Fix cutthrough bug with body lines having a single dot. The dot was
716 incorrectly not doubled on cutthrough transmission, hence seen as a
717 body-termination at the receiving system - resulting in truncated mails.
718 Commonly the sender saw a TCP-level error, and retransmitted the message
719 via the normal store-and-forward channel. This could result in duplicates
720 received - but deduplicating mailstores were liable to retain only the
721 initial truncated version.
723 JH/39 Bug 1781: Fix use of DKIM private-keys having trailing '=' in the base-64.
725 JH/40 Fix crash in queryprogram router when compiled with EXPERIMENTAL_SRS.
727 JH/41 Bug 1792: Fix selection of headers to sign for DKIM: bottom-up. While
728 we're in there, support oversigning also; bug 1309.
730 JH/42 Bug 1796: Fix error logged on a malware scanner connection failure.
732 HS/04 Add support for keep_environment and add_environment options.
734 JH/43 Tidy coding issues detected by gcc --fsanitize=undefined. Some remain;
735 either intentional arithmetic overflow during PRNG, or testing config-
738 JH/44 Bug 1800: The combination of a -bhc commandline option and cutthrough
739 delivery resulted in actual delivery. Cancel cutthrough before DATA
742 JH/45 Fix cutthrough, when connection not opened by verify and target hard-
743 rejects a recipient: pass the reject to the originator.
745 JH/46 Multiple issues raised by Coverity. Some were obvious or plausible bugs.
746 Many were false-positives and ignorable, but it's worth fixing the
749 JH/47 Fix build on HP-UX and older Solaris, which need (un)setenv now also
750 for the new environment-manipulation done at startup. Move the routines
751 from being local to tls.c to being global via the os.c file.
753 JH/48 Bug 1807: Fix ${extract } for the numeric/3-string case. While preparsing
754 an extract embedded as result-arg for a map, the first arg for extract
755 is unavailable so we cannot tell if this is a numbered or keyed
756 extraction. Accept either.
762 JH/01 Bug 1545: The smtp transport option "retry_include_ip_address" is now
765 JH/02 The smtp transport option "multi_domain" is now expanded.
767 JH/03 The smtp transport now requests PRDR by default, if the server offers
770 JH/04 Certificate name checking on server certificates, when exim is a client,
771 is now done by default. The transport option tls_verify_cert_hostnames
772 can be used to disable this per-host. The build option
773 EXPERIMENTAL_CERTNAMES is withdrawn.
775 JH/05 The value of the tls_verify_certificates smtp transport and main options
776 default to the word "system" to access the system default CA bundle.
777 For GnuTLS, only version 3.0.20 or later.
779 JH/06 Verification of the server certificate for a TLS connection is now tried
780 (but not required) by default. The verification status is now logged by
781 default, for both outbound TLS and client-certificate supplying inbound
784 JH/07 Changed the default rfc1413 lookup settings to disable calls. Few
787 JH/08 The EXPERIMENTAL_DSN compile option is no longer needed; all Delivery
788 Status Notification (bounce) messages are now MIME format per RFC 3464.
789 Support for RFC 3461 DSN options NOTIFY,ENVID,RET,ORCPT can be advertised
790 under the control of the dsn_advertise_hosts option, and routers may
791 have a dsn_lasthop option.
793 JH/09 A timeout of 2 minutes is now applied to all malware scanner types by
794 default, modifiable by a malware= option. The list separator for
795 the options can now be changed in the usual way. Bug 68.
797 JH/10 The smtp_receive_timeout main option is now expanded before use.
799 JH/11 The incoming_interface log option now also enables logging of the
800 local interface on delivery outgoing connections.
802 JH/12 The cutthrough-routing facility now supports multi-recipient mails,
803 if the interface and destination host and port all match.
805 JH/13 Bug 344: The verify = reverse_host_lookup ACL condition now accepts a
808 JH/14 Bug 1573: The spam= ACL condition now additionally supports Rspamd.
809 Patch from Andrew Lewis.
811 JH/15 Bug 670: The spamd_address main option (for the spam= ACL condition)
812 now supports optional time-restrictions, weighting, and priority
813 modifiers per server. Patch originally by <rommer@active.by>.
815 JH/16 The spamd_address main option now supports a mixed list of local
816 and remote servers. Remote servers can be IPv6 addresses, and
817 specify a port-range.
819 JH/17 Bug 68: The spamd_address main option now supports an optional
820 timeout value per server.
822 JH/18 Bug 1581: Router and transport options headers_add/remove can
823 now have the list separator specified.
825 JH/19 Bug 392: spamd_address, and clamd av_scanner, now support retry
828 JH/20 Bug 1571: Ensure that $tls_in_peerdn is set, when verification fails
831 JH/21 Support for the A6 type of dns record is withdrawn.
833 JH/22 Bug 608: The result of a QUIT or not-QUIT toplevel ACL now matters
834 rather than the verbs used.
836 JH/23 Bug 1572: Increase limit on SMTP confirmation message copy size
837 from 255 to 1024 chars.
839 JH/24 Verification callouts now attempt to use TLS by default.
841 HS/01 DNSSEC options (dnssec_require_domains, dnssec_request_domains)
842 are generic router options now. The defaults didn't change.
844 JH/25 Bug 466: Add RFC2322 support for MIME attachment filenames.
845 Original patch from Alexander Shikoff, worked over by JH.
847 HS/02 Bug 1575: exigrep falls back to autodetection of compressed
848 files if ZCAT_COMMAND is not executable.
850 JH/26 Bug 1539: Add timeout/retry options on dnsdb lookups.
852 JH/27 Bug 286: Support SOA lookup in dnsdb lookups.
854 JH/28 Bug 1588: Do not use the A lookup following an AAAA for setting the FQDN.
855 Normally benign, it bites when the pair was led to by a CNAME;
856 modern usage is to not canonicalize the domain to a CNAME target
857 (and we were inconsistent anyway for A-only vs AAAA+A).
859 JH/29 Bug 1632: Removed the word "rejected" from line logged for ACL discards.
861 JH/30 Check the forward DNS lookup for DNSSEC, in addition to the reverse,
862 when evaluating $sender_host_dnssec.
864 JH/31 Check the HELO verification lookup for DNSSEC, adding new
865 $sender_helo_dnssec variable.
867 JH/32 Bug 1397: Enable ECDHE on OpenSSL, just the NIST P-256 curve.
869 JH/33 Bug 1346: Note MAIL cmd seen in -bS batch, to avoid smtp_no_mail log.
871 JH/34 Bug 1648: Fix a memory leak seen with "mailq" and large queues.
873 JH/35 Bug 1642: Fix support of $spam_ variables at delivery time. Was
874 documented as working, but never had. Support all but $spam_report.
876 JH/36 Bug 1659: Guard checking of input smtp commands again pseudo-command
877 added for tls authenticator.
879 HS/03 Add perl_taintmode main config option
885 TL/01 When running the test suite, the README says that variables such as
886 no_msglog_check are global and can be placed anywhere in a specific
887 test's script, however it was observed that placement needed to be near
888 the beginning for it to behave that way. Changed the runtest perl
889 script to read through the entire script once to detect and set these
890 variables, reset to the beginning of the script, and then run through
891 the script parsing/test process like normal.
893 TL/02 The BSD's have an arc4random API. One of the functions to induce
894 adding randomness was arc4random_stir(), but it has been removed in
895 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
896 function when detected.
898 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
899 cause callback expansion.
901 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
902 syntax errors in an expansion can be treated as a string instead of
903 logging or causing an error, due to the internal use of bool_lax
904 instead of bool when processing it.
906 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
907 server certificates when making smtp deliveries.
909 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
911 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
913 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
915 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
916 Merged patch from Sebastian Wiedenroth.
918 JH/05 Fix results-pipe from transport process. Several recipients, combined
919 with certificate use, exposed issues where response data items split
920 over buffer boundaries were not parsed properly. This eventually
921 resulted in duplicates being sent. This issue only became common enough
922 to notice due to the introduction of connection certificate information,
923 the item size being so much larger. Found and fixed by Wolfgang Breyha.
925 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
926 size buffer was used, resulting in syntax errors when an expansion
929 JH/07 Add support for directories of certificates when compiled with a GnuTLS
930 version 3.3.6 or later.
932 JH/08 Rename the TPDA experimental facility to Event Actions. The #ifdef
933 is EXPERIMENTAL_EVENT, the main-configuration and transport options
934 both become "event_action", the variables become $event_name, $event_data
935 and $event_defer_errno. There is a new variable $verify_mode, usable in
936 routers, transports and related events. The tls:cert event is now also
937 raised for inbound connections, if the main configuration event_action
940 TL/06 In test suite, disable OCSP for old versions of openssl which contained
941 early OCSP support, but no stapling (appears to be less than 1.0.0).
943 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
944 server certificate names available under the smtp transport option
945 "tls_verify_cert_hostname" now do not permit multi-component wildcard
948 JH/10 Time-related extraction expansions from certificates now use the main
949 option "timezone" setting for output formatting, and are consistent
950 between OpenSSL and GnuTLS compilations. Bug 1541.
952 JH/11 Fix a crash in mime ACL when meeting a zero-length, quoted or RFC2047-
953 encoded parameter in the incoming message. Bug 1558.
955 JH/12 Bug 1527: Autogrow buffer used in reading spool files. Since they now
956 include certificate info, eximon was claiming there were spoolfile
959 JH/13 Bug 1521: Fix ldap lookup for single-attr request, multiple-attr return.
961 JH/14 Log delivery-related information more consistently, using the sequence
962 "H=<name> [<ip>]" wherever possible.
964 TL/07 Bug 1547: Omit RFCs from release. Draft and RFCs have licenses which
965 are problematic for Debian distribution, omit them from the release
968 JH/15 Updates and fixes to the EXPERIMENTAL_DSN feature.
970 JH/16 Fix string representation of time values on 64bit time_t architectures.
973 JH/17 Fix a null-indirection in certextract expansions when a nondefault
974 output list separator was used.
979 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
980 checkers that were complaining about end of non-void function with no
983 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
984 This was a regression introduced in 4.83 by another bugfix.
986 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
988 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
989 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
995 TF/01 Correctly close the server side of TLS when forking for delivery.
997 When a message was received over SMTP with TLS, Exim failed to clear up
998 the incoming connection properly after forking off the child process to
999 deliver the message. In some situations the subsequent outgoing
1000 delivery connection happened to have the same fd number as the incoming
1001 connection previously had. Exim would try to use TLS and fail, logging
1002 a "Bad file descriptor" error.
1004 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
1005 utilities have not been installed.
1007 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
1008 temporary space as the ACL may create new global variables.
1010 TL/01 LDAP support uses per connection or global context settings, depending
1011 upon the detected version of the libraries at build time.
1013 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
1014 to extract and use the src ip:port in logging and expansions as if it
1015 were a direct connection from the outside internet. PPv2 support was
1016 updated based on HAProxy spec change in May 2014.
1018 JH/02 Add ${listextract {number}{list}{success}{fail}}.
1020 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
1021 Properly escape header and check for NULL return.
1023 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
1026 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
1028 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
1029 characters in header names, implemented as a verify condition.
1030 Contributed by Michael Fischer v. Mollard.
1032 TL/05 Rename SPF condition results err_perm and err_temp to standardized
1033 results permerror and temperror. Previous values are deprecated but
1034 still accepted. In a future release, err_perm and err_temp will be
1035 completely removed, which will be a backward incompatibility if the
1036 ACL tests for either of these two old results. Patch contributed by
1037 user bes-internal on the mailing list.
1039 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
1041 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
1042 selectors, in both main and reject logs.
1044 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
1047 JH/07 Add malware type "sock" for talking to simple daemon.
1049 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
1051 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
1052 routers/transports under cutthrough routing.
1054 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
1055 numbers. Touch up "bool" conditional to keep the same definition.
1057 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
1059 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
1061 JH/12 Expand items in router/transport headers_add or headers_remove lists
1062 individually rather than the list as a whole. Bug 1452.
1064 Required for reasonable handling of multiple headers_ options when
1065 they may be empty; requires that headers_remove items with embedded
1066 colons must have them doubled (or the list-separator changed).
1068 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
1069 view the policy declared in the DMARC record. Currently, $dmarc_status
1070 is a combined value of both the record presence and the result of the
1073 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
1075 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
1076 dnslookup router and the smtp transport (applying to the forward
1079 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
1080 of ldap servers used for a specific lookup. Patch provided by Heiko
1083 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
1084 New variable $lookup_dnssec_authenticated for observability.
1086 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
1087 Patch submitted by Lars Timman.
1089 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
1091 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
1092 Requires trusted mode and valid format message id, aborts otherwise.
1093 Patch contributed by Heiko Schlichting.
1095 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
1096 certextract with support for various fields. Bug 1358.
1098 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
1099 is requested by default, modifiable by smtp transport option
1102 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
1103 operate on certificate variables to give certificate fingerprints
1104 Also new ${sha256:cert_variable}.
1106 JH/23 The PRDR feature is moved from being Experimental into the mainline.
1108 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
1109 Christian Aistleitner.
1111 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
1113 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
1114 file. Patch from Wolfgang Breyha.
1116 JH/25 Expand the coverage of the delivery $host and $host_address to
1117 client authenticators run in verify callout. Bug 1476.
1119 JH/26 Port service names are now accepted for tls_on_connect_ports, to
1120 align with daemon_smtp_ports. Bug 72.
1122 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
1123 support and error reporting did not work properly.
1125 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
1126 and is readable. Patch from Andrew Colin Kissa.
1128 TL/14 Enhance documentation of ${run expansion and how it parses the
1129 commandline after expansion, particularly in the case when an
1130 unquoted variable expansion results in an empty value.
1132 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
1134 PP/02 Fix internal collision of T_APL on systems which support RFC3123
1135 by renaming away from it. Addresses GH issue 15, reported by
1138 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
1140 TL/15 SECURITY: prevent double expansion in math comparison functions
1141 (can expand unsanitized data). Not remotely exploitable.
1148 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
1150 PP/02 Make -n do something, by making it not do something.
1151 When combined with -bP, the name of an option is not output.
1153 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
1156 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
1157 $sender_host_name and config options to manage this, and basic check
1160 PP/05 DSCP support for outbound connections and control modifier for inbound.
1162 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
1163 (Only plugin which currently uses this is kerberos4, which nobody should
1164 be using, but we should make it available and other future plugins might
1165 conceivably use it, even though it would break NAT; stuff *should* be
1166 using channel bindings instead).
1168 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
1169 name; added for Sendmail compatibility; requires admin caller.
1170 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
1171 just ignore it); requires trusted caller.
1172 Also parse but ignore: -Ac -Am -X<logfile>
1175 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
1177 TL/02 Add +smtp_confirmation as a default logging option.
1179 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
1180 Patch by Magnus Holmgren from 2007-02-20.
1182 TL/04 Bugzilla 1281 - Spec typo.
1183 Bugzilla 1283 - Spec typo.
1184 Bugzilla 1290 - Spec grammar fixes.
1186 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
1188 TL/06 Add Experimental DMARC support using libopendmarc libraries.
1190 TL/07 Fix an out of order global option causing a segfault. Reported to dev
1191 mailing list by by Dmitry Isaikin.
1193 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
1195 JH/02 Support "G" suffix to numbers in ${if comparisons.
1197 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
1199 NM/01 Bugzilla 1197 - Spec typo
1200 Bugzilla 1196 - Spec examples corrections
1202 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
1204 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
1205 gnutls_enable_pkcs11, but renamed to more accurately indicate its
1208 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
1209 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
1211 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
1212 "acl {{name}{arg}...}", and optional args on acl condition
1215 JH/05 Permit multiple router/transport headers_add/remove lines.
1217 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
1219 JH/07 Avoid using a waiting database for a single-message-only transport.
1220 Performance patch from Paul Fisher. Bugzilla 1262.
1222 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
1225 JH/09 Add $headers_added variable, with content from use of ACL modifier
1226 add_header (but not yet added to the message). Bugzilla 199.
1228 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
1229 Pulled from Bugzilla 817 by Wolfgang Breyha.
1231 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
1233 (nb: this is the same fix as in Exim 4.80.1)
1235 JH/11 Add A= logging on delivery lines, and a client_set_id option on
1238 JH/12 Add optional authenticated_sender logging to A= and a log_selector
1241 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
1243 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
1244 advertise SMTP AUTH mechanism to us, instead of a generic
1245 protocol violation error. Also, make Exim more robust to bad
1246 data from the Dovecot auth socket.
1248 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
1250 When a queue runner is handling a message, Exim first routes the
1251 recipient addresses, during which it prunes them based on the retry
1252 hints database. After that it attempts to deliver the message to
1253 any remaining recipients. It then updates the hints database using
1256 So if a recipient address works intermittently, it can get repeatedly
1257 deferred at routing time. The retry hints record remains fresh so the
1258 address never reaches the final cutoff time.
1260 This is a fairly common occurrence when a user is bumping up against
1261 their storage quota. Exim had some logic in its local delivery code
1262 to deal with this. However it did not apply to per-recipient defers
1263 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
1265 This change adds a proper retry rule check during routing so that the
1266 final cutoff time is checked against the message's age. We only do
1267 this check if there is an address retry record and there is not a
1268 domain retry record; this implies that previous attempts to handle
1269 the address had the retry_use_local_parts option turned on. We use
1270 this as an approximation for the destination being like a local
1271 delivery, as in LMTP.
1273 I suspect this new check makes the old local delivery cutoff check
1274 redundant, but I have not verified this so I left the code in place.
1276 TF/02 Correct gecos expansion when From: is a prefix of the username.
1278 Test 0254 submits a message to Exim with the header
1282 When I ran the test suite under the user fanf2, Exim expanded
1283 the header to contain my full name, whereas it should have added
1284 a Resent-Sender: header. It erroneously treats any prefix of the
1285 username as equal to the username.
1287 This change corrects that bug.
1289 GF/01 DCC debug and logging tidyup
1290 Error conditions log to paniclog rather than rejectlog.
1291 Debug lines prefixed by "DCC: " to remove any ambiguity.
1293 TF/03 Avoid unnecessary rebuilds of lookup-related code.
1295 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
1296 Bug spotted by Jeremy Harris; was flawed since initial commit.
1297 Would have resulted in OCSP responses post-SNI triggering an Exim
1298 NULL dereference and crash.
1300 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
1302 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
1303 Bug detection, analysis and fix by Samuel Thibault.
1304 Bugzilla 1331, Debian bug #698092.
1306 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
1308 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
1309 Server implementation by Todd Lyons, client by JH.
1310 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
1311 config variable "prdr_enable" controls whether the server
1312 advertises the facility. If the client requests PRDR a new
1313 acl_data_smtp_prdr ACL is called once for each recipient, after
1314 the body content is received and before the acl_smtp_data ACL.
1315 The client is controlled by both of: a hosts_try_prdr option
1316 on the smtp transport, and the server advertisement.
1317 Default client logging of deliveries and rejections involving
1318 PRDR are flagged with the string "PRDR".
1320 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
1321 fclose(). Diagnosis by Todd Lyons.
1323 PP/17 Update configure.default to handle IPv6 localhost better.
1324 Patch by Alain Williams (plus minor tweaks).
1327 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
1328 This is now consistent with GnuTLS, and is now documented: the
1329 previous undocumented portable approach to treating the option as
1330 unset was to force an expansion failure. That still works, and
1331 an empty string is now equivalent.
1333 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
1334 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
1335 not performing validation itself.
1337 PP/20 Added force_command boolean option to pipe transport.
1338 Patch from Nick Koston, of cPanel Inc.
1340 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
1343 TF/04 Added udpsend ACL modifier and hexquote expansion operator
1345 PP/21 Fix eximon continuous updating with timestamped log-files.
1346 Broken in a format-string cleanup in 4.80, missed when I repaired the
1347 other false fix of the same issue.
1348 Report and fix from Heiko Schlichting.
1351 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
1352 Report from Prashanth Katuri.
1354 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
1355 It's SecureTransport, so affects any MacOS clients which use the
1356 system-integrated TLS libraries, including email clients.
1358 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
1359 using a MIME ACL for non-SMTP local injection.
1360 Report and assistance in diagnosis by Warren Baker.
1362 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
1364 JH/16 Fix comparisons for 64b. Bugzilla 1385.
1366 TL/09 Add expansion variable $authenticated_fail_id to keep track of
1367 last id that failed so it may be referenced in subsequent ACL's.
1369 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
1372 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
1373 ldap library initialization, allowing self-signed CA's to be
1374 used. Also properly sets require_cert option later in code by
1375 using NULL (global ldap config) instead of ldap handle (per
1376 session). Bug diagnosis and testing by alxgomz.
1378 TL/12 Enhanced documentation in the ratelimit.pl script provided in
1379 the src/util/ subdirectory.
1381 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
1382 renamed to Transport Post Delivery Action by Jeremy Harris, as
1385 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
1386 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
1387 redis_servers = needs to be configured which will be used by the redis
1388 lookup. Patch from Warren Baker, of The Packet Hub.
1390 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
1392 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
1393 hostname or reverse DNS when processing a host list. Used suggestions
1394 from multiple comments on this bug.
1396 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
1398 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
1399 Missed a few lines, added it to make the runtest require no keyboard
1402 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
1403 contains upper case chars. Make router use caseful_local_part.
1405 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
1406 support when GnuTLS has been built with p11-kit.
1412 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
1414 This, or similar/improved, will also be change PP/11 of 4.82.
1420 PP/01 Handle short writes when writing local log-files.
1421 In practice, only affects FreeBSD (8 onwards).
1422 Bugzilla 1053, with thanks to Dmitry Isaikin.
1424 NM/01 Bugzilla 949 - Documentation tweak
1426 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
1429 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
1431 PP/02 Implemented gsasl authenticator.
1433 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
1435 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
1436 `pkg-config foo` for cflags/libs.
1438 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
1439 with rest of GSASL and with heimdal_gssapi.
1441 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
1442 `pkg-config foo` for cflags/libs for the TLS implementation.
1444 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
1445 properties get this fed in as external SSF. A number of robustness
1446 and debugging improvements to the cyrus_sasl authenticator.
1448 PP/08 cyrus_sasl server now expands the server_realm option.
1450 PP/09 Bugzilla 1214 - Log authentication information in reject log.
1451 Patch by Jeremy Harris.
1453 PP/10 Added dbmjz lookup type.
1455 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
1457 PP/12 MAIL args handles TAB as well as SP, for better interop with
1458 non-compliant senders.
1459 Analysis and variant patch by Todd Lyons.
1461 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
1462 Bug report from Lars Müller <lars@samba.org> (via SUSE),
1463 Patch from Dirk Mueller <dmueller@suse.com>
1465 PP/13 tls_peerdn now print-escaped for spool files.
1466 Observed some $tls_peerdn in wild which contained \n, which resulted
1467 in spool file corruption.
1469 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
1470 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
1471 or write after TLS renegotiation, which otherwise led to messages
1474 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
1475 as a tracking header (ie: a signed header comes before the signature).
1476 Patch from Wolfgang Breyha.
1478 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
1479 comma-sep list; embedded commas doubled.
1481 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
1483 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
1485 Report and patch from Dmitry Banschikov.
1487 PP/16 Removed "dont_insert_empty_fragments" from "openssl_options".
1488 Removed SSL_clear() after SSL_new() which led to protocol negotiation
1489 failures. We appear to now support TLS1.1+ with Exim.
1491 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
1492 lets Exim select keys and certificates based upon TLS SNI from client.
1493 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
1494 before an outbound SMTP session. New log_selector, +tls_sni.
1496 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
1497 NULL dereference. Report and patch from Alun Jones.
1499 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
1500 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
1501 Not seeing resolver debug output on NetBSD, but suspect this is a
1502 resolver implementation change.
1504 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
1505 Left warnings. Added "eximon gdb" invocation mode.
1507 PP/21 Defaulting "accept_8bitmime" to true, not false.
1509 PP/22 Added -bw for inetd wait mode support.
1511 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
1512 locate the relevant includes and libraries. Made this the default.
1514 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
1515 Bugzilla 1246, report and most of solution from Tomasz Kusy.
1517 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
1518 This may cause build issues on older platforms.
1520 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
1521 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
1522 gnutls_require_mac & gnutls_require_protocols (no longer supported).
1523 Added SNI support via GnuTLS too.
1524 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
1526 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
1528 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
1529 Applied second patch from Janne, implementing suggestion to default
1530 multiple-strings-in-record handling to match SPF spec.
1532 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
1534 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
1535 read-only, out of scope).
1536 Patch from Wolfgang Breyha, report from Stuart Northfield.
1538 PP/29 Fix three issues highlighted by clang analyser static analysis.
1539 Only crash-plausible issue would require the Cambridge-specific
1540 iplookup router and a misconfiguration.
1541 Report from Marcin Mirosław.
1543 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
1545 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
1546 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
1547 As part of this, removing so much warning spew let me fix some minor
1548 real issues in debug logging.
1550 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
1551 assignment on my part. Fixed.
1553 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
1554 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
1555 Janne Snabb (who went above and beyond: thank you).
1557 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
1558 string otherwise requires a connection and a bunch more work and it's
1559 relatively easy to get wrong. Should also expose TLS library linkage
1562 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
1563 64-bit ${eval} (JH/03).
1565 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
1566 GNU libc to support some of the 64-bit stuff, should not lead to
1567 conflicts. Defined before os.h is pulled in, so if a given platform
1568 needs to override this, it can.
1570 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
1571 protection layer was required, which is not implemented.
1572 Bugzilla 1254, patch from Wolfgang Breyha.
1574 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
1575 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
1576 tls_dhparam take prime identifiers. Also unbreak combination of
1577 OpenSSL+DH_params+TLSSNI.
1579 PP/39 Disable SSLv2 by default in OpenSSL support.
1585 PP/01 Solaris build fix for Oracle's LDAP libraries.
1586 Bugzilla 1109, patch from Stephen Usher.
1588 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
1590 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
1593 TF/02 Fix a couple more cases where we did not log the error message
1594 when unlink() failed. See also change 4.74-TF/03.
1596 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
1597 lock up or crash if it happened to be inside a call to libc when it
1598 got a SIGUSR1 from exiwhat.
1600 The SIGUSR1 handler appends the current process status to the process
1601 log which is later printed by exiwhat. It used to use the general
1602 purpose logging code to do this, but several functions it calls are
1603 not safe for signals.
1605 The new output code in the SIGUSR1 handler is specific to the process
1606 log, and simple enough that it's easy to inspect for signal safety.
1607 Removing some special cases also simplifies the general logging code.
1608 Removing the spurious timestamps from the process log simplifies
1611 TF/04 Improved ratelimit ACL condition.
1613 The /noupdate option has been deprecated in favour of /readonly which
1614 has clearer semantics. The /leaky, /strict, and /readonly update modes
1615 are mutually exclusive. The update mode is no longer included in the
1616 database key; it just determines when the database is updated. (This
1617 means that when you upgrade Exim will forget old rate measurements.)
1619 Exim now checks that the per_* options are used with an update mode that
1620 makes sense for the current ACL. For example, when Exim is processing a
1621 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
1622 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
1623 must specify per_mail/readonly. If you omit the update mode it defaults to
1624 /leaky where that makes sense (as before) or /readonly where required.
1626 The /noupdate option is now undocumented but still supported for
1627 backwards compatibility. It is equivalent to /readonly except that in
1628 ACLs where /readonly is required you may specify /leaky/noupdate or
1629 /strict/noupdate which are treated the same as /readonly.
1631 A useful new feature is the /count= option. This is a generalization
1632 of the per_byte option, so that you can measure the throughput of other
1633 aggregate values. For example, the per_byte option is now equivalent
1634 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
1636 The per_rcpt option has been generalized using the /count= mechanism
1637 (though it's more complicated than the per_byte equivalence). When it is
1638 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
1639 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
1640 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
1641 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
1642 Note that using per_rcpt with a non-readonly update mode in more than
1643 one ACL will cause the recipients to be double-counted. (The per_mail
1644 and per_byte options don't have this problem.)
1646 The handling of very low rates has changed slightly. If the computed rate
1647 is less than the event's count (usually one) then this event is the first
1648 after a long gap. In this case the rate is set to the same as this event's
1649 count, so that the first message of a spam run is counted properly.
1651 The major new feature is a mechanism for counting the rate of unique
1652 events. The new per_addr option counts the number of different
1653 recipients that someone has sent messages to in the last time period. It
1654 behaves like per_rcpt if all the recipient addresses are different, but
1655 duplicate recipient addresses do not increase the measured rate. Like
1656 the /count= option this is a general mechanism, so the per_addr option
1657 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
1658 example, measure the rate that a client uses different sender addresses
1659 with the options per_mail/unique=$sender_address. There are further
1660 details in the main documentation.
1662 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
1664 TF/06 Removed a few PCRE remnants.
1666 TF/07 Automatically extract Exim's version number from tags in the git
1667 repository when doing development or release builds.
1669 PP/02 Raise smtp_cmd_buffer_size to 16kB.
1670 Bugzilla 879. Patch from Paul Fisher.
1672 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
1673 Heavily based on revision 40f9a89a from Simon Arlott's tree.
1676 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
1678 PP/05 Variable $av_failed, true if the AV scanner deferred.
1679 Bugzilla 1078. Patch from John Horne.
1681 PP/06 Stop make process more reliably on build failure.
1682 Bugzilla 1087. Patch from Heiko Schlittermann.
1684 PP/07 Make maildir_use_size_file an _expandable_ boolean.
1685 Bugzilla 1089. Patch from Heiko Schlittermann.
1687 PP/08 Handle ${run} returning more data than OS pipe buffer size.
1688 Bugzilla 1131. Patch from Holger Weiß.
1690 PP/09 Handle IPv6 addresses with SPF.
1691 Bugzilla 860. Patch from Wolfgang Breyha.
1693 PP/10 GnuTLS: support TLS 1.2 & 1.1.
1695 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
1698 PP/11 match_* no longer expand right-hand-side by default.
1699 New compile-time build option, EXPAND_LISTMATCH_RHS.
1700 New expansion conditions, "inlist", "inlisti".
1702 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
1704 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
1706 PP/14 fix log_write() format string regression from TF/03.
1707 Bugzilla 1152. Patch from Dmitry Isaikin.
1713 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
1715 PP/02 Harmonised TLS library version reporting; only show if debugging.
1716 Layout now matches that introduced for other libraries in 4.74 PP/03.
1718 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
1720 PP/04 New "dns_use_edns0" global option.
1722 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
1725 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
1726 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
1728 TK/01 Updated PolarSSL code to 0.14.2.
1729 Bugzilla 1097. Patch from Andreas Metzler.
1731 PP/07 Catch divide-by-zero in ${eval:...}.
1732 Fixes bugzilla 1102.
1734 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
1737 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
1738 format-string attack -- SECURITY: remote arbitrary code execution.
1740 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
1741 time unintentionally subject to list matching rules, letting the header
1742 cause arbitrary Exim lookups (of items which can occur in lists, *not*
1743 arbitrary string expansion). This allowed for information disclosure.
1745 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
1746 INT_MIN/-1 -- value coerced to INT_MAX.
1752 NM/01 Workaround for PCRE version dependency in version reporting
1755 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
1756 This fixes portability to compilers other than gcc, notably
1757 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
1759 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
1760 makefiles for portability to HP-UX and POSIX correctness.
1762 PP/01 Permit LOOKUP_foo enabling on the make command-line.
1763 Also via indented variable definition in the Makefile.
1764 (Debugging by Oliver Heesakkers).
1766 PP/02 Restore caching of spamd results with expanded spamd_address.
1767 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
1769 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
1770 Improves build reliability. Fix from: Frank Elsner
1772 NM/02 Fix wide character breakage in the rfc2047 coding
1773 Fixes bug 1064. Patch from Andrey N. Oktyabrski
1775 NM/03 Allow underscore in dnslist lookups
1776 Fixes bug 1026. Patch from Graeme Fowler
1778 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
1779 Code patches from Adam Ciarcinski of NetBSD.
1781 NM/04 Fixed exiqgrep to cope with mailq missing size issue
1784 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
1785 is logged, to avoid truncation. Patch from John Horne.
1787 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
1788 Patch from Jakob Hirsch.
1790 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
1791 SQL string expansion failure details.
1792 Patch from Andrey Oktyabrski.
1794 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
1795 Patch from Simon Arlott.
1797 PP/09 New lookups functionality failed to compile on old gcc which rejects
1798 extern declarations in function scope.
1799 Patch from Oliver Fleischmann
1801 PP/10 Use sig_atomic_t for flags set from signal handlers.
1802 Check getgroups() return and improve debugging.
1803 Fixed developed for diagnosis in bug 927 (which turned out to be
1806 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
1807 Patch from Mark Zealey.
1809 PP/12 Bugzilla 1056: Improved spamd server selection.
1810 Patch from Mark Zealey.
1812 PP/13 Bugzilla 1086: Deal with maildir quota file races.
1813 Based on patch from Heiko Schlittermann.
1815 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
1816 Patch from Uwe Doering, sign-off by Michael Haardt.
1818 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
1819 variable declaration deep within a block. Bug and patch from
1822 PP/15 lookups-Makefile IRIX compatibility coercion.
1824 PP/16 Make DISABLE_DKIM build knob functional.
1826 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
1827 Patch by Simon Arlott
1829 TF/03 Fix valgrind.h portability to C89 compilers that do not support
1830 variable argument macros. Our copy now differs from upstream.
1836 TF/01 Failure to get a lock on a hints database can have serious
1837 consequences so log it to the panic log.
1839 TF/02 Log LMTP confirmation messages in the same way as SMTP,
1840 controlled using the smtp_confirmation log selector.
1842 TF/03 Include the error message when we fail to unlink a spool file.
1844 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
1845 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
1846 for maintaining out-of-tree patches for some time.
1848 PP/01 Bugzilla 139: Documentation and portability issues.
1849 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
1850 Handle per-OS dynamic-module compilation flags.
1852 PP/02 Let /dev/null have normal permissions.
1853 The 4.73 fixes were a little too stringent and complained about the
1854 permissions on /dev/null. Exempt it from some checks.
1855 Reported by Andreas M. Kirchwitz.
1857 PP/03 Report version information for many libraries, including
1858 Exim version information for dynamically loaded libraries. Created
1859 version.h, now support a version extension string for distributors
1860 who patch heavily. Dynamic module ABI change.
1862 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
1863 privilege escalation vulnerability whereby the Exim run-time user
1864 can cause root to append content of the attacker's choosing to
1867 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
1870 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
1871 If dropping privileges for untrusted macros, we disabled normal logging
1872 on the basis that it would fail; for the Exim run-time user, this is not
1873 the case, and it resulted in successful deliveries going unlogged.
1874 Fixed. Reported by Andreas Metzler.
1880 PP/01 Date: & Message-Id: revert to normally being appended to a message,
1881 only prepend for the Resent-* case. Fixes regression introduced in
1882 Exim 4.70 by NM/22 for Bugzilla 607.
1884 PP/02 Include check_rfc2047_length in configure.default because we're seeing
1885 increasing numbers of administrators be bitten by this.
1887 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
1889 PP/03 Bugzilla 994: added openssl_options main configuration option.
1891 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
1893 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
1895 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
1897 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
1898 without a peer certificate, leading to a segfault because of an
1899 assumption that peers always have certificates. Be a little more
1900 paranoid. Problem reported by Martin Tscholak.
1902 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
1903 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
1904 NB: ClamAV planning to remove STREAM in "middle of 2010".
1905 CL also introduces -bmalware, various -d+acl logging additions and
1906 more caution in buffer sizes.
1908 PP/09 Implemented reverse_ip expansion operator.
1910 PP/10 Bugzilla 937: provide a "debug" ACL control.
1912 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
1914 PP/12 Bugzilla 973: Implement --version.
1916 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
1918 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
1920 PP/15 Bugzilla 816: support multiple condition rules on Routers.
1922 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
1923 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
1924 ignore trailing whitespace.
1926 JJ/02 prevent non-panic DKIM error from being sent to paniclog
1928 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
1931 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
1932 Notification from Dr Andrew Aitchison.
1934 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
1935 ExtendedDetectionInfo response format.
1936 Notification from John Horne.
1938 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
1941 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
1942 XSL and documented dependency on system catalogs, with examples of how
1945 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
1948 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
1949 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
1950 configuration file which is writeable by the Exim user or group.
1952 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
1953 of configuration files to cover files specified with the -C option if
1954 they are going to be used with root privileges, not just the default
1957 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
1958 option (effectively making it always true).
1960 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
1961 files to be used while preserving root privileges.
1963 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
1964 that rogue child processes cannot use them.
1966 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
1967 run-time user, instead of root.
1969 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
1970 Exim run-time user without dropping privileges.
1972 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
1973 result string, instead of calling string_vformat() twice with the same
1976 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
1977 for other users. Others should always drop root privileges if they use
1978 -C on the command line, even for a whitelisted configure file.
1980 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
1982 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
1988 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
1989 $data_path, and $header_path variables; fixed documentation bugs and
1992 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
1993 exipick to access non-standard spools, including the "frozen" queue
1996 NM/01 Bugzilla 965: Support mysql stored procedures.
1997 Patch from Alain Williams
1999 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
2001 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
2002 Patch from Andreas Metzler
2004 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
2005 Patch from Kirill Miazine
2007 NM/05 Bugzilla 671: Added umask to procmail example.
2009 JJ/03 installed exipick 20100323.0, fixing doc bug
2011 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
2012 directory. Notification and patch from Dan Rosenberg.
2014 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
2016 TK/02 Improve log output when DKIM signing operation fails.
2018 MH/01 Treat the transport option dkim_domain as a colon separated
2019 list, not as a single string, and sign the message with each element,
2020 omitting multiple occurences of the same signer.
2022 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
2023 Bugzilla 985, 986. Patch by Simon Arlott
2025 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
2026 Patch by Simon Arlott
2028 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
2029 MBX locking. Notification from Dan Rosenberg.
2035 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
2037 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
2039 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
2041 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
2043 NM/04 Bugzilla 915: Flag broken perl installation during build.
2049 TK/01 Added patch by Johannes Berg that expands the main option
2050 "spamd_address" if it starts with a dollar sign.
2052 TK/02 Write list of recipients to X-Envelope-Sender header when building
2053 the mbox-format spool file for content scanning (suggested by Jakob
2056 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
2057 (http://www.dcc-servers.net/) support via dccifd. Activated by
2058 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
2060 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
2061 by Mark Daniel Reidel <mr@df.eu>.
2063 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
2064 When building exim an external PCRE library is now needed -
2065 PCRE is a system library on the majority of modern systems.
2066 See entry on PCRE_LIBS in EDITME file.
2068 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
2069 conversation. Added nologin parameter to request.
2070 Patch contributed by Kirill Miazine.
2072 TF/01 Do not log submission mode rewrites if they do not change the address.
2074 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
2076 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
2077 log files in place. Contributed by Roberto Lima.
2079 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
2081 TF/03 Bugzilla 615: When checking the local_parts router precondition
2082 after a local_part_suffix or local_part_prefix option, Exim now
2083 does not use the address's named list lookup cache, since this
2084 contains cached lookups for the whole local part.
2086 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
2087 Robert Millan. Documentation is in experimental-spec.txt.
2089 TF/04 Bugzilla 668: Fix parallel build (make -j).
2091 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
2093 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
2094 Patch provided by Jan Srzednicki.
2096 TF/05 Leading white space used to be stripped from $spam_report which
2097 wrecked the formatting. Now it is preserved.
2099 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
2100 that they are available at delivery time.
2102 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
2104 TF/08 TLS error reporting now respects the incoming_interface and
2105 incoming_port log selectors.
2107 TF/09 Produce a more useful error message if an SMTP transport's hosts
2108 setting expands to an empty string.
2110 NM/06 Bugzilla 744: EXPN did not work under TLS.
2111 Patch provided by Phil Pennock.
2113 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
2114 Patch provided by Richard Godbee.
2116 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
2117 acl_smtp_notquit, added index entry.
2119 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
2120 Patch provided by Eugene Bujak.
2122 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
2123 accept(). Patch provided by Maxim Dounin.
2125 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
2126 Patch provided by Phil Pennock.
2128 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
2130 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
2131 Patch provided by Brad "anomie" Jorsch.
2133 NM/14 Bugzilla 622: Added timeout setting to transport filter.
2134 Patch provided by Dean Brooks.
2136 TK/05 Add native DKIM support (does not depend on external libraries).
2138 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
2139 Patch provided by Graeme Fowler.
2141 NM/16 Bugzilla 851: Documentation example syntax fix.
2143 NM/17 Changed NOTICE file to remove references to embedded PCRE.
2145 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
2148 NM/19 Bugzilla 745: TLS version reporting.
2149 Patch provided by Phil Pennock.
2151 NM/20 Bugzilla 167: bool: condition support.
2152 Patch provided by Phil Pennock.
2154 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
2155 clients. Patch provided by Phil Pennock.
2157 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
2158 Patch provided by Brad "anomie" Jorsch.
2160 NM/23 Bugzilla 687: Fix misparses in eximstats.
2161 Patch provided by Heiko Schlittermann.
2163 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
2164 Patch provided by Heiko Schlittermann.
2166 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
2167 plus update to original patch.
2169 NM/26 Bugzilla 799: Documentation correction for ratelimit.
2171 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
2172 Patch provided by David Brownlee.
2174 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
2176 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
2178 NM/30 Bugzilla 888: TLS documentation bugfixes.
2180 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
2182 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
2183 Unlike the original bugzilla I have changed all shell scripts in src tree.
2185 NM/33 Bugzilla 898: Transport filter timeout fix.
2186 Patch by Todd Rinaldo.
2188 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
2189 Patch by Serge Demonchaux.
2191 NM/35 Bugzilla 39: Base64 decode bug fixes.
2192 Patch by Jakob Hirsch.
2194 NM/36 Bugzilla 909: Correct connect() call in dcc code.
2196 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
2198 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
2200 NM/39 Bugzilla 911: Fixed MakeLinks build script.
2206 TK/01 Add preliminary DKIM support. Currently requires a forked version of
2207 ALT-N's libdkim that I have put here:
2208 http://duncanthrax.net/exim-experimental/
2210 Note to Michael Haardt: I had to rename some vars in sieve.c. They
2211 were called 'true' and it seems that C99 defines that as a reserved
2212 keyword to be used with 'bool' variable types. That means you could
2213 not include C99-style headers which use bools without triggering
2214 build errors in sieve.c.
2216 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
2217 as mailq or other aliases. Changed the --help handling significantly
2218 to do whats expected. exim_usage() emits usage/help information.
2220 SC/01 Added the -bylocaldomain option to eximstats.
2222 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
2224 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
2226 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
2232 PH/01 Another patch from the Sieve maintainer.
2234 PH/02 When an IPv6 address is converted to a string for single-key lookup
2235 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
2236 dots are used instead of colons so that keys in lsearch files need not
2237 contain colons. This was done some time before quoting was made available
2238 in lsearch files. However, iplsearch files do require colons in IPv6 keys
2239 (notated using the quote facility) so as to distinguish them from IPv4
2240 keys. This meant that lookups for IP addresses in host lists did not work
2241 for iplsearch lookups.
2243 This has been fixed by arranging for IPv6 addresses to be expressed with
2244 colons if the lookup type is iplsearch. This is not incompatible, because
2245 previously such lookups could never work.
2247 The situation is now rather anomalous, since one *can* have colons in
2248 ordinary lsearch keys. However, making the change in all cases is
2249 incompatible and would probably break a number of configurations.
2251 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
2254 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
2255 conversion specification without a maximum field width, thereby enabling
2256 a rogue spamd server to cause a buffer overflow. While nobody in their
2257 right mind would setup Exim to query an untrusted spamd server, an
2258 attacker that gains access to a server running spamd could potentially
2259 exploit this vulnerability to run arbitrary code as the Exim user.
2261 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
2262 $primary_hostname instead of what libspf2 thinks the hosts name is.
2264 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
2265 a directory entry by the name of the lookup key. Previously, if a
2266 symlink pointed to a non-existing file or a file in a directory that
2267 Exim lacked permissions to read, a lookup for a key matching that
2268 symlink would fail. Now it is enough that a matching directory entry
2269 exists, symlink or not. (Bugzilla 503.)
2271 PH/03 The body_linecount and body_zerocount variables are now exported in the
2274 PH/04 Added the $dnslist_matched variable.
2276 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
2277 This means they are set thereafter only if the connection becomes
2280 PH/06 Added the client_condition to authenticators so that some can be skipped
2281 by clients under certain conditions.
2283 PH/07 The error message for a badly-placed control=no_multiline_responses left
2284 "_responses" off the end of the name.
2286 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
2288 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
2289 (without spaces) instead of just copying the configuration text.
2291 PH/10 Added the /noupdate option to the ratelimit ACL condition.
2293 PH/11 Added $max_received_linelength.
2295 PH/12 Added +ignore_defer and +include_defer to host lists.
2297 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
2298 way in which PCRE > 7.0 is built.
2300 PH/14 Implemented queue_only_load_latch.
2302 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
2303 MAIL command. The effect was to mangle the value on 64-bit systems.
2305 PH/16 Another patch from the Sieve maintainer.
2307 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
2309 PH/18 If a system quota error occurred while trying to create the file for
2310 a maildir delivery, the message "Mailbox is full" was not appended to the
2311 bounce if the delivery eventually timed out. Change 4.67/27 below applied
2312 only to a quota excession during the actual writing of the file.
2314 PH/19 It seems that peer DN values may contain newlines (and other non-printing
2315 characters?) which causes problems in log lines. The DN values are now
2316 passed through string_printing() before being added to log lines.
2318 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
2319 and InterBase are left for another time.)
2321 PH/21 Added message_body_newlines option.
2323 PH/22 Guard against possible overflow in moan_check_errorcopy().
2325 PH/23 POSIX allows open() to be a macro; guard against that.
2327 PH/24 If the recipient of an error message contained an @ in the local part
2328 (suitably quoted, of course), incorrect values were put in $domain and
2329 $local_part during the evaluation of errors_copy.
2335 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
2336 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
2339 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
2340 issue a MAIL command.
2342 PH/02 In an ACL statement such as
2344 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
2346 if a client was not listed at all, or was listed with a value other than
2347 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
2348 the condition was not true (as it should be), so access was not denied.
2349 The bug was that the ! inversion was incorrectly passed on to the second
2350 item. This has been fixed.
2352 PH/03 Added additional dnslists conditions == and =& which are different from
2353 = and & when the dns lookup returns more than one IP address.
2355 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
2356 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
2358 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
2359 FSYNC, which compiles an option called disable_fsync that allows for
2360 bypassing fsync(). The documentation is heavily laced with warnings.
2362 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
2364 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
2365 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
2366 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
2367 including adding "make clean"; (3) Added -fPIC when compiling the test
2368 dynamically loaded module, to get rid of a warning.
2370 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
2371 message fails, move_frozen_messages = true and ignore_bounce_errors_after
2372 = 0s. The bug is otherwise harmless.
2374 PH/07 There was a bug in the dovecot authenticator such that the value of
2375 $auth1 could be overwritten, and so not correctly preserved, after a
2376 successful authentication. This usually meant that the value preserved by
2377 the server_setid option was incorrect.
2379 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
2381 PH/09 Installed PCRE release 7.0.
2383 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
2384 run for batched SMTP input. It is now run at the start of every message
2385 in the batch. While fixing this I discovered that the process information
2386 (output by running exiwhat) was not always getting set for -bs and -bS
2387 input. This is fixed, and it now also says "batched" for BSMTP.
2389 PH/11 Added control=no_pipelining.
2391 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
2392 patch, slightly modified), and move the expansion of helo_data till after
2393 the connection is made in the smtp transport (so it can use these
2396 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
2398 PH/14 Added log_selector = +pid.
2400 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
2402 PH/16 Add ${if forany and ${if forall.
2404 PH/17 Added dsn_from option to vary the From: line in DSNs.
2406 PH/18 Flush SMTP output before performing a callout, unless control =
2407 no_callout_flush is set.
2409 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
2410 was true (the default) a successful delivery failed to delete the retry
2411 item, thus causing premature timeout of the address. The bug is now
2414 PH/20 Added hosts_avoid_pipelining to the smtp transport.
2416 PH/21 Long custom messages for fakedefer and fakereject are now split up
2417 into multiline responses in the same way that messages for "deny" and
2418 other ACL rejections are.
2420 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
2421 with slight modification.
2423 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
2424 draft, changing the syntax and factoring some duplicate code".
2426 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
2427 for deliveries of the second and subsequent messages over the same SMTP
2430 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
2431 ${reduce, with only minor "tidies".
2433 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
2435 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
2436 expansion side effects.
2438 PH/27 When a message times out after an over-quota error from an Exim-imposed
2439 quota, the bounce message says "mailbox is full". This message was not
2440 being given when it was a system quota that was exceeded. It now should
2443 MH/03 Made $recipients available in local_scan(). local_scan() already has
2444 better access to the recipient list through recipients_list[], but
2445 $recipients can be useful in postmaster-provided expansion strings.
2447 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
2448 in the case of a MAIL command with additional options following the
2449 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
2450 were accidentally chopped off.
2452 PH/29 SMTP synchronization checks are implemented when a command is read -
2453 there is a check that no more input is waiting when there shouldn't be
2454 any. However, for some commands, a delay in an ACL can mean that it is
2455 some time before the response is written. In this time, more input might
2456 arrive, invalidly. So now there are extra checks after an ACL has run for
2457 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
2458 pipelining has not been advertised.
2460 PH/30 MH's patch to allow iscntrl() characters to be list separators.
2462 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
2463 returned in the SMTP response when smtp_return_error_details was false.
2464 This has been fixed.
2466 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
2467 instead of the C I/O that was originally supplied, because problems were
2468 reported on Solaris.
2470 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
2471 Exim which did not show up earlier: it was assuming that a call to
2472 SSL_CTX_set_info_callback() might give an error value. In fact, there is
2473 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
2474 was a macro that became an assignment, so it seemed to work. This has
2475 changed to a proper function call with a void return, hence the compile
2476 error. Exim's code has been fixed.
2478 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
2481 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
2483 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
2485 PH/37 If a message is not accepted after it has had an id assigned (e.g.
2486 because it turns out to be too big or there is a timeout) there is no
2487 "Completed" line in the log. When some messages of this type were
2488 selected by exigrep, they were listed as "not completed". Others were
2489 picked up by some special patterns. I have improved the selection
2490 criteria to be more general.
2492 PH/38 The host_find_failed option in the manualroute router can now be set
2493 to "ignore", to completely ignore a host whose IP address cannot be
2494 found. If all hosts are ignored, the behaviour is controlled by the new
2495 host_all_ignored option.
2497 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
2498 homing or because of multiple MX records with /mx) generated more than
2499 one IP address, and the following item turned out to be the local host,
2500 all the secondary addresses of the first item were incorrectly removed
2501 from the list, along with the local host and any following hosts (which
2502 is what is supposed to happen).
2504 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
2505 whoever called Exim into the -H file. In the case of the daemon it was
2506 behaving confusingly. When first started, it used values for whoever
2507 started the daemon, but after a SIGHUP it used the Exim user (because it
2508 calls itself on a restart). I have changed the code so that it now always
2511 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
2512 message are rejected with the same error (e.g. no authentication or bad
2513 sender address), and a DATA command is nevertheless sent (as can happen
2514 with PIPELINING or a stupid MUA), the error message that was given to the
2515 RCPT commands is included in the rejection of the DATA command. This is
2516 intended to be helpful for MUAs that show only the final error to their
2519 PH/42 Another patch from the Sieve maintainer.
2521 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
2522 Eximstats - Fixed some broken HTML links and added missing column headers
2524 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
2525 columns for Rejects, Temp Rejects, Ham, and Spam rows.
2527 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
2529 PH/43 Yet another patch from the Sieve maintainer.
2531 PH/44 I found a way to check for a TCP/IP connection going away before sending
2532 the response to the final '.' that terminates a message, but only in the
2533 case where the client has not sent further data following the '.'
2534 (unfortunately, this is allowed). However, in many cases there won't be
2535 any further data because there won't be any more messages to send. A call
2536 to select() can be used: if it shows that the input is "ready", there is
2537 either input waiting, or the socket has been closed. An attempt to read
2538 the next input character can distinguish the two cases. Previously, Exim
2539 would have sent an OK response which the client would never have see.
2540 This could lead to message repetition. This fix should cure that, at
2541 least in a lot of common cases.
2543 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
2544 advertised in response to EHLO.
2550 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
2551 fixed by 4.65/MH/01 (is this a record?) are fixed:
2553 (i) An empty string was always treated as zero by the numeric comparison
2554 operators. This behaviour has been restored.
2556 (ii) It is documented that the numeric comparison operators always treat
2557 their arguments as decimal numbers. This was broken in that numbers
2558 starting with 0 were being interpreted as octal.
2560 While fixing these problems I realized that there was another issue that
2561 hadn't been noticed. Values of message_size_limit (both the global option
2562 and the transport option) were treated as octal if they started with 0.
2563 The documentation was vague. These values are now always treated as
2564 decimal, and I will make that clear in the documentation.
2570 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
2571 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
2574 MH/01 Don't check that the operands of numeric comparison operators are
2575 integers when their expansion is in "skipping" mode (fixes bug
2576 introduced by 4.64-PH/07).
2578 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
2579 child addresses, Exim now panics and dies. Previously, because the count
2580 is held in a short int, deliveries were likely to be lost. As such a
2581 large number of recipients for a single message is ridiculous
2582 (performance will be very, very poor), I have chosen to impose a limit
2583 rather than extend the field.
2589 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
2590 leftover -K file (the existence of which was triggered by #402).
2591 While we were at it, introduced process PID as part of the -K
2592 filename. This should rule out race conditions when creating
2595 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
2596 processing considerably. Previous code took too long for large mails,
2597 triggering a timeout which in turn triggers #401.
2599 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
2600 in the DK code in transports.c. sendfile() is not really portable,
2601 hence the _LINUX specificness.
2603 TF/01 In the add_headers option to the mail command in an Exim filter,
2604 there was a bug that Exim would claim a syntax error in any
2605 header after the first one which had an odd number of characters
2608 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
2609 callout verification, Exim cached a "reject" for the entire domain. This
2610 is correct for most verifications, but it is not correct for a recipient
2611 verification with use_sender or use_postmaster set, because in that case
2612 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
2613 case of MAIL FROM:<> rejection from other early rejections (e.g.
2614 rejection of HELO). When verifying a recipient using a non-null MAIL
2615 address, the cache is ignored if it shows MAIL FROM:<> rejection.
2616 Whatever the result of the callout, the value of the domain cache is
2617 left unchanged (for any other kind of callout, getting as far as trying
2618 RCPT means that the domain itself is ok).
2620 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
2623 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
2624 manifest itself as EPIPE rather than ECONNECT. When tidying away a
2625 session, the daemon ignores ECONNECT errors and logs others; it now
2626 ignores EPIPE as well.
2628 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
2629 (quoted-printable decoding).
2631 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
2632 later the small subsequent patch to fix an introduced bug.
2634 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
2636 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
2638 PH/08 An error is now given if message_size_limit is specified negative.
2640 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
2641 to be given (somewhat) arbitrary names.
2643 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
2646 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
2647 miscellaneous code fixes
2649 PH/10 Added the log_reject_target ACL modifier to specify where to log
2652 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
2653 hostname. This is wrong, because it relates to the incoming message (and
2654 probably the interface on which it is arriving) and not to the outgoing
2655 callout (which could be using a different interface). This has been
2656 changed to use the value of the helo_data option from the smtp transport
2657 instead - this is what is used when a message is actually being sent. If
2658 there is no remote transport (possible with a router that sets up host
2659 addresses), $smtp_active_hostname is used.
2661 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
2662 tweaks were necessary in order to get it to work (see also 21 below):
2663 (a) The code assumed that strncpy() returns a negative number on buffer
2664 overflow, which isn't the case. Replaced with Exim's string_format()
2666 (b) There were several signed/unsigned issues. I just did the minimum
2667 hacking in of casts. There is scope for a larger refactoring.
2668 (c) The code used strcasecmp() which is not a standard C function.
2669 Replaced with Exim's strcmpic() function.
2670 (d) The code set only $1; it now sets $auth1 as well.
2671 (e) A simple test gave the error "authentication client didn't specify
2672 service in request". It would seem that Dovecot has changed its
2673 interface. Fortunately there's a specification; I followed it and
2674 changed what the client sends and it appears to be working now.
2676 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
2679 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
2680 address A is aliased to B and C, where B exists and C does not. Without
2681 -v the output is "A verified" because verification stops after a
2682 successful redirection if more than one address is generated. However,
2683 with -v the child addresses are also verified. Exim was outputting "A
2684 failed to verify" and then showing the successful verification for C,
2685 with its parentage. It now outputs "B failed to verify", showing B's
2686 parentage before showing the successful verification of C.
2688 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
2689 look up a TXT record in a specific list after matching in a combined
2692 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
2693 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
2694 they consult the DNS. I had assumed they would set it the way they
2695 wanted; and indeed my experiments on Linux seem to show that in some
2696 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
2697 To be on the safe side, however, I have now made the interface to
2698 host_find_byname() similar to host_find_bydns(), with an argument
2699 containing the DNS resolver options. The host_find_byname() function now
2700 sets these options at its start, just as host_find_bydns() does. The smtp
2701 transport options dns_qualify_single and dns_search_parents are passed to
2702 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
2703 of host_find_byname() use the default settings of RES_DEFNAMES
2704 (qualify_single) but not RES_DNSRCH (search_parents).
2706 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
2707 spool_read_header() do less string testing, by means of a preliminary
2708 switch on the second character of optional "-foo" lines. (This is
2709 overdue, caused by the large number of possibilities that now exist.
2710 Originally there were few.) While I was there, I also converted the
2711 str(n)cmp tests so they don't re-test the leading "-" and the first
2712 character, in the hope this might squeeze out yet more improvement.
2714 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
2715 flag allowing group syntax was set by the header_syntax check but not
2716 turned off, possible causing trouble later; (2) The flag was not being
2717 set at all for the header_verify test, causing "group"-style headers to
2718 be rejected. I have now set it in this case, and also caused header_
2719 verify to ignore an empty address taken from a group. While doing this, I
2720 came across some other cases where the code for allowing group syntax
2721 while scanning a header line wasn't quite right (mostly, not resetting
2722 the flag correctly in the right place). These bugs could have caused
2723 trouble for malformed header lines. I hope it is now all correct.
2725 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
2726 with the "reply" argument non-NULL. The code, however (which originally
2727 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
2728 but it didn't always do it. This confused somebody who was copying the
2729 code for some other use. I have removed all the tests.
2731 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
2732 feature that was used to support insecure browsers during the U.S. crypto
2733 embargo. It requires special client support, and Exim is probably the
2734 only MTA that supported it -- and would never use it because real RSA is
2735 always available. This code has been removed, because it had the bad
2736 effect of slowing Exim down by computing (never used) parameters for the
2737 RSA_EXPORT functionality.
2739 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
2740 authenticator to fail if there's a tab character in the incoming data
2741 (there should never be unless someone is messing about, as it's supposed
2742 to be base64-encoded). Also added, on Timo's advice, the "secured" option
2743 if the connection is using TLS or if the remote IP is the same as the
2744 local IP, and the "valid-client-cert option" if a client certificate has
2747 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
2748 authenticators. This can be used for authorization after authentication
2749 succeeds. (In the case of plaintext, it servers for both authentication
2752 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
2753 if any retry times were supplied.
2755 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
2756 connection, where there is no client IP address to check. In this
2757 situation, the verify now always succeeds.
2759 PH/25 Applied John Jetmore's -Mset patch.
2761 PH/26 Added -bem to be like -Mset, but loading a message from a file.
2763 PH/27 In a string expansion for a processed (not raw) header when multiple
2764 headers of the same name were present, leading whitespace was being
2765 removed from all of them, but trailing whitespace was being removed only
2766 from the last one. Now trailing whitespace is removed from each header
2767 before concatenation. Completely empty headers in a concatenation (as
2768 before) are ignored.
2770 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
2771 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
2773 PH/29 [Removed. This was a change that I later backed out, and forgot to
2774 correct the ChangeLog entry (that I had efficiently created) before
2775 committing the later change.]
2777 PH/30 Exim was sometimes attempting to deliver messages that had suffered
2778 address errors (4xx response to RCPT) over the same connection as other
2779 messages routed to the same hosts. Such deliveries are always "forced",
2780 so retry times are not inspected. This resulted in far too many retries
2781 for the affected addresses. The effect occurred only when there were more
2782 hosts than the hosts_max_try setting in the smtp transport when it had
2783 the 4xx errors. Those hosts that it had tried were not added to the list
2784 of hosts for which the message was waiting, so if all were tried, there
2785 was no problem. Two fixes have been applied:
2787 (i) If there are any address or message errors in an SMTP delivery, none
2788 of the hosts (tried or untried) are now added to the list of hosts
2789 for which the message is waiting, so the message should not be a
2790 candidate for sending over the same connection that was used for a
2791 successful delivery of some other message. This seems entirely
2792 reasonable: after all the message is NOT "waiting for some host".
2793 This is so "obvious" that I'm not sure why it wasn't done
2794 previously. Hope I haven't missed anything, but it can't do any
2795 harm, as the worst effect is to miss an optimization.
2797 (ii) If, despite (i), such a delivery is accidentally attempted, the
2798 routing retry time is respected, so at least it doesn't keep
2799 hammering the server.
2801 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
2802 in ${readsocket because some servers need this prod.
2804 PH/32 Added some extra debug output when updating a wait-xxx database.
2806 PH/33 The hint "could be header name not terminated by colon", which has been
2807 given for certain expansion errors for a long time, was not being given
2808 for the ${if def:h_colon_omitted{... case.
2810 PH/34 The spec says: "With one important exception, whenever a domain list is
2811 being scanned, $domain contains the subject domain." There was at least
2812 one case where this was not true.
2814 PH/35 The error "getsockname() failed: connection reset by peer" was being
2815 written to the panic log as well as the main log, but it isn't really
2816 panic-worthy as it just means the connection died rather early on. I have
2817 removed the panic log writing for the ECONNRESET error when getsockname()
2820 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
2821 runs only) independently of the message's sender address. This meant
2822 that, if the 4xx error was in fact related to the sender, a different
2823 message to the same recipient with a different sender could confuse
2824 things. In particular, this can happen when sending to a greylisting
2825 server, but other circumstances could also provoke similar problems.
2826 I have changed the default so that the retry time for these errors is now
2827 based a combination of the sender and recipient addresses. This change
2828 can be overridden by setting address_retry_include_sender=false in the
2831 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
2832 remote server are returned as part of bounce messages. This was not
2833 happening for LMTP over a pipe (the lmtp transport), but now it is the
2834 same for both kinds of LMTP.
2836 PH/38 Despite being documented as not happening, Exim was rewriting addresses
2837 in header lines that were in fact CNAMEs. This is no longer the case.
2839 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
2840 and queue runs started by the daemon processed all messages. This has
2841 been fixed so that -R and -S can now usefully be given with -q<time>.
2843 PH/40 Import PCRE release 6.7 (fixes some bugs).
2845 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
2847 PH/42 Give an error if -q is specified more than once.
2849 PH/43 Renamed the variables $interface_address and $interface_port as
2850 $received_ip_address and $received_port, to make it clear that these
2851 values apply to message reception, and not to the outgoing interface when
2852 a message is delivered. (The old names remain recognized, of course.)
2854 PH/44 There was no timeout on the connect() call when using a Unix domain
2855 socket in the ${readsocket expansion. There now is.
2857 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
2858 be meaningful with "accept".
2860 SC/01 Eximstats V1.43
2861 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
2863 SC/02 Eximstats V1.44
2864 Use a glob alias rather than an array ref in the generated
2865 parser. This improves both readability and performance.
2867 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
2868 Collect SpamAssassin and rejection statistics.
2869 Don't display local sender or destination tables unless
2870 there is data to show.
2871 Added average volumes into the top table text output.
2873 SC/04 Eximstats V1.46
2874 Collect data on the number of addresses (recipients)
2875 as well as the number of messages.
2877 SC/05 Eximstats V1.47
2878 Added 'Message too big' to the list of mail rejection
2879 reasons (thanks to Marco Gaiarin).
2881 SC/06 Eximstats V1.48
2882 Mainlog lines which have GMT offsets and are too short to
2883 have a flag are now skipped.
2885 SC/07 Eximstats V1.49 (Alain Williams)
2886 Added the -emptyok flag.
2888 SC/08 Eximstats V1.50
2889 Fixes for obtaining the IP address from reject messages.
2891 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
2892 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
2893 whitespace changes from 4.64-PH/27
2895 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
2898 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
2899 are found, allow negative numbers in numeric criteria)
2901 JJ/06 exipick.20061117.2, added new $message_body_missing variable
2903 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
2904 to match changes made in 4.64-PH/43
2906 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
2908 PH/47 Put in an explicit test for a DNS lookup of an address record where the
2909 "domain" is actually an IP address, and force a failure. This locks out
2910 those revolvers/nameservers that support "A-for-A" lookups, in
2911 contravention of the specifications.
2913 PH/48 When a host name was looked up from an IP address, and the subsequent
2914 forward lookup of the name timed out, the host name was left in
2915 $sender_host_name, contrary to the specification.
2917 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
2918 restricted to single-key lookups, Exim was not diagnosing an error if
2919 * or *@ was used with a query-style lookup.
2921 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
2923 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
2924 long ago, but noone interested enough thought of it. Let's just say that
2925 the "1.1" means that there are some new functions that weren't there at
2926 some point in the past.
2928 PH/51 Error processing for expansion failure of helo_data from an smtp
2929 transport during callout processing was broken.
2931 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
2932 tested/used via the -bh/-bhc/-bs options.
2934 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
2935 bug, fixed in subsequent PCRE releases).
2937 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
2938 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
2940 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
2946 SC/01 Use a glob alias rather than an array ref in eximstats generated
2947 parser. This improves both readability and performance.
2949 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
2950 Don't display local sender or destination tables in eximstats unless
2951 there is data to show.
2952 Added average volumes into the eximstats top table text output.
2954 SC/03 Collect data on the number of addresses (recipients) as well
2955 as the number of messages in eximstats.
2957 TF/01 Correct an error in the documentation for the redirect router. Exim
2958 does (usually) call initgroups() when daemonizing.
2960 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
2961 with consistent privilege compared to when running as a daemon.
2963 TF/03 Note in the spec that $authenticated_id is not set for local
2964 submissions from trusted users.
2966 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
2967 Thanks to Dean Brooks <dean@iglou.com> for the patch.
2969 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
2970 by adding some example configuration directives to the default
2971 configuration file. A little bit of work is required to uncomment the
2972 directives and define how usernames and passwords are checked, but
2973 there is now a framework to start from.
2975 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
2976 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
2977 without this. I don't know how relevant this is to other LDAP libraries.
2979 PH/02 Add the verb name to the "unknown ACL verb" error.
2981 PH/03 Magnus Holmgren's patch for filter_prepend_home.
2983 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
2985 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
2986 directory not expanded when it should be if an expanded home directory
2987 was set for the address (which is overridden by the transport).
2989 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
2992 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
2993 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
2994 because it is too late at that time, and has no effect.
2996 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
2997 security issue with \' (bugzilla #107). I could not use the
2998 PQescapeStringConn() function, because it needs a PGconn value as one of
3001 PH/08 When testing addresses using -bt, indicate those final addresses that
3002 are duplicates that would not cause an additional delivery. At least one
3003 person was confused, thinking that -bt output corresponded to deliveries.
3004 (Suppressing duplicates isn't a good idea as you lose the information
3005 about possibly different redirections that led to the duplicates.)
3007 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
3008 systems where poll() doesn't work, in particular OS X.
3010 PH/10 Added more information to debugging output for retry time not reached.
3012 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
3013 operations in malware.c.
3015 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
3018 PH/13 If write_rejectlog was set false when logging was sent to syslog with
3019 syslog_duplication set false, log lines that would normally be written
3020 both the the main log and to the reject log were not written to syslog at
3023 PH/14 In the default configuration, change the use of "message" in ACL warn
3024 statements to "add_header".
3026 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
3027 not followed by a command (e.g. "seen endif").
3029 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
3030 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
3033 PH/17 Added extra conditions to the default value of delay_warning_condition
3037 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
3038 { match{$h_precedence:}{(?i)bulk|list|junk} } \
3039 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
3042 The Auto-Submitted: and various List- headers are standardised, whereas I
3043 don't think Precedence: ever was.
3045 PH/18 Refactored debugging code in route_finduser() to show more information,
3046 in particular, the error code if getpwnam() issues one.
3048 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
3049 This is apparently needed in addition to the PH/07 change above to avoid
3050 any possible encoding problems.
3052 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
3053 but not after initializing Perl.
3055 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
3056 output them only if debugging. By default they are written stderr,
3057 apparently, which is not desirable.
3059 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
3062 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
3065 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
3067 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
3068 authenticated or an ident call has been made. Suppress the default
3069 values for $authenticated_id and $authenticated_sender (but permit -oMai
3070 and -oMas) when testing with -bh.
3072 PH/24 Re-jigged the order of the tests in the default configuration so that the
3073 tests for valid domains and recipients precede the DNS black list and CSA
3074 tests, on the grounds that those ones are more expensive.
3076 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
3077 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
3078 This bug exists in every version of Exim that I still have, right back to
3081 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
3082 However, an attempt to turn on case-sensitivity in a regex key by
3083 including (?-i) didn't work because the subject string was already
3084 lowercased, and the effects were non-intuitive. It turns out that a
3085 one-line patch can be used to allow (?-i) to work as expected.
3091 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
3092 other effects) broke the use of negated acl sub-conditions.
3094 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
3097 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
3098 "Deny" causes Exim to reject the incoming connection with a 554 error.
3099 Unfortunately, if there is a major crisis, such as a disk failure,
3100 tcp-wrappers gives "deny", whereas what one would like would be some
3101 kind of temporary error. A kludge has been added to help with this.
3102 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
3103 554 error is used if errno is still zero or contains ENOENT (which occurs
3104 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
3107 PH/03 Add -lutil to the default FreeBSD LIBS setting.
3109 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
3110 errors. Otherwise a message that provokes a temporary error (when other
3111 messages do not) can cause a whole host to time out.
3113 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
3114 addresses were routed directly to files or pipes from a redirect router.
3115 File deliveries just didn't batch; pipe deliveries might have suffered
3118 PH/06 A failure to get a lock for a hints database would erroneously always say
3119 "Failed to get write lock", even when it was really a read lock.
3121 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
3122 of 0600. This has been changed to use the value of the lockfile_mode
3123 option (which defaults to 0600).
3125 PH/08 Applied small patch from the Sieve maintainer.
3127 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
3128 folder from quota calculations, a direct delivery into this folder messed
3129 up the contents of the maildirsize file. This was because the regex was
3130 used only to exclude .Trash (or whatever) when the size of the mailbox
3131 was calculated. There was no check that a delivery was happening into an
3132 excluded directory. This bug has been fixed by ignoring all quota
3133 processing for deliveries into excluded directories.
3135 PH/10 Added the maildirfolder_create_regex option to appendfile.
3141 PH/01 The code for finding all the local interface addresses on a FreeBSD
3142 system running IPv6 was broken. This may well have applied to all BSD
3143 systems, as well as to others that have similar system calls. The broken
3144 code found IPv4 interfaces correctly, but gave incorrect values for the
3145 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
3146 that it would not match correctly against @[] and not recognize the IPv6
3149 PH/02 The ipliteral router was not recognizing addresses of the form user@
3150 [ipv6:....] because it didn't know about the "ipv6:" prefix.
3152 PH/03 Added disable_ipv6.
3154 PH/04 Changed $reply_address to use the raw form of the headers instead of the
3155 decoded form, because it is most often used to construct To: headers
3156 lines in autoreplies, and the decoded form may well be syntactically
3157 invalid. However, $reply_address has leading white space removed, and all
3158 newlines turned into spaces so that the autoreply transport does not
3161 PH/05 If group was specified without a user on a router, and no group or user
3162 was specified on a transport, the group from the router was ignored.
3164 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
3165 for visible compile-time settings that can be used to change these
3166 numbers, for those that want even more. Backwards compatibility with old
3167 spool files has been maintained. However, going back to a previous Exim
3168 release will lost any variables that are in spool files.
3170 PH/07 Two small changes when running in the test harness: increase delay when
3171 passing a TCP/IP connection to a new process, in case the original
3172 process has to generate a bounce, and remove special handling of
3173 127.0.0.2 (sic), which is no longer necessary.
3175 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
3176 be the same on different OS.
3178 PH/09 Moved a debug statement in filter processing to avoid a race problem when
3181 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
3182 whether --show-vars was specified or not
3184 JJ/02 exipick: Added support for new ACL variable spool format introduced
3187 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
3188 syntactically invalid From: or Reply-to: line, and a filter used this to
3189 generate an autoreply, and therefore failed to obtain an address for the
3190 autoreply, Exim could try to deliver to a non-existent relative file
3191 name, causing unrelated and misleading errors. What now happens is that
3192 it logs this as a hard delivery error, but does not attempt to create a
3195 PH/11 The exinext utility has a -C option for testing purposes, but although
3196 the given file was scanned by exinext itself; it wasn't being passed on
3197 when Exim was called.
3199 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
3200 an end-of-file indication when reading a command response.
3202 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
3203 compiled. In many other places in Exim, IPv6 addresses are always
3204 recognized, so I have changed this. It also means that IPv4 domain
3205 literals of the form [IPV4:n.n.n.n] are now always recognized.
3207 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
3208 used if the router is not running as root, for example, when verifying at
3209 ACL time, or when using -bh. The debugging output from this situation was
3210 non-existent - all you got was a failure to exec. I have made two
3213 (a) Failures to set uid/gid, the current directory, or a process leader
3214 in a subprocess such as that created by queryprogram now generate
3215 suitable debugging output when -d is set.
3217 (b) The queryprogram router detects when it is not running as root,
3218 outputs suitable debugging information if -d is set, and then runs
3219 the subprocess without attempting to change uid/gid.
3221 PH/15 Minor change to Makefile for building test_host (undocumented testing
3224 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
3225 additional section of a DNS packet that returns MX or SRV records.
3226 Instead, it always explicitly searches for A/AAAA records. This avoids
3227 major problems that occur when a DNS server includes only records of one
3228 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
3229 fixed another bug: if SRV records were looked up and the corresponding
3230 address records were *not* found in the additional section, the port
3231 values from the SRV records were lost.
3233 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
3234 using the correct key (the original address) when searching the retry
3235 rules in order to find which one to use for generating the retry hint.
3237 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
3238 adding the default one. Similarly, if it contains a Reply-To: header, the
3239 errors_reply_to option, if set, is not used.
3241 PH/19 When calculating a retry time, Exim used to measure the "time since
3242 failure" by looking at the "first failed" field in the retry record. Now
3243 it does not use this if it is later than than the arrival time of the
3244 message. Instead it uses the arrival time. This makes for better
3245 behaviour in cases where some deliveries succeed, thus re-setting the
3246 "first failed" field. An example is a quota failure for a huge message
3247 when small messages continue to be delivered. Without this change, the
3248 "time since failure" will always be short, possible causing more frequent
3249 delivery attempts for the huge message than are intended.
3250 [Note: This change was subsequently modified - see PH/04 for 4.62.]
3252 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
3253 $1, $2, $3) because the numerical variables can be reset during some
3254 expansion items (e.g. "match"), thereby losing the authentication data.
3256 PH/21 Make -bV show the size of off_t variables so that the test suite can
3257 decide whether to run tests for quotas > 2G.
3259 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
3260 mailbox_size, and mailbox_filecount in the appendfile transport. If a
3261 filecount value is greater than 2G or if a quota value is greater than 2G
3262 on a system where the size of off_t is not greater than 4, a panic error
3265 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
3266 never match. The debug and -bh output now contains an explicit error
3267 message indicating a malformed IPv4 address or mask.
3269 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
3270 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
3271 PH/23 above applies.
3273 PH/25 Do not write to syslog when running in the test harness. The only
3274 occasion when this arises is a failure to open the main or panic logs
3275 (for which there is an explicit test).
3277 PH/26 Added the /no_tell option to "control=freeze".
3279 PH/27 If a host name lookup failed very early in a connection, for example, if
3280 the IP address matched host_lookup and the reverse lookup yielded a name
3281 that did not have a forward lookup, an error message of the form "no IP
3282 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
3283 could be logged. Now it outputs the IP address instead of "NULL".
3285 PH/28 An enabling patch from MH: add new function child_open_exim2() which
3286 allows the sender and the authenticated sender to be set when
3287 submitting a message from within Exim. Since child_open_exim() is
3288 documented for local_scan(), the new function should be too.
3290 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
3291 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
3292 results in an empty string is now treated as unset.
3294 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
3296 PH/31 Added sender_verify_fail logging option.
3298 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
3299 needed by RFC 822 but not by RFC 2822 was commented out. I have now
3300 tidied the source and removed it altogether.
3302 PH/33 When a queue run was abandoned because the load average was too high, a
3303 log line was always written; now it is written only if the queue_run log
3304 selector is set. In addition, the log line for abandonment now contains
3305 information about the queue run such as the pid. This is always present
3306 in "start" and "stop" lines but was omitted from the "abandon" line.
3308 PH/34 Omit spaces between a header name and the colon in the error message that
3309 is given when verify = headers_syntax fails (if there are lots of them,
3310 the message gets confusing).
3312 PH/35 Change the default for dns_check_names_pattern to allow slashes within
3313 names, as there are now some PTR records that contain slashes. This check
3314 is only to protect against broken name servers that fall over on strange
3315 characters, so the fact that it applies to all lookups doesn't matter.
3317 PH/36 Now that the new test suite is complete, we can remove some of the
3318 special code in Exim that was needed for the old test suite. For example,
3319 sorting DNS records because real resolvers return them in an arbitrary
3320 order. The new test suite's fake resolver always returns records in the
3323 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
3324 bounces) except when queue_only is set, to avoid logging races between
3325 the different processes.
3327 PH/38 Panic-die if .include specifies a non-absolute path.
3329 PH/39 A tweak to the "H" retry rule from its user.
3331 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
3332 a label. They prevented compilation on older perls.
3334 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
3335 a warning to be raised on newish perls.
3337 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
3338 on queue. Changes to match documented behaviour of showing count of
3339 messages matching specified criteria.
3341 PH/40 Changed the default ident timeout from 30s to 5s.
3343 PH/41 Added support for the use of login_cap features, on those BSD systems
3344 that have them, for controlling the resources used by pipe deliveries.
3346 PH/42 The content-scanning code uses fopen() to create files in which to put
3347 message data. Previously it was not paying any attention to the mode of
3348 the files. Exim runs with umask(0) because the rest of the code creates
3349 files with open(), and sets the required mode explicitly. Thus, these
3350 files were ending up world-writeable. This was not a big issue, because,
3351 being within the spool directory, they were not world-accessible. I have
3352 created a function called modefopen, which takes an additional mode
3353 argument. It sets umask(777), creates the file, chmods it to the required
3354 mode, then resets the umask. All the relevant calls to fopen() in the
3355 content scanning code have been changed to use this function.
3357 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
3358 to 24 hours. This avoids potential overflow problems when processing G
3359 and H retry rules. I suspect nobody ever tinkers with this value.
3361 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
3363 PH/45 When the plaintext authenticator is running as a client, the server's
3364 challenges are checked to ensure they are valid base64 strings. By
3365 default, the authentication attempt is cancelled if an invalid string is
3366 received. Setting client_ignore_invalid_base64 true ignores these errors.
3367 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
3368 they are received. Thus, the responses can be made to depend on the
3369 challenges. If an invalid string is ignored, an empty string is placed in
3372 PH/46 Messages that are created by the autoreply transport now contains a
3373 References: header, in accordance with RFCs 2822 and 3834.
3375 PH/47 Added authenticated_sender_force to the smtp transport.
3377 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
3379 PH/49 Installed latest patch from the Sieve maintainer.
3381 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
3382 was also set, the appendfile transport was unnecessarily scanning a
3383 directory of message files (e.g. for maildir delivery) to find the count
3384 of files (along with the size), even though it did not need this
3385 information. It now does the scan only if it needs to find either the
3386 size of the count of files.
3388 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
3390 PH/52 Two bugs concerned with error handling when the smtp transport is
3393 (i) Exim was not creating retry information for temporary errors given
3394 for individual recipients after the DATA command when the smtp transport
3395 was used in LMTP mode. This meant that they could be retried too
3396 frequently, and not timed out correctly.
3398 (ii) Exim was setting the flag that allows error details to be returned
3399 for LMTP errors on RCPT commands, but not for LMTP errors for individual
3400 recipients that were returned after the DATA command.
3402 PH/53 This is related to PH/52, but is more general: for any failing address,
3403 when detailed error information was permitted to be returned to the
3404 sender, but the error was temporary, then after the final timeout, only
3405 "retry timeout exceeded" was returned. Now it returns the full error as
3406 well as "retry timeout exceeded".
3408 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
3409 do this, and (what is worse) MTAs that accept it.
3411 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
3412 will now be deprecated.
3414 PH/56 New os.c-cygwin from the Cygwin maintainer.
3416 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
3417 formats (previously only available in exim formats via -bpr, -bpru,
3418 and -bpra. Now also available in native and exiqgrep formats)
3420 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
3421 with very large, slow to parse queues
3423 JJ/08 exipick: added ! as generic prefix to negate any criteria format
3425 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
3427 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
3428 responses to authentication challenges, though it was showing the
3429 challenges; (ii) I've removed the CR characters from the debug output for
3432 PH/58 Allow for the insertion of a newline as well as a space when a string
3433 is turned into more than one encoded-word during RFC 2047 encoding. The
3434 Sieve code now uses this.
3436 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
3437 data_4xx, lost_connection, tls_required.
3439 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
3440 message was being sent as an SMTP response.
3442 PH/61 Add -l and -k options to exicyclog.
3444 PH/62 When verifying, if an address was redirected to one new address, so that
3445 verification continued, and the new address failed or deferred after
3446 having set something in $address_data, the value of $address_data was not
3447 passed back to the ACL. This was different to the case when no
3448 redirection occurred. The value is now passed back in both cases.
3450 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
3451 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
3452 use login_cap.h, so on its own it isn't the distinguishing feature. The
3453 new name refers directly to the setclassresources() function.
3455 PH/65 Added configuration files for NetBSD3.
3457 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
3459 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
3460 is preferred over IPv4.
3462 PH/68 The bounce_return_message and bounce_return_body options were not being
3463 honoured for bounces generated during the reception of non-SMTP messages.
3464 In particular, this applied to messages rejected by the ACL. This bug has
3465 been fixed. However, if bounce_return_message is true and bounce_return_
3466 body is false, the headers that are returned for a non-SMTP message
3467 include only those that have been read before the error was detected.
3468 (In the case of an ACL rejection, they have all been read.)
3470 PH/69 The HTML version of the specification is now built in a directory called
3471 spec_html instead of spec.html, because the latter looks like a path with
3472 a MIME-type, and this confuses some software.
3474 PH/70 Catch two compiler warnings in sieve.c.
3476 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
3477 function verify_get_ident() calls ip_connect() to connect a socket, but
3478 if the "connect()" function timed out, ip_connect() used to close the
3479 socket. However, verify_get_ident() also closes the socket later, and in
3480 between Exim writes to the log, which may get opened at this point. When
3481 the socket was closed in ip_connect(), the log could get the same file
3482 descriptor number as the socket. This naturally causes chaos. The fix is
3483 not to close the socket in ip_connect(); the socket should be closed by
3484 the function that creates it. There was only one place in the code where
3485 this was missing, in the iplookup router, which I don't think anybody now
3486 uses, but I've fixed it anyway.
3488 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
3489 well as to direct DNS lookups. Otherwise the handling of names in host
3490 lists is inconsistent and therefore confusing.
3496 PH/01 Two changes to the default runtime configuration:
3498 (1) Move the checks for relay_from_hosts and authenticated clients from
3499 after to before the (commented out) DNS black list checks.
3501 (2) Add control=submission to the relay_from_hosts and authenticated
3502 clients checks, on the grounds that messages accepted by these
3503 statements are most likely to be submissions.
3505 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
3507 (1) Generate an error if the third argument for the ${prvs expansion is
3510 (2) Treat a missing third argument of ${prvscheck as if it were an empty
3513 (3) Reset the variables that are obtained from the first argument of
3514 ${prvscheck and used in the second argument before leaving the code,
3515 because their memory is reclaimed, so using them afterwards may do
3518 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
3519 one (it's much easier than Tom thought :-).
3521 (5) Because of (4), we can now allow for the use of $prvscheck_result
3522 inside the third argument.
3524 PH/03 For some reason, the default setting of PATH when running a command from
3525 a pipe transport was just "/usr/bin". I have changed it to
3528 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
3529 anything to be listed in the output from -bV.
3531 PH/05 When a filter generated an autoreply, the entire To: header line was
3532 quoted in the delivery log line, like this:
3534 => >A.N.Other <ano@some.domain> <original@ddress> ...
3536 This has been changed so that it extracts the operative address. There
3537 may be more than one such address. If so, they are comma-separated, like
3540 => >ano@some.domain,ona@other.domain <original@ddress> ...
3542 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
3543 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
3544 not being looked up in the rDNS to get a host name, Exim was showing the
3545 IP address twice in Received: lines, even though the IP addresses were
3546 identical. For example:
3548 Received: from [1.2.3.4] (helo=[1.2.3.4])
3550 However, if the real host name was known, it was omitting the HELO data
3551 if it matched the actual IP address. This has been tidied up so that it
3552 doesn't show the same IP address twice.
3554 PH/07 When both +timestamp and +memory debugging was on, the value given by
3555 $tod_xxx expansions could be wrong, because the tod_stamp() function was
3556 called by the debug printing, thereby overwriting the timestamp buffer.
3557 Debugging no longer uses the tod_stamp() function when +timestamp is set.
3559 PH/08 When the original message was included in an autoreply transport, it
3560 always said "this is a copy of the message, including all the headers",
3561 even if body_only or headers_only was set. It now gives an appropriate
3564 PH/09 Applied a patch from the Sieve maintainer which:
3566 o fixes some comments
3567 o adds the (disabled) notify extension core
3568 o adds some debug output for the result of if/elsif tests
3569 o points to the current vacation draft in the documentation
3570 and documents the missing references header update
3574 o fixes a bug in processing the envelope test (when testing
3575 multiple envelope elements, the last element determined the
3578 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
3579 Electronic Mail") by including:
3581 Auto-submitted: auto-generated
3583 in the messages that it generates (bounce messages and others, such as
3584 warnings). In the case of bounce messages for non-SMTP messages, there was
3585 also a typo: it was using "Auto_submitted" (underscore instead of
3586 hyphen). Since every message generated by Exim is necessarily in response
3587 to another message, thes have all been changed to:
3589 Auto-Submitted: auto-replied
3591 in accordance with these statements in the RFC:
3593 The auto-replied keyword:
3595 - SHOULD be used on messages sent in direct response to another
3596 message by an automatic process,
3598 - MUST NOT be used on manually-generated messages,
3600 - MAY be used on Delivery Status Notifications (DSNs) and Message
3601 Disposition Notifications (MDNs),
3603 - MUST NOT be used on messages generated by automatic or periodic
3604 processes, except for messages which are automatic responses to
3607 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
3608 to the default Received: header definition.
3610 PH/12 Added log selector acl_warn_skipped (default on).
3612 PH/13 After a successful wildlsearch lookup, discard the values of numeric
3613 variables because (a) they are in the wrong storage pool and (b) even if
3614 they were copied, it wouldn't work properly because of the caching.
3616 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
3617 checking when decoding. Apparently there are clients that generate
3618 overlong encoded strings. Why am I not surprised?
3620 PH/15 If the first argument of "${if match_address" was not empty, but did not
3621 contain an "@" character, Exim crashed. Now it writes a panic log message
3622 and treats the condition as false.
3624 PH/16 In autoreply, treat an empty string for "once" the same as unset.
3626 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
3627 extension "envelope-auth". The code is finished and in agreement with
3628 other implementations, but there is no documentation so far and in fact,
3629 nobody wrote the draft yet. This extension is currently #undef'ed, thus
3630 not changing the active code.
3632 Print executed "if" and "elsif" statements when debugging is used. This
3633 helps a great deal to understand what a filter does.
3635 Document more things not specified clearly in RFC3028. I had all this
3636 sorted out, when out of a sudden new issues came to my mind. Oops."
3638 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
3641 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
3642 canonical form (as documented). However, after a host name lookup from
3643 the IP address, check_host() was doing a simple string comparison with
3644 addresses acquired from the DNS when checking that the found name did
3645 have the original IP as one of its addresses. Since any found IPv6
3646 addresses are likely to be in abbreviated form, the comparison could
3647 fail. Luckily, there already exists a function for doing the comparison
3648 by converting both addresses to binary, so now that is used instead of
3649 the text comparison.
3651 PH/20 There was another similar case to PH/19, when a complete host name was
3652 given in a host list; looking up its IP address could give an abbreviated
3653 form, whereas the current host's name might or might not be abbreviated.
3654 The same fix has been applied.
3660 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
3661 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
3664 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
3665 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
3667 PH/03 Typo: missing ".o" in src/pcre/Makefile.
3669 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
3670 header line, restrict the check to what is listed in RFCs 2369 and 2929.
3671 Also, for "Auto-Submitted", treat anything other than "no" as
3672 non-personal, in accordance with RFC 3834. (Previously it treated
3673 anything starting "auto-" as non-personal.)
3675 TF/01 The control=submission/name=... option had a problem with syntax
3676 errors if the name included a slash character. The /name= option
3677 now slurps the rest of the string, so it can include any characters
3678 but it must come last in the list of options (after /sender_retain
3681 PH/05 Some modifications to the interface to the fake nameserver for the new
3689 TK/01 Added the "success_on_redirect" address verification option. See
3690 NewStuff for rationale and an example.
3692 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
3694 PH/02 Patch to exigrep to allow it to work on syslog lines.
3696 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
3697 fread() to skip over the body file's header line, because in Cygwin the
3698 header line is locked and is inaccessible.
3700 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
3701 co-exist for some time) to make it clear that it is the Exim ID that is
3702 referenced, not the Message-ID: header line.
3704 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
3705 string_format() function, because snprintf() does not exist on all
3708 PH/06 The use of forbid_filter_existstest now also locks out the use of the
3709 ${stat: expansion item.
3711 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
3712 protocol synchronization error", to keep the pedants happy.
3714 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
3715 well as for IRIX systems, when gcc is being used. See the host.c source
3718 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
3720 PH/10 Named domain lists were not working if used in a queue_smtp_domains
3723 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
3724 transport and to the smtp transport in LMTP mode.
3726 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
3728 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
3729 run a filter in a subprocess. This could lead to confusion in subsequent
3730 lookups in the parent process. There should also be a search_tidyup() at
3731 the end of the subprocess.
3733 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
3734 only if the host matched helo_try_verify_hosts, which caused the
3735 verification to occur when the EHLO/HELO command was issued. The ACL just
3736 tested the remembered result. Now, if a previous verification attempt has
3737 not happened, "verify = helo" does it there and then.
3739 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
3741 TK/03 Fix log output including CR from clamd.
3743 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
3744 exist provoked a memory error which could cause a segfault.
3746 PH/15 Installed PCRE 6.2
3748 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
3750 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
3751 of the problem. Specifically, suggested +O2 rather than +O1 for the
3754 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
3756 PH/20 If a delivery was routed to a non-standard port by means of an SRV
3757 record, the port was not correctly logged when the outgoing_port log
3758 selector was set (it logged the transort's default port).
3760 PH/21 Added support for host-specific ports to manualroute, queryprogram,
3761 fallback_hosts, and "hosts" in the smtp transport.
3763 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
3764 host errors such as "Connection refused".
3766 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
3767 authentication with radiusclient 0.4.9:
3769 - Error returned from rc_read_config was caught wrongly
3770 - Username/password not passed on to radius server due to wrong length.
3772 The presumption is that some radiusclient API changes for 4.51/PH/17
3773 were not taken care of correctly. The code is still untested by me (my
3774 Linux distribution still has 0.3.2 of radiusclient), but it was
3775 contributed by a Radius user.
3777 PH/24 When doing a callout, the value of $domain wasn't set correctly when
3778 expanding the "port" option of the smtp transport.
3780 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
3781 while reading a MIME header. Thanks to Tom Hughes for a patch.
3783 PH/24 Include config.h inside local_scan.h so that configuration settings are
3786 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
3787 that in an ACL for RCPT (for example), you can examine exactly what was
3790 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
3791 commands, but it was not correctly comparing the address with the actual
3792 client host address. Thus, it would show the EHLO address in Received:
3793 header lines when this was not necessary.
3795 PH/27 Added the % operator to ${eval:}.
3797 PH/28 Exim tries to create and chdir to its spool directory when it starts;
3798 it should be ignoring failures (because with -C, for example, it has lost
3799 privilege). It wasn't ignoring creation failures other than "already
3802 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
3805 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
3806 because an input line was too long, either on its own, or by virtue of
3807 too many continuations, the temporary file was not being removed, and the
3808 return code was incorrect.
3810 PH/31 Missing "BOOL" in function definition in filtertest.c.
3812 PH/32 Applied Sieve patches from the maintainer.
3814 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
3816 PH/33 Added "verify = not_blind".
3818 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
3819 Local/Makefile (with some defaults set). These are used in built scripts
3820 such as exicyclog, but they have never been used in the exim_install
3821 script (though there are many overriding facilities there). I have
3822 arranged that the exim_install script now takes note of these two
3825 PH/35 Installed configuration files for Dragonfly.
3827 PH/36 When a locally submitted message by a trusted user did not contain a
3828 From: header, and the sender address was obtained from -f or from an SMTP
3829 MAIL command, and the trusted user did not use -F to supply a sender
3830 name, $originator_name was incorrectly used when constructing a From:
3831 header. Furthermore, $originator_name was used for submission mode
3832 messages from external hosts without From: headers in a similar way,
3833 which is clearly wrong.
3835 PH/37 Added control=suppress_local_fixups.
3837 PH/38 When log_selector = +received_sender was set, and the addition of the
3838 sender made the log line's construction buffer exactly full, or one byte
3839 less than full, an overflow happened when the terminating "\n" was
3842 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
3843 when the result of a list match is failure because a DNS lookup failed.
3845 PH/40 RM_COMMAND is now used in the building process.
3847 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
3848 the "build-* directories that it finds.
3850 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
3851 address in a domain literal was a prefix of an interface address.
3853 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
3854 when verifying a sender address, unless rewrite_headers is false.
3856 PH/44 Wrote a long comment about why errors_to addresses are verified as
3857 recipients, not senders.
3859 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
3860 the ratelimit ACL was added.
3862 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
3864 PH/46 Added extra information about PostgreSQL errors to the error string.
3866 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
3867 avoiding the need to install special zones in a real server. This is
3868 backwards compatible; if it can't find the fake resolver, it drops back.
3869 Thus, both old and new test suites can be run.
3871 TF/02 Added util/ratelimit.pl
3873 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
3874 clock is set back in time.
3876 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
3877 Candler <B.Candler@pobox.com>.
3879 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
3880 OK for addresses that are the result of redirections.
3882 PH/48 A number of further additions for the benefit of the new test suite,
3883 including a fake gethostbyname() that interfaces to the fake DNS resolver
3886 TF/06 The fix for widen_domains has also been applied to qualify_single and
3887 search_parents which are the other dnslookup options that can cause
3890 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
3893 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
3895 TF/07 Exim produced the error message "an SRV record indicated no SMTP
3896 service" if it encountered an MX record with an empty target hostname.
3897 The message is now "an MX or SRV record indicated no SMTP service".
3899 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
3900 if the DNS of the sending site is misconfigured. This is quite a
3901 common situation. This change restores the behaviour of treating a
3902 helo verification defer as a failure.
3904 PH/51 If self=fail was set on a router, the bounce message did not include the
3905 actual error message.
3911 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
3913 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
3914 command itself ended in error, the underlying message about the transport
3915 filter timeout was being overwritten with the pipe command error. Now the
3916 underlying error message should be appended to the second error message.
3918 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
3920 PH/02 Reduce the amount of output that "make" produces by default. Full output
3921 can still be requested.
3923 PH/03 The warning log line about a condition test deferring for a "warn" verb
3924 was being output only once per connection, rather than after each
3925 occurrence (because it was using the same function as for successful
3926 "warn" verbs). This seems wrong, so I have changed it.
3928 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
3929 it should not have, which might have caused a crash in the right
3930 circumstances, but probably never did.
3932 PH/04 Installed a modified version of Tony Finch's patch to make submission
3933 mode fix the return path as well as the Sender: header line, and to
3934 add a /name= option so that you can make the user's friendly name appear
3937 TF/03 Added the control = fakedefer ACL modifier.
3939 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
3940 Mark Lowes for thorough testing.
3942 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
3944 TK/03 Merged latest SRS patch from Miles Wilton.
3946 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
3947 with the definition in sysexits.h (which is #included earlier).
3948 Fortunately, Exim does not actually use EX_OK. The code used to try to
3949 preserve the sysexits.h value, by assuming that macro definitions were
3950 scanned for macro replacements. I have been disabused of this notion,
3951 so now the code just undefines EX_OK before #including unistd.h.
3953 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
3954 in the smtp transport. When a block could not be written in a single
3955 write() function, the timeout was being re-applied to each part-write.
3956 This seems wrong - if the receiver was accepting one byte at a time it
3957 would take for ever. The timeout is now adjusted when this happens. It
3958 doesn't have to be particularly precise.
3960 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
3961 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
3963 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
3964 without a domain if the check to <postmaster@domain> fails.
3966 SC/01 Eximstats: added -xls and the ability to specify output files
3967 (patch written by Frank Heydlauf).
3969 SC/02 Eximstats: use FileHandles for outputting results.
3971 SC/03 Eximstats: allow any combination of xls, txt, and html output.
3973 SC/04 Eximstats: fixed display of large numbers with -nvr option
3975 SC/05 Eximstats: fixed merging of reports with empty tables.
3977 SC/06 Eximstats: added the -include_original_destination flag
3979 SC/07 Eximstats: removed tabs and trailing whitespace.
3981 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
3983 TK/06 MBOX spool code: Add real "From " MBOX separator line
3984 so the .eml file is really in mbox format (even though
3985 most programs do not really care). Patch from Alex Miller.
3987 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
3988 The latter is generated from $received_to and is only set if the
3989 message has one envelope recipient. SA can use these headers,
3990 obviously out-of-the-box. Patch from Alex Miller.
3992 PH/08 The ${def test on a variable was returning false if the variable's
3993 value was "0", contrary to what the specification has always said!
3994 The result should be true unless the variable is empty.
3996 PH/09 The syntax error of a character other than { following "${if
3997 def:variable_name" (after optional whitespace) was not being diagnosed.
3998 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
3999 accidental colon was present, for example, could give incorrect results.
4001 PH/10 Tidied the code in a number of places where the st_size field of a stat()
4002 result is used (not including appendfile, where other changes are about
4005 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
4006 This involved changing a lot of size variables from int to off_t. It
4007 should work with maildirs and everything.
4009 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
4010 spamd dying while we are connected to it.
4012 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
4015 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
4016 function for BDB 4.3.
4018 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
4020 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
4021 the sources for PCRE have been re-arranged and more files are now
4024 PH/15 The code I had for printing potentially long long variables in PH/11
4025 above was not the best (it lost precision). The length of off_t variables
4026 is now inspected at build time, and an appropriate printing format (%ld
4027 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
4028 to be "long long int" or "long int". This is needed for the internal
4029 formatting function string_vformat().
4031 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
4032 the configuration file to be ":syslog", then the script "guesses" where
4033 the logs files are, rather than using the compiled in default. In our
4034 case the guess is not the same as the compiled default, so the script
4035 suddenly stopped working when I started to use syslog. The patch checks
4036 to see if log_file_path is "". If so, it attempts to read it from exim
4037 with no configuration file to get the compiled in version, before it
4038 falls back to the previous guessing code."
4040 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
4041 implementing BATV in an Exim configuration. See NewStuff for the gory
4044 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
4045 Makefile that are specific to HP-UX.
4047 PH/18 If the "use_postmaster" option was set for a recipient callout together
4048 with the "random" option, the postmaster address was used as the MAIL
4049 FROM address for the random test, but not for the subsequent recipient
4050 test. It is now used for both.
4052 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
4053 patch removes a few documentation additions to RFC 3028, because the
4054 latest draft now contains them. It adds the new en;ascii-case comparator
4055 and a new error check for 8bit text in MIME parts. Comparator and
4056 require names are now matched exactly. I enabled the subaddress
4057 extension, but it is not well tested yet (read: it works for me)."
4059 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
4060 rework some of the code of TK/09 above to avoid the hardwired use of
4061 "%lld" and "long long". Replaced the call to snprintf() with a call to
4064 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
4065 records point to non-existent hosts", "retry timeout exceeded", and
4066 "retry time not reached for any host after a long failure period".
4068 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
4069 experimental DomainKeys support:
4071 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
4072 (2) On an error such as an illegally used "control", the wrong name for
4073 the control was given.
4075 These problems did NOT occur unless DomainKeys support was compiled.
4077 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
4079 PH/24 Added ${if match_ip condition.
4081 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
4082 databases so that it will be absolutely obvious if a crash occurs in the
4083 DB library. This is a regular occurrence (often caused by mis-matched
4086 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
4087 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
4088 picked up on a user's system that detects such things. There doesn't seem
4089 to be a gcc warning option for this - only an attribute that has to be
4090 put on the function's prototype. It seems that in Fedora Core 4 they have
4091 set this on a number of new functions. No doubt there will be more in due
4094 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
4095 specify a transport. However, if an address that was verified by such a
4096 router was the subject of a callout, Exim crashed because it tried to
4097 read the rcpt_include_affixes from the non-existent transport. Now it
4098 just assumes that the setting of that option is false. This bug was
4099 introduced by 4.51/PH/31.
4101 PH/28 Changed -d+all to exclude +memory, because that information is very
4102 rarely of interest, but it makes the output a lot bigger. People tend to
4103 do -d+all out of habit.
4105 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
4106 code in os-type was giving problems when libc.so lives in lib64, like on
4109 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
4110 aren't the modern standard, and it seems that some systems' include files
4111 don't always have them. Exim was already checking for some of the newer
4112 ones like T_AAAA, and defining it itself. I've added checks for all the
4113 record types that Exim uses.
4115 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
4116 not automatically generating a new one, as it is supposed to. This
4117 prevented TLS from working. If the file did exist, but contained invalid
4118 data, a new version was generated, as expected. It was only the case of a
4119 non-existent file that was broken.
4121 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
4122 with a change in libdomainkeys > 0.64.
4124 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
4125 from DNS. If the selector record carries the flag, it now has
4126 precedence over the domain-wide flag.
4128 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
4130 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
4131 the use of a port name that isn't defined in /etc/services) occurred, the
4132 message was deferred as in a normal delivery, and thus remained on the
4133 spool, instead of being failed because of the mua_wrapper setting. This
4134 is now fixed, and I tidied up some of the mua_wrapper messages at the
4137 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
4138 the messages in a hash of arrays rather than using individual hashes.
4139 This is a bit cleaner and results in dramatic memory savings, albeit
4140 at a slight CPU cost.
4142 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
4143 as requested by Marc Sherman.
4145 SC/10 Eximstats: added histograms for user specified patterns as requested
4148 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
4150 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
4151 fopen() in the content-scanning modules that did not already have it.
4157 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
4158 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
4160 TK/02 Fix ACL "control" statement not being available in MIME ACL.
4162 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
4164 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
4165 to test Sieve filters that use "vacation".
4167 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
4168 that changes the way the GnuTLS parameters are stored in the cache file.
4169 The new format can be generated externally. For backward compatibility,
4170 if the data in the cache doesn't make sense, Exim assumes it has read an
4171 old-format file, and it generates new data and writes a new file. This
4172 means that you can't go back to an older release without removing the
4175 PH/03 A redirect router that has both "unseen" and "one_time" set does not
4176 work if there are any delivery delays because "one_time" forces the
4177 parent to be marked "delivered", so its unseen clone is never tried
4178 again. For this reason, Exim now forbids the simultaneous setting of
4181 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
4182 redirected to themselves ("homonym" addresses). Read the long ChangeLog
4183 entry if you want to know the details. The fix, however, neglected to
4184 consider the case when local delivery batching is involved. The test for
4185 "previously delivered" was not happening when checking to see if an
4186 address could be batched with a previous (undelivered) one; under
4187 certain circumstances this could lead to multiple deliveries to the same
4190 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
4191 in its include files, and this causes problems building Exim.
4193 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
4194 header_syntax) but Exim was just ignoring anything given after a slash.
4195 In particular, this caused confusion with an attempt to use "verify =
4196 reverse_host_lookup/defer_ok". An error is now given when options are
4197 supplied for verify items that do not have them. (Maybe reverse_host_
4198 lookup should have a defer_ok option, but that's a different point.)
4200 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
4201 defined by RFC 821) to 2048, because there were problems with some AUTH
4202 commands, and RFC 1869 says the size should be increased for extended
4203 SMTP commands that take arguments.
4205 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
4208 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
4209 "unknown" error; now it says that the functionality isn't in the binary.
4211 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
4212 an address' error message when a string expansion fails (syntax or
4213 whatever). Otherwise the password may appear in the log. Following change
4214 PH/42 below, there is no longer a chance of it appearing in a bounce
4217 PH/11 Installed exipick version 20050225.0 from John Jetmore.
4219 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
4220 of its addresses was ever tried. (Bugzilla bug #2.)
4222 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
4223 the result incorrectly in the debug output. (It correctly added a newline
4224 to what was transported.)
4226 TF/01 Added $received_time.
4228 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
4229 commented out examples of how to interface to a virus scanner and to
4230 SpamAssassin. Also added commented examples of av_scanner and
4231 spamd_address settings.
4233 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
4234 and controls are allowed in which ACLs. There were a couple of minor
4235 errors. Some of the entries in the conditions table (which is a table of
4236 where they are NOT allowed) were getting very unwieldy; rewrote them as a
4237 negation of where the condition IS allowed.
4239 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
4241 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
4242 header file does not have a version number, so I've had to invent a new
4243 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
4244 API. The code is untested by me (my Linux distribution still has 0.3.2 of
4245 radiusclient), but it was contributed by a Radius user.
4247 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
4248 files or directories, for OpenSSL.
4250 PH/19 When an Exim process that is running as root has to create an Exim log
4251 file, it does so in a subprocess that runs as exim:exim so as to get the
4252 ownership right at creation (otherwise, other Exim processes might see
4253 the file with the wrong ownership). There was no test for failure of this
4254 fork() call, which would lead to the process getting stuck as it waited
4255 for a non-existent subprocess. Forks do occasionally fail when resources
4256 run out. I reviewed all the other calls to fork(); they all seem to check
4259 PH/20 When checking for unexpected SMTP input at connect time (before writing
4260 the banner), Exim was not dealing correctly with a non-positive return
4261 from the read() function. If the client had disconnected by this time,
4262 the result was a log entry for a synchronization error with an empty
4263 string after "input=" when read() returned zero. If read() returned -1
4264 (an event I could not check), uninitialized data bytes were printed.
4265 There were reports of junk text (parts of files, etc) appearing after
4268 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
4270 PH/22 Added support for macro redefinition, and (re)definition in between
4271 driver and ACL definitions.
4273 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
4274 forgetting to use the resulting value; it was using the unexpanded value.
4276 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
4277 hadn't been configured. The fix is from Juergen Kreileder, who
4278 understands it better than I do:
4280 "Here's what I see happening with three configured cyrus_sasl
4281 authenticators configured (plain, login, cram-md5):
4283 On startup auth_cyrus_sasl_init() gets called for each of these.
4284 This means three calls to sasl_listmech() without a specified mech_list.
4285 => SASL tests which mechs of all available mechs actually work
4286 => three warnings about OTP not working
4287 => the returned list contains: plain, login, cram-md5, digest-md5, ...
4289 With the patch, sasl_listmech() also gets called three times. But now
4290 SASL's mech_list option is set to the server_mech specified in the the
4291 authenticator. Or in other words, the answer from sasl_listmech()
4292 gets limited to just the mech you're testing for (which is different
4294 => the return list contains just 'plain' or 'login', 'cram-md5' or
4295 nothing depending on the value of ob->server_mech.
4297 I've just tested the patch: Authentication still works fine,
4298 unavailable mechs specified in the exim configuration are still
4299 caught, and the auth.log warnings about OTP are gone."
4301 PH/25 When debugging is enabled, the contents of the command line are added
4302 to the debugging output, even when log_selector=+arguments is not
4305 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
4306 answer is "GNU", and only if the return is "GNU/something" is the answer
4309 PH/27 $acl_verify_message is now set immediately after the failure of a
4310 verification in an ACL, and so is available in subsequent modifiers. In
4311 particular, the message can be preserved by coding like this:
4313 warn !verify = sender
4314 set acl_m0 = $acl_verify_message
4316 Previously, $acl_verify_message was set only while expanding "message"
4317 and "log_message" when a very denied access.
4319 PH/28 Modified OS/os.c-Linux with
4321 -#ifndef OS_LOAD_AVERAGE
4322 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
4324 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
4325 nomenclature these days.)
4327 PH/29 Installed patch from the Sieve maintainer that adds the options
4328 sieve_useraddress and sieve_subaddress to the redirect router.
4330 PH/30 In these circumstances:
4331 . Two addresses routed to the same list of hosts;
4332 . First host does not offer TLS;
4333 . First host accepts first address;
4334 . First host gives temporary error to second address;
4335 . Second host offers TLS and a TLS session is established;
4336 . Second host accepts second address.
4337 Exim incorrectly logged both deliveries with the TLS parameters (cipher
4338 and peerdn, if requested) that were in fact used only for the second
4341 PH/31 When doing a callout as part of verifying an address, Exim was not paying
4342 attention to any local part prefix or suffix that was matched by the
4343 router that accepted the address. It now behaves in the same way as it
4344 does for delivery: the affixes are removed from the local part unless
4345 rcpt_include_affixes is set on the transport.
4347 PH/32 Add the sender address, as F=<...>, to the log line when logging a
4348 timeout during the DATA phase of an incoming message.
4350 PH/33 Sieve envelope tests were broken for match types other than :is. I have
4351 applied a patch sanctioned by the Sieve maintainer.
4353 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
4354 the uid or gid is negative. A case of a negative gid caused this to be
4355 noticed. The fix allows for either to be negative.
4357 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
4358 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
4361 PH/36 The change PH/12 above was broken. Fixed it.
4363 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
4364 the grounds that routing the same address twice would always produce the
4365 same answer. This might have been true once, but it is certainly no
4366 longer true now. Routing a child address may depend on the previous
4367 routing that produced that child. Some complicated redirection strategies
4368 went wrong when messages had multiple recipients, and made Exim's
4369 behaviour dependent on the order in which the addresses were given.
4371 I have moved the duplicate checking until after the routing is complete.
4372 Exim scans the addresses that are assigned to local and remote
4373 transports, and removes any duplicates. This means that more work will be
4374 done, as duplicates will always all be routed, but duplicates are
4375 presumably rare, so I don't expect this is of any significance.
4377 For deliveries to pipes, files, and autoreplies, the duplicate checking
4378 still happens during the routing process, since they are not going to be
4381 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
4382 It corrects a timeout issue with spamd. This is Ian's comment: "The
4383 background is that sometimes spamd either never reads data from a
4384 connection it has accepted, or it never writes response data. The exiscan
4385 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
4386 blindly assumes that writes won't block so it may never time out."
4388 PH/39 Allow G after quota size as well as K and M.
4390 PH/40 The value set for $authenticated_id in an authenticator may not contain
4391 binary zeroes or newlines because the value is written to log lines and
4392 to spool files. There was no check on this. Now the value is run through
4393 the string_printing() function so that such characters are converted to
4394 printable escape sequences.
4396 PH/41 $message_linecount is a new variable that contains the total number of
4397 lines in the message. Compare $body_linecount, which is the count for the
4400 PH/42 Exim no longer gives details of delivery errors for specific addresses in
4401 bounce and delay warning messages, except in certain special cases, which
4404 (a) An SMTP error message from a remote host;
4405 (b) A message specified in a :fail: redirection;
4406 (c) A message specified in a "fail" command in a system filter;
4407 (d) A message specified in a FAIL return from the queryprogram router;
4408 (e) A message specified by the cannot_route_message router option.
4410 In these cases only, Exim does include the error details in bounce and
4411 warning messages. There are also a few cases where bland messages such
4412 as "unrouteable address" or "local delivery error" are given.
4414 PH/43 $value is now also set for the "else" part of a ${run expansion.
4416 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
4417 being worked on, but at least Exim now implements the latest version to
4420 PH/45 In a pipe transport, although a timeout while waiting for the pipe
4421 process to complete was treated as a delivery failure, a timeout while
4422 writing the message to the pipe was logged, but erroneously treated as a
4423 successful delivery. Such timeouts include transport filter timeouts. For
4424 consistency with the overall process timeout, these timeouts are now
4425 treated as errors, giving rise to delivery failures by default. However,
4426 there is now a new Boolean option for the pipe transport called
4427 timeout_defer, which, if set TRUE, converts the failures into defers for
4428 both kinds of timeout. A transport filter timeout is now identified in
4431 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
4432 systems where "make" and "gmake" are different, calling "gmake" at top
4433 level broke things. I've arranged for the value of $(MAKE) to be passed
4434 from the Makefile to this script so that it can call the same version of
4438 A note about Exim versions 4.44 and 4.50
4439 ----------------------------------------
4441 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
4442 changes of various kinds. As a consequence, a big documentation update was
4443 needed. This delayed the release for rather longer than seemed good, especially
4444 in the light of a couple of (minor) security issues. Therefore, the changes
4445 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
4446 release. So 4.44 and 4.50 are in effect two different branches that both start
4449 I have left the 4.50 change log unchanged; it contains all the changes since
4450 4.43. The change log for 4.44 is below; many of its items are identical to
4451 those for 4.50. This seems to be the most sensible way to preserve the
4452 historical information.
4458 1. Minor wording change to the doc/README.SIEVE file.
4460 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
4461 computation of the current number of files was incorrect.
4463 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4464 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4467 4. Give more explanation in the error message when the command for a transport
4468 filter fails to execute.
4470 5. There are several places where Exim runs a non-Exim command in a
4471 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4472 was being done only for the command run by the queryprogram router. It is
4473 now done for all such subprocesses. The other cases are: ${run, transport
4474 filters, and the commands run by the lmtp and pipe transports.
4476 6. Added CONFIGURE_GROUP build-time option.
4478 7. Some older OS have a limit of 256 on the maximum number of file
4479 descriptors. Exim was using setrlimit() to set 1000 as a large value
4480 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4481 systems. I've change it so that if it can't get 1000, it tries for 256.
4483 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4484 was an oversight, and furthermore, ever since the addition of extra
4485 controls (e.g. 4.43/32), the checks on when to allow different forms of
4486 "control" were broken. There should now be diagnostics for all cases when a
4487 control that does not make sense is encountered.
4489 9. Added the /retain_sender option to "control=submission".
4491 10. $recipients is now available in the predata ACL (oversight).
4493 11. Tidy the search cache before the fork to do a delivery from a message
4494 received from the command line. Otherwise the child will trigger a lookup
4495 failure and thereby defer the delivery if it tries to use (for example) a
4496 cached ldap connection that the parent has called unbind on.
4498 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4499 of $address_data from the recipient verification was clobbered by the
4500 sender verification.
4502 13. The value of address_data from a sender verification is now available in
4503 $sender_address_data in subsequent conditions in the ACL statement.
4505 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
4507 15. Added a new option "connect=<time>" to callout options, to set a different
4510 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4511 was its contents. (It was OK if the option was not defined at all.)
4513 17. A "Completed" log line is now written for messages that are removed from
4514 the spool by the -Mrm option.
4516 18. New variables $sender_verify_failure and $recipient_verify_failure contain
4517 information about exactly what failed.
4519 19. Added -dd to debug only the daemon process.
4521 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
4522 handles timeouts, both on the server side and network timeouts. Renamed the
4523 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
4525 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
4526 It is now set to "smtps".
4528 22. $host_address is now set to the target address during the checking of
4529 ignore_target_hosts.
4531 23. When checking ignore_target_hosts for an ipliteral router, no host name was
4532 being passed; this would have caused $sender_host_name to have been used if
4533 matching the list had actually called for a host name (not very likely,
4534 since this list is usually IP addresses). A host name is now passed as
4537 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4538 code that specifies a non-restarting handler (typically sigaction() in
4539 modern systems) in an attempt to fix a rare and obscure crash bug.
4541 25. Narrowed the window for a race in the daemon that could cause it to ignore
4542 SIGCHLD signals. This is not a major problem, because they are used only to
4543 wake it up if nothing else does.
4545 26. A malformed maildirsize file could cause Exim to calculate negative values
4546 for the mailbox size or file count. Odd effects could occur as a result.
4547 The maildirsize information is now recalculated if the size or filecount
4550 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4551 support for a long time. Removed HAVE_SYS_VFS_H.
4553 28. Installed the latest version of exipick from John Jetmore.
4555 29. In an address list, if the pattern was not a regular expression, an empty
4556 subject address (from a bounce message) matched only if the pattern was an
4557 empty string. Non-empty patterns were not even tested. This was the wrong
4558 because it is perfectly reasonable to use an empty address as part of a
4559 database query. An empty address is now tested by patterns that are
4560 lookups. However, all the other forms of pattern expect the subject to
4561 contain a local part and a domain, and therefore, for them, an empty
4562 address still always fails if the pattern is not itself empty.
4564 30. Exim went into a mad DNS loop when attempting to do a callout where the
4565 host was specified on an smtp transport, and looking it up yielded more
4566 than one IP address.
4568 31. Re-factored the code for checking spool and log partition space into a
4569 function that finds that data and another that does the check. The former
4570 is then used to implement four new variables: $spool_space, $log_space,
4571 $spool_inodes, and $log_inodes.
4573 32. The RFC2047 encoding function was originally intended for short strings
4574 such as real names; it was not keeping to the 75-character limit for
4575 encoded words that the RFC imposes. It now respects the limit, and
4576 generates multiple encoded words if necessary. To be on the safe side, I
4577 have increased the buffer size for the ${rfc2047: expansion operator from
4580 33. It is now permitted to omit both strings after an "if" condition; if the
4581 condition is true, the result is "true". As before, when the second string
4582 is omitted, a false condition yields an empty string. This makes it less
4583 cumbersome to write custom ACL and router conditions.
4585 34. Failure to deliver a bounce message always caused it to be frozen, even if
4586 there was an errors_to setting on the router. The errors_to setting is now
4589 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4590 canonical form (fully expanded) before being placed in
4591 $sender_host_address.
4593 36. The table in the code that translates DNS record types into text (T_A to
4594 "A" for instance) was missing entries for NS and CNAME. It is just possible
4595 that this could have caused confusion if both these types were looked up
4596 for the same domain, because the text type is used as part of Exim's
4597 per-process caching. But the chance of anyone hitting this buglet seems
4600 37. The dnsdb lookup has been extended in a number of ways.
4602 (1) There is a new type, "zns", which walks up the domain tree until it
4603 finds some nameserver records. It should be used with care.
4605 (2) There is a new type, "mxh", which is like "mx" except that it returns
4606 just the host names, not the priorities.
4608 (3) It is now possible to give a list of domains (or IP addresses) to be
4609 looked up. The behaviour when one of the lookups defers can be
4610 controlled by a keyword.
4612 (4) It is now possible to specify the separator character for use when
4613 multiple records are returned.
4615 38. The dnslists ACL condition has been extended: it is now possible to supply
4616 a list of IP addresses and/or domains to be looked up in a particular DNS
4619 39. Added log_selector=+queue_time_overall.
4621 40. When running the queue in the test harness, wait just a tad after forking a
4622 delivery process, to get repeatability of debugging output.
4624 41. Include certificate and key file names in error message when GnuTLS fails
4625 to set them up, because the GnuTLS error message doesn't include the name
4626 of the failing file when there is a problem reading it.
4628 42. Allow both -bf and -bF in the same test run.
4630 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
4632 44. The "Exiscan patch" is now merged into the mainline Exim source.
4634 45. Sometimes the final signoff response after QUIT could fail to get
4635 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4636 < 0 before doing a fflush(). This bug looks as though it goes back to the
4637 introduction of TLS in release 3.20, but "sometimes" must have been rare
4638 because the tests only now provoked it.
4640 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
4641 (this can affect the format of dates).
4643 47. exim_tidydb, when checking for the continued existence of a message for
4644 which it has found a message-specific retry record, was not finding
4645 messages that were in split spool directories. Consequently, it was
4646 deleting retry records that should have stayed in existence.
4648 48. Steve fixed some bugs in eximstats.
4650 49. The SPA authentication driver was not abandoning authentication and moving
4651 on to the next authenticator when an expansion was forced to fail,
4652 contradicting the general specification for all authenticators. Instead it
4653 was generating a temporary error. It now behaves as specified.
4655 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
4656 (the order specifies the preference for clients). The order is now AES256,
4657 AES128, 3DES, ARCFOUR128.
4659 51. Small patch to Sieve code - explicitly set From: when generating an
4662 52. Exim crashed if a remote delivery caused a very long error message to be
4663 recorded - for instance if somebody sent an entire SpamAssassin report back
4664 as a large number of 550 error lines. This bug was coincidentally fixed by
4665 increasing the size of one of Exim's internal buffers (big_buffer) that
4666 happened as part of the Exiscan merge. However, to be on the safe side, I
4667 have made the code more robust (and fixed the comments that describe what
4670 53. Now that there can be additional text after "Completed" in log lines (if
4671 the queue_time_overall log selector is set), a one-byte patch to exigrep
4672 was needed to allow it to recognize "Completed" as not the last thing in
4675 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
4676 patch that reportedly fixes this has been added. I am not expert enough to
4677 create a test for it. This is what the patch creator wrote:
4679 "I found a little strange behaviour of ldap code when working with
4680 Windows 2003 AD Domain, where users was placed in more than one
4681 Organization Units. When I tried to give exim partial DN, the exit code
4682 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
4683 But simultaneously result of request was absolutely normal ldap result,
4684 so I produce this patch..."
4686 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
4687 REFERENCE, so I have modified the code to exclude the patch when that macro
4690 55. Some experimental protocols are using DNS PTR records for new purposes. The
4691 keys for these records are domain names, not reversed IP addresses. The
4692 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4693 leaves it alone. Component reversal etc. now happens only for IP addresses.
4696 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
4698 57. Double the size of the debug message buffer (to 2048) so that more of very
4699 long debug lines gets shown.
4701 58. The exicyclog utility now does better if the number of log files to keep
4702 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
4704 59. Two changes related to the smtp_active_hostname option:
4706 (1) $smtp_active_hostname is now available as a variable.
4707 (2) The default for smtp_banner uses $smtp_active_hostname instead
4708 of $primary_hostname.
4710 60. The host_aton() function is supposed to be passed a string that is known
4711 to be a valid IP address. However, in the case of IPv6 addresses, it was
4712 not checking this. This is a hostage to fortune. Exim now panics and dies
4713 if the condition is not met. A case was found where this could be provoked
4714 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4715 components; fortuitously, this particular loophole had already been fixed
4716 by change 4.50/55 above.
4718 If there are any other similar loopholes, the new check in host_aton()
4719 itself should stop them being exploited. The report I received stated that
4720 data on the command line could provoke the exploit when Exim was running as
4721 exim, but did not say which command line option was involved. All I could
4722 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4723 running as the user.
4726 61. There was a buffer overflow vulnerability in the SPA authentication code
4727 (which came originally from the Samba project). I have added a test to the
4728 spa_base64_to_bits() function which I hope fixes it.
4731 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
4732 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
4734 63. The daemon start-up calls getloadavg() while still root for those OS that
4735 need the first call to be done as root, but it missed one case: when
4736 deliver_queue_load_max is set with deliver_drop_privilege. This is
4737 necessary for the benefit of the queue runner, because there is no re-exec
4738 when deliver_drop_privilege is set.
4740 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
4741 This has been fixed.
4743 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4744 was in use, was not putting the data itself into the right store pool;
4745 consequently, it could be overwritten for a subsequent message in the same
4746 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4749 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
4751 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
4752 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
4753 the function were treating the return as a boolean value, which happened to
4754 work because 0=false and not-0=true, but is not correct code.
4756 68. The host_aton() function was not handling scoped IPv6 addresses (those
4757 with, for example, "%eth0" on the end) correctly.
4759 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
4760 negated items (that is, ~something) in unsigned ints. Some compilers
4761 apparently mutter when there is no cast.
4763 70. If an address verification called from an ACL failed, and did not produce a
4764 user-specific message (i.e. there was only a "system" message), nothing was
4765 put in $acl_verify_message. In this situation, it now puts the system
4768 71. Change 4.23/11 added synchronization checking at the start of an SMTP
4769 session; change 4.31/43 added the unwanted input to the log line - except
4770 that it did not do this in the start of session case. It now does.
4772 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
4773 This is wrong and can cause the other end to generate a synchronization
4774 error if it is another Exim or anything else that does the synchronization
4775 check. A QUIT command is no longer sent after a timeout.
4777 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
4778 during host lookups.
4780 74. The defer_ok option of callout verification was not working if it was used
4781 when verifying addresses in header lines, that is, for this case:
4783 verify = header_sender/callout=defer_ok
4785 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
4786 those file descriptors could be used for SMTP connections. If anything
4787 wrote to stderr (the example that came up was "warn" in embedded Perl), it
4788 could be sent to the SMTP client, causing chaos. The daemon now opens
4789 stdin, stdout, and stderr to /dev/null when it puts itself into the
4792 76. Arrange for output from Perl's "warn" command to be written to Exim's main
4793 log by default. The user can override this with suitable Perl magic.
4795 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
4796 the log message when discard triggers, was not working for the DATA ACL or
4797 for the non-SMTP ACL.
4799 78. Error message wording change in sieve.c.
4801 79. If smtp_accept_max_per_host was set, the number of connections could be
4802 restricted to fewer than expected, because the daemon was trying to set up
4803 a new connection before checking whether the processes handling previous
4804 connections had finished. The check for completed processes is now done
4805 earlier. On busy systems, this bug wouldn't be noticed because something
4806 else would have woken the daemon, and it would have reaped the completed
4809 80. If a message was submitted locally by a user whose login name contained one
4810 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
4811 It caused a spool format error. I have fixed the spool reading code. A
4812 related problem was that the "from" clause in the Received: line became
4813 illegal because of the space(s). It is now covered by ${quote_local_part.
4815 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
4818 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
4820 83. Patch from Sieve maintainer to fix unterminated string problem in
4821 "vacation" handling.
4823 84. Some minor changes to the Linux configuration files to help with other
4824 OS variants using glibc.
4826 85. One more patch for Sieve to update vacation handling to latest spec.
4829 ----------------------------------------------------
4830 See the note above about the 4.44 and 4.50 releases.
4831 ----------------------------------------------------
4837 1. Change 4.43/35 introduced a bug that caused file counts to be
4838 incorrectly computed when quota_filecount was set in an appendfile
4841 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4842 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4845 3. Give more explanation in the error message when the command for a transport
4846 filter fails to execute.
4848 4. There are several places where Exim runs a non-Exim command in a
4849 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4850 was being done only for the command run by the queryprogram router. It is
4851 now done for all such subprocesses. The other cases are: ${run, transport
4852 filters, and the commands run by the lmtp and pipe transports.
4854 5. Some older OS have a limit of 256 on the maximum number of file
4855 descriptors. Exim was using setrlimit() to set 1000 as a large value
4856 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4857 systems. I've change it so that if it can't get 1000, it tries for 256.
4859 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4860 was an oversight, and furthermore, ever since the addition of extra
4861 controls (e.g. 4.43/32), the checks on when to allow different forms of
4862 "control" were broken. There should now be diagnostics for all cases when a
4863 control that does not make sense is encountered.
4865 7. $recipients is now available in the predata ACL (oversight).
4867 8. Tidy the search cache before the fork to do a delivery from a message
4868 received from the command line. Otherwise the child will trigger a lookup
4869 failure and thereby defer the delivery if it tries to use (for example) a
4870 cached ldap connection that the parent has called unbind on.
4872 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4873 of $address_data from the recipient verification was clobbered by the
4874 sender verification.
4876 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4877 was its contents. (It was OK if the option was not defined at all.)
4879 11. A "Completed" log line is now written for messages that are removed from
4880 the spool by the -Mrm option.
4882 12. $host_address is now set to the target address during the checking of
4883 ignore_target_hosts.
4885 13. When checking ignore_target_hosts for an ipliteral router, no host name was
4886 being passed; this would have caused $sender_host_name to have been used if
4887 matching the list had actually called for a host name (not very likely,
4888 since this list is usually IP addresses). A host name is now passed as
4891 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4892 code that specifies a non-restarting handler (typically sigaction() in
4893 modern systems) in an attempt to fix a rare and obscure crash bug.
4895 15. Narrowed the window for a race in the daemon that could cause it to ignore
4896 SIGCHLD signals. This is not a major problem, because they are used only to
4897 wake it up if nothing else does.
4899 16. A malformed maildirsize file could cause Exim to calculate negative values
4900 for the mailbox size or file count. Odd effects could occur as a result.
4901 The maildirsize information is now recalculated if the size or filecount
4904 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4905 support for a long time. Removed HAVE_SYS_VFS_H.
4907 18. Updated exipick to current release from John Jetmore.
4909 19. Allow an empty sender to be matched against a lookup in an address list.
4910 Previously the only cases considered were a regular expression, or an
4913 20. Exim went into a mad DNS lookup loop when doing a callout where the
4914 host was specified on the transport, if the DNS lookup yielded more than
4917 21. The RFC2047 encoding function was originally intended for short strings
4918 such as real names; it was not keeping to the 75-character limit for
4919 encoded words that the RFC imposes. It now respects the limit, and
4920 generates multiple encoded words if necessary. To be on the safe side, I
4921 have increased the buffer size for the ${rfc2047: expansion operator from
4924 22. Failure to deliver a bounce message always caused it to be frozen, even if
4925 there was an errors_to setting on the router. The errors_to setting is now
4928 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4929 canonical form (fully expanded) before being placed in
4930 $sender_host_address.
4932 24. Updated eximstats to version 1.33
4934 25. Include certificate and key file names in error message when GnuTLS fails
4935 to set them up, because the GnuTLS error message doesn't include the name
4936 of the failing file when there is a problem reading it.
4938 26. Expand error message when OpenSSL has problems setting up cert/key files.
4941 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
4942 (this can affect the format of dates).
4944 28. exim_tidydb, when checking for the continued existence of a message for
4945 which it has found a message-specific retry record, was not finding
4946 messages that were in split spool directories. Consequently, it was
4947 deleting retry records that should have stayed in existence.
4949 29. eximstats updated to version 1.35
4950 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
4951 1.35 - bugfix such that pie charts by volume are generated correctly
4953 30. The SPA authentication driver was not abandoning authentication and moving
4954 on to the next authenticator when an expansion was forced to fail,
4955 contradicting the general specification for all authenticators. Instead it
4956 was generating a temporary error. It now behaves as specified.
4958 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
4959 (the order specifies the preference for clients). The order is now AES256,
4960 AES128, 3DES, ARCFOUR128.
4962 31. Small patch to Sieve code - explicitly set From: when generating an
4965 32. Exim crashed if a remote delivery caused a very long error message to be
4966 recorded - for instance if somebody sent an entire SpamAssassin report back
4967 as a large number of 550 error lines. This bug was coincidentally fixed by
4968 increasing the size of one of Exim's internal buffers (big_buffer) that
4969 happened as part of the Exiscan merge. However, to be on the safe side, I
4970 have made the code more robust (and fixed the comments that describe what
4973 33. Some experimental protocols are using DNS PTR records for new purposes. The
4974 keys for these records are domain names, not reversed IP addresses. The
4975 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4976 leaves it alone. Component reversal etc. now happens only for IP addresses.
4979 34. The host_aton() function is supposed to be passed a string that is known
4980 to be a valid IP address. However, in the case of IPv6 addresses, it was
4981 not checking this. This is a hostage to fortune. Exim now panics and dies
4982 if the condition is not met. A case was found where this could be provoked
4983 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4984 components; fortuitously, this particular loophole had already been fixed
4985 by change 4.50/55 or 4.44/33 above.
4987 If there are any other similar loopholes, the new check in host_aton()
4988 itself should stop them being exploited. The report I received stated that
4989 data on the command line could provoke the exploit when Exim was running as
4990 exim, but did not say which command line option was involved. All I could
4991 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4992 running as the user.
4995 35. There was a buffer overflow vulnerability in the SPA authentication code
4996 (which came originally from the Samba project). I have added a test to the
4997 spa_base64_to_bits() function which I hope fixes it.
5000 36. The daemon start-up calls getloadavg() while still root for those OS that
5001 need the first call to be done as root, but it missed one case: when
5002 deliver_queue_load_max is set with deliver_drop_privilege. This is
5003 necessary for the benefit of the queue runner, because there is no re-exec
5004 when deliver_drop_privilege is set.
5006 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
5007 was in use, was not putting the data itself into the right store pool;
5008 consequently, it could be overwritten for a subsequent message in the same
5009 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
5012 38. Sometimes the final signoff response after QUIT could fail to get
5013 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
5014 < 0 before doing a fflush(). This bug looks as though it goes back to the
5015 introduction of TLS in release 3.20, but "sometimes" must have been rare
5016 because the tests only now provoked it.
5022 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
5023 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
5024 Now it returns an integer. A number of calls were still expecting a T/F
5025 return. Fortuitously, in all cases, the tests worked in OK situations,
5026 which is the norm. However, things would have gone wrong on any write
5027 failures on the smtp file descriptor. This function is used when sending
5028 messages over SMTP and also when doing verify callouts.
5030 2. When Exim is called to do synchronous delivery of a locally submitted
5031 message (the -odf or -odi options), it no longer closes stderr before doing
5034 3. Implemented the mua_wrapper option.
5036 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
5038 5. Implemented the functions header_remove(), header_testname(),
5039 header_add_at_position(), and receive_remove_recipient(), and exported them
5042 6. If an ACL "warn" statement specified the addition of headers, Exim already
5043 inserted X-ACL-Warn: at the start if there was no header name. However, it
5044 was not making this test for the second and subsequent header lines if
5045 there were newlines in the string. This meant that an invalid header could
5046 be inserted if Exim was badly configured.
5048 7. Allow an ACL "warn" statement to add header lines at the start or after all
5049 the Received: headers, as well as at the end.
5051 8. Added the rcpt_4xx retry error code.
5053 9. Added postmaster_mailfrom=xxx to callout verification option.
5055 10. Added mailfrom=xxxx to the callout verification option, for verify=
5058 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
5059 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
5061 12. Inserted some casts to stop certain compilers warning when using pointer
5062 differences as field lengths or precisions in printf-type calls (mostly
5063 affecting debugging statements).
5065 13. Added optional readline() support for -be (dynamically loaded).
5067 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
5068 same clock tick as a message's arrival, so that its received time was the
5069 same as the "first fail" time on the retry record, and that message
5070 remained on the queue past the ultimate address timeout, every queue runner
5071 would try a delivery (because it was past the ultimate address timeout) but
5072 after another failure, the ultimate address timeout, which should have then
5073 bounced the address, did not kick in. This was a "< instead of <=" error;
5074 in most cases the first failure would have been in the next clock tick
5075 after the received time, and all would be well.
5077 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
5078 being recognized when the domain list was tested by the match_domain
5079 condition in an expansion string.
5081 16. Added the ${str2b64: operator.
5083 17. Exim was always calling setrlimit() to set a large limit for the number of
5084 processes, without checking whether the existing limit was already
5085 adequate. (It did check for the limit on file descriptors.) Furthermore,
5086 errors from getrlimit() and setrlimit() were being ignored. Now they are
5087 logged to the main and panic logs, but Exim does carry on, to try to do its
5088 job under whatever limits there are.
5090 18. Imported PCRE 5.0.
5092 19. Trivial typo in log message " temporarily refused connection" (the leading
5095 20. If the log selector return_path_on_delivery was set and an address was
5096 redirected to /dev/null, the delivery process crashed because it assumed
5097 that a return path would always be set for a "successful" delivery. In this
5098 case, the whole delivery is bypassed as an optimization, and therefore no
5101 21. Internal re-arrangement: the function for sending a challenge and reading
5102 a response while authentication was assuming a zero-terminated challenge
5103 string. It's now changed to take a pointer and a length, to allow for
5104 binary data in such strings.
5106 22. Added the cyrus_sasl authenticator (code supplied by MBM).
5108 23. Exim was not respecting finduser_retries when seeking the login of the
5109 uid under which it was called; it was always trying 10 times. (The default
5110 setting of finduser_retries is zero.) Also, it was sleeping after the final
5111 failure, which is pointless.
5113 24. Implemented tls_on_connect_ports.
5115 25. Implemented acl_smtp_predata.
5117 26. If the domain in control=submission is set empty, Exim assumes that the
5118 authenticated id is a complete email address when it generates From: or
5119 Sender: header lines.
5121 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
5122 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
5123 chown and chgrp in /bin and hostname in /usr/bin.
5125 28. Exim was keeping the "process log" file open after each use, just as it
5126 does for the main log. This opens the possibility of it remaining open for
5127 long periods when the USR1 signal hits a daemon. Occasional processlog
5128 errors were reported, that could have been caused by this. Anyway, it seems
5129 much more sensible not to leave this file open at all, so that is what now
5132 29. The long-running daemon process does not normally write to the log once it
5133 has entered its main loop, and it closes the log before doing so. This is
5134 so that log files can straightforwardly be renamed and moved. However,
5135 there are a couple of unusual error situations where the daemon does write
5136 log entries, and I had neglected to close the log afterwards.
5138 30. The text of an SMTP error response that was received during a remote
5139 delivery was being truncated at 512 bytes. This is too short for some of
5140 the long messages that one sometimes sees. I've increased the limit to
5143 31. It is now possible to make retry rules that apply only when a message has a
5144 specific sender, in particular, an empty sender.
5146 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
5147 it possible to be selective about when SMTP synchronization is enforced.
5149 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
5151 32. Implemented hosts_connection_nolog.
5153 33. Added an ACL for QUIT.
5155 34. Setting "delay_warning=" to disable warnings was not working; it gave a
5158 35. Added mailbox_size and mailbox_filecount to appendfile.
5160 36. Added control = no_multiline_responses to ACLs.
5162 37. There was a bug in the logic of the code that waits for the clock to tick
5163 in the case where the clock went backwards by a substantial amount such
5164 that the microsecond fraction of "now" was more than the microsecond
5165 fraction of "then" (but the whole seconds number was less).
5167 38. Added support for the libradius Radius client library this is found on
5168 FreeBSD (previously only the radiusclient library was supported).
5174 1. When certain lookups returned multiple values in the form name=value, the
5175 quoting of the values was not always being done properly. Specifically:
5176 (a) If the value started with a double quote, but contained no whitespace,
5178 (b) If the value contained whitespace other than a space character (i.e.
5179 tabs or newlines or carriage returns) it was not quoted.
5180 This fix has been applied to the mysql and pgsql lookups by writing a
5181 separate quoting function and calling it from the lookup code. The fix
5182 should probably also be applied to nisplus, ibase and oracle lookups, but
5183 since I cannot test any of those, I have not disturbed their existing code.
5185 2. A hit in the callout cache for a specific address caused a log line with no
5186 reason for rejecting RCPT. Now it says "Previous (cached) callout
5187 verification failure".
5189 3. There was an off-by-one bug in the queryprogram router. An over-long
5190 return line was truncated at 256 instead of 255 characters, thereby
5191 overflowing its buffer with the terminating zero. As well as fixing this, I
5192 have increased the buffer size to 1024 (and made a note to document this).
5194 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
5195 when Exim is waiting for an SMTP response from a remote server, Exim
5196 restarts its select() call on the socket, thereby resetting its timeout.
5197 This is not a problem when such interrupts are rare. Somebody set up a cron
5198 job to run exiwhat every 2 minutes, which is less than the normal select()
5199 timeout (5 or 10 minutes). This meant that the select() timeout never
5200 kicked in because it was always reset. I have fixed this by comparing the
5201 time when an interrupt arrives with the time at the start of the first call
5202 to select(). If more time than the timeout has elapsed, the interrupt is
5203 treated as a timeout.
5205 5. Some internal re-factoring in preparation for the addition of Sieve
5206 extensions (by MH). In particular, the "personal" test is moved to a
5207 separate function, and given an option for scanning Cc: and Bcc: (which is
5208 not set for Exim filters).
5210 6. When Exim created an email address using the login of the caller as the
5211 local part (e.g. when creating a From: or Sender: header line), it was not
5212 quoting the local part when it contained special characters such as @.
5214 7. Installed new OpenBSD configuration files.
5216 8. Reworded some messages for syntax errors in "and" and "or" conditions to
5217 try to make them clearer.
5219 9. Callout options, other than the timeout value, were being ignored when
5220 verifying sender addresses in header lines. For example, when using
5222 verify = header_sender/callout=no_cache
5224 the cache was (incorrectly) being used.
5226 10. Added a missing instance of ${EXE} to the exim_install script; this affects
5227 only the Cygwin environment.
5229 11. When return_path_on_delivery was set as a log selector, if different remote
5230 addresses in the same message used different return paths and parallel
5231 remote delivery occurred, the wrong values would sometimes be logged.
5232 (Whenever a remote delivery process finished, the return path value from
5233 the most recently started remote delivery process was logged.)
5235 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
5236 lines when AUTH and/or TLS are in use. This is the "received protocol"
5237 field. Exim used to use "asmtp" for authenticated SMTP, without any
5238 indication (in the protocol name) for TLS use. Now it follows the RFC and
5239 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
5240 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
5241 names appear in log lines as well as in Received: header lines.
5243 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
5244 extensions, and comparison tests, and to fix some bugs.
5246 14. Changes to the "personal" filter test:
5248 (1) The test was buggy in that it was just doing the equivalent of
5249 "contains" tests on header lines. For example, if a user's address was
5250 anne@some.where, the "personal" test would incorrectly be true for
5252 To: susanne@some.where
5254 This test is now done by extracting each address from the header in turn,
5255 and checking the entire address. Other tests that are part of "personal"
5256 are now done using regular expressions (for example, to check local parts
5257 of addresses in From: header lines).
5259 (2) The list of non-personal local parts in From: addresses has been
5260 extended to include "listserv", "majordomo", "*-request", and "owner-*",
5261 taken from the Sieve specification recommendations.
5263 (3) If the message contains any header line starting with "List-" it is
5264 treated as non-personal.
5266 (4) The test for "circular" in the Subject: header line has been removed
5267 because it now seems ill-conceived.
5269 15. Minor typos in src/EDITME comments corrected.
5271 16. Installed latest exipick from John Jetmore.
5273 17. If headers_add on a router specified a text string that was too long for
5274 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
5275 of string_sprintf() is now avoided.
5277 18. $message_body_size was not set (it was always zero) when running the DATA
5278 ACL and the local_scan() function.
5280 19. For the "mail" command in an Exim filter, no default was being set for
5281 the once_repeat time, causing a random time value to be used if "once" was
5282 specified. (If the value happened to be <= 0, no repeat happened.) The
5283 default is now 0s, meaning "never repeat". The "vacation" command was OK
5284 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
5285 (I found it when inspecting the code).
5287 20. There is now an overall timeout for performing a callout verification. It
5288 defaults to 4 times the callout timeout, which applies to individual SMTP
5289 commands during the callout. The overall timeout applies when there is more
5290 than one host that can be tried. The timeout is checked before trying the
5291 next host. This prevents very long delays if there are a large number of
5292 hosts and all are timing out (e.g. when the network connections are timing
5293 out). The value of the overall timeout can be changed by specifying an
5294 additional sub-option for "callout", called "maxwait". For example:
5296 verify = sender/callout=5s,maxwait=20s
5298 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
5299 to the end before writing, but this should make it even safer).
5301 22. Exim was forgetting that it had advertised PIPELINING for the second and
5302 subsequent messages on an SMTP connection. It was also not resetting its
5303 memory on STARTTLS and an internal HELO.
5305 23. When Exim logs an SMTP synchronization error within a session, it now
5306 records whether PIPELINING has been advertised or not.
5308 24. Added 3 instances of "(long int)" casts to time_t variables that were being
5309 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
5310 rather than long int.
5312 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
5314 26. Added the never_mail option to autoreply.
5320 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
5321 crash if the getsockname() call failed; this can happen if a connection is
5322 closed very soon after it is established. The problem was simply in the
5323 order in which certain operations were done, causing Exim to try to write
5324 to the SMTP stream before it had set up the file descriptor. The bug has
5325 been fixed by making things happen in the correct order.
5331 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
5332 before the connection was closed, thus losing the rejection response.
5334 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
5335 some early Solaris releases, but causes trouble in current releases where
5336 socklen_t is defined.
5338 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
5341 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
5344 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
5345 that releasing the top of it at the end releases what was used for sub-
5346 expansions (unless the block got too big). However, discard this block if
5347 the first thing is a variable or header, so that we can use its block when
5348 it is dynamic (useful for very large $message_headers, for example).
5350 6. Lookups now cache *every* query, not just the most recent. A new, separate
5351 store pool is used for this. It can be recovered when all lookup caches are
5352 flushed. Lookups now release memory at the end of their result strings.
5353 This has involved some general refactoring of the lookup sources.
5355 7. Some code has been added to the store_xxx() functions to reduce the amount
5356 of flapping under certain conditions.
5358 8. log_incoming_interface used to affect only the <= reception log lines. Now
5359 it causes the local interface and port to be added to several more SMTP log
5360 lines, for example "SMTP connection from", and rejection lines.
5362 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
5364 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
5366 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
5367 could be overwritten at the end of the current message (or the start of a
5368 new message if it was set in a HELO ACL). The value is now preserved for
5369 the duration of the SMTP connection.
5371 12. If a transport had a headers_rewrite setting, and a matching header line
5372 contained an unqualified address, that address was qualified, even if it
5373 did not match any rewriting rules. The underlying bug was that the values
5374 of the flags that permit the existence of unqualified sender and recipient
5375 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
5376 non-local messages, and by -bnq for local messages) were not being
5377 preserved with the message after it was received.
5379 13. When Exim was logging an SMTP synchronization error, it could sometimes log
5380 "next input=" as part of the text comprising the host identity instead of
5381 the correct text. The code was using the same buffer for two different
5382 strings. However, depending on which order the printing function evaluated
5383 its arguments, the bug did not always show up. Under Linux, for example, my
5384 test suite worked just fine.
5386 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
5387 doesn't work with some older versions of Perl. It has been changed to "my",
5388 which in any case is probably the better facility to use.
5390 15. A really picky compiler found some instances of statements for creating
5391 error messages that either had too many or two few arguments for the format
5394 16. The size of the buffer for calls to the DNS resolver has been increased
5395 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
5396 for addresses that have a lot of PTR records. This alleviates a problem; it
5397 does not fully solve it.
5399 17. A dnsdb lookup for PTR records that receives more data than will fit in the
5400 buffer now truncates the list and logs the incident, which is the same
5401 action as happens when Exim is looking up a host name and its aliases.
5402 Previously in this situation something unpredictable would happen;
5403 sometimes it was "internal error: store_reset failed".
5405 18. If a server dropped the connection unexpectedly when an Exim client was
5406 using GnuTLS and trying to read a response, the client delivery process
5407 crashed while trying to generate an error log message.
5409 19. If a "warn" verb in an ACL added multiple headers to a message in a single
5410 string, for example:
5412 warn message = H1: something\nH2: something
5414 the text was added as a single header line from Exim's point of view
5415 though it ended up OK in the delivered message. However, searching for the
5416 second and subsequent header lines using $h_h2: did not work. This has been
5417 fixed. Similarly, if a system filter added multiple headers in this way,
5418 the routers could not see them.
5420 20. Expanded the error message when iplsearch is called with an invalid key to
5421 suggest using net-iplsearch in a host list.
5423 21. When running tests using -bh, any delays imposed by "delay" modifiers in
5424 ACLs are no longer actually imposed (and a message to that effect is
5427 22. If a "gecos" field in a passwd entry contained escaped characters, in
5428 particular, if it contained a \" sequence, Exim got it wrong when building
5429 a From: or a Sender: header from that name. A second bug also caused
5430 incorrect handling when an unquoted " was present following a character
5431 that needed quoting.
5433 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
5434 was not being matched caselessly.
5436 24. Arranged for all hyphens in the exim.8 source to be escaped with
5439 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
5440 properly. Recipient callout cache records were still being keyed to include
5441 the sender, even when use_sender was set false. This led to far more
5442 callouts that were necessary. The sender is no longer included in the key
5443 when use_sender is false.
5445 26. Added "control = submission" modifier to ACLs.
5447 27. Added the ${base62d: operator to decode base 62 numbers.
5449 28. dnsdb lookups can now access SRV records.
5451 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
5452 the configuration file.
5454 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
5455 (-v) mode. This makes the output for a verbose queue run more intelligible.
5457 31. Added a use_postmaster feature to recipient callouts.
5459 32. Added the $body_zerocount variable, containing the number of binary zero
5460 bytes in the message body.
5462 33. The time of last modification of the "new" subdirectory is now used as the
5463 "mailbox time last read" when there is a quota error for a maildir
5466 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
5468 35. Added +ignore_unknown as a special item in host lists.
5470 36. Code for decoding IPv6 addresses in host lists is now included, even if
5471 IPv6 support is not being compiled. This fixes a bug in which an IPv6
5472 address was recognized as an IP address, but was then not correctly decoded
5473 into binary, causing unexpected and incorrect effects when compared with
5480 1. Very minor rewording of debugging text in manualroute to say "list of
5481 hosts" instead of "hostlist".
5483 2. If verify=header_syntax was set, and a header line with an unqualified
5484 address (no domain) and a large number of spaces between the end of the
5485 name and the colon was received, the reception process suffered a buffer
5486 overflow, and (when I tested it) crashed. This was caused by some obsolete
5487 code that should have been removed. The fix is to remove it!
5489 3. When running in the test harness, delay a bit after writing a bounce
5490 message to get a bit more predictability in the log output.
5492 4. Added a call to search_tidyup() just before forking a reception process. In
5493 theory, someone could use a lookup in the expansion of smtp_accept_max_
5494 per_host which, without the tidyup, could leave open a database connection.
5496 5. Added the variables $recipient_data and $sender_data which get set from a
5497 lookup success in an ACL "recipients" or "senders" condition, or a router
5498 "senders" option, similar to $domain_data and $local_part_data.
5500 6. Moved the writing of debug_print from before to after the "senders" test
5503 7. Change 4.31/66 (moving the time when the Received: is generated) caused
5504 problems for message scanning, either using a data ACL, or using
5505 local_scan() because the Received: header was not generated till after they
5506 were called (in order to set the time as the time of reception completion).
5507 I have revised the way this works. The header is now generated after the
5508 body is received, but before the ACL or local_scan() are called. After they
5509 are run, the timestamp in the header is updated.
5515 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
5516 before starting a queue runner without re-exec. This happened only when
5517 deliver_drop_privilege was set or when the Exim user was set to root. The
5518 effect of the bug was that timeouts during subsequent deliveries caused
5519 crashes instead of being properly handled. The handler is now left at its
5520 default (and expected) setting.
5522 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
5523 message, again when deliver_drop_privilege is set or Exim is run as root.
5524 The bug described in (1) was not present in this case, but the tidying up
5525 of the other signals was missing. I have made the two cases consistent.
5527 3. The ignore_target_hosts setting on a manualroute router was being ignored
5528 for hosts that were looked up using the /MX notation.
5530 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
5533 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
5534 operated on the sender address. After changing the $sender_address to <>
5535 for the sender address verify, Exim was re-instated it as the original
5536 (before rewriting) address, but remembering that it had rewritten it, so it
5537 wasn't rewriting it again. This bug also had the effect of breaking the
5538 sender address verification caching when the sender address was rewritten.
5540 6. The ignore_target_hosts option was being ignored by the ipliteral router.
5541 This has been changed so that if the ip literal address matches
5542 ignore_target_hosts, the router declines.
5544 7. Added expansion conditions match_domain, match_address, and match_local_
5545 part (NOT match_host).
5547 8. The placeholder for the Received: header didn't have a length field set.
5549 9. Added code to Exim itself and to exim_lock to test for a specific race
5550 condition that could lead to file corruption when using MBX delivery. The
5551 issue is with the lockfile that is created in /tmp. If this file is removed
5552 after a process has opened it but before that process has acquired a lock,
5553 there is the potential for a second process to recreate the file and also
5554 acquire a lock. This could lead to two Exim processes writing to the file
5555 at the same time. The added code performs the same test as UW imapd; it
5556 checks after acquiring the lock that its file descriptor still refers to
5557 the same named file.
5559 10. The buffer for building added header lines was of fixed size, 8192 bytes.
5560 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
5563 11. Added the smtp_active_hostname option. If used, this will typically be made
5564 to depend on the incoming interface address. Because $interface_address is
5565 not set up until the daemon has forked a reception process, error responses
5566 that can happen earlier (such as "too many connections") no longer contain
5569 12. If an expansion in a condition on a "warn" statement fails because a lookup
5570 defers, the "warn" statement is abandoned, and the next ACL statement is
5571 processed. Previously this caused the whole ACL to be aborted.
5573 13. Added the iplsearch lookup type.
5575 14. Added ident_timeout as a log selector.
5577 15. Added tls_certificate_verified as a log selector.
5579 16. Added a global option tls_require_ciphers (compare the smtp transport
5580 option of the same name). This controls incoming TLS connections.
5582 17. I finally figured out how to make tls_require_ciphers do a similar thing
5583 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
5584 before starting the TLS session.
5586 18. Tabs are now shown as \t in -bP output.
5588 19. If the log selector return_path_on_delivery was set, Exim crashed when
5589 bouncing a message because it had too many Received: header lines.
5591 20. If two routers both had headers_remove settings, and the first one included
5592 a superfluous trailing colon, the final name in the first list and the
5593 first name in the second list were incorrectly joined into one item (with a
5594 colon in the middle).
5600 1. Added -C and -D options to the exinext utility, mainly to make it easier
5601 to include in the automated testing, but these could be helpful when
5602 multiple configurations are in use.
5604 2. The exinext utility was not formatting the output nicely when there was
5605 an alternate port involved in the retry record key, nor when there was a
5606 message id as well (for retries that were specific to a specific message
5607 and a specific host). It was also confused by IPv6 addresses, because of
5608 the additional colons they contain. I have fixed the IPv4 problem, and
5609 patched it up to do a reasonable job for IPv6.
5611 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
5612 delivery, the log line now contains "pipelined" if PIPELINING was used.
5614 4. An SMTP transport process used to panic and die if the bind() call to set
5615 an explicit outgoing interface failed. This has been changed; it is now
5616 treated in the same way as a connect() failure.
5618 5. A reference to $sender_host_name in the part of a conditional expansion
5619 that was being skipped was still causing a DNS lookup. This no longer
5622 6. The def: expansion condition was not recognizing references to header lines
5623 that used bh_ and bheader_.
5625 7. Added the _cache feature to named lists.
5627 8. The code for checking quota_filecount in the appendfile transport was
5628 allowing one more file than it should have been.
5630 9. For compatibility with Sendmail, the command line option
5638 and sets the incoming protocol and host name (for trusted callers). The
5639 host name and its colon can be omitted when only the protocol is to be set.
5640 Note the Exim already has two private options, -pd and -ps, that refer to
5641 embedded Perl. It is therefore impossible to set a protocol value of "d" or
5642 "s", but I don't think that's a major issue.
5644 10. A number of refactoring changes to the code, none of which should affect
5647 (a) The number of logging options was getting close to filling up the
5648 32-bit word that was used as a bit map. I have split them into two classes:
5649 those that are passed in the argument to log_write(), and those that are
5650 only ever tested independently outside of that function. These are now in
5651 separate 32-bit words, so there is plenty of room for expansion again.
5652 There is no change in the user interface or the logging behaviour.
5654 (b) When building, for example, log lines, the code previously used a
5655 macro that called string_cat() twice, in order to add two strings. This is
5656 not really sufficiently general. Furthermore, there was one instance where
5657 it was actually wrong because one of the argument was used twice, and in
5658 one call a function was used. (As it happened, calling the function twice
5659 did not affect the overall behaviour.) The macro has been replaced by a
5660 function that can join an arbitrary number of extra strings onto a growing
5663 (c) The code for expansion conditions now uses a table and a binary chop
5664 instead of a serial search (which was left over from when there were very
5665 few conditions). Also, it now recognizes conditions like "pam" even when
5666 the relevant support is not compiled in: a suitably worded error message is
5667 given if an attempt is made to use such a condition.
5669 11. Added ${time_interval:xxxxx}.
5671 12. A bug was causing one of the ddress fields not to be passed back correctly
5672 from remote delivery subprocesses. The field in question was not being
5673 subsequently used, so this caused to problems in practice.
5675 13. Added new log selectors queue_time and deliver_time.
5677 14. Might have fixed a bug in maildirsizefile handling that threw up
5678 "unexpected character" debug warnings, and recalculated the data
5679 unnecessarily. In any case, I expanded the warning message to give more
5682 15. Added the message "Restricted characters in address" to the statements in
5683 the default ACL that block characters like @ and % in local parts.
5685 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
5686 Three changes have been made:
5688 (a) There was a serious bug; a negative response to MAIL caused the whole
5689 recipient domain to be cached as invalid, thereby blocking all messages
5690 to all local parts at the same domain, from all senders. This bug has
5691 been fixed. The domain is no longer cached after a negative response to
5692 MAIL if the sender used is not empty.
5694 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
5697 (c) A new callout option, "use_sender" has been added for people who want
5698 the modified behaviour.
5704 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
5707 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
5708 indeed breaks things for older releases.
5710 3. Added additional logging to the case where there is a problem reading data
5711 from a filter that is running in a subprocess using a pipe, in order to
5712 try to track down a specific problem.
5714 4. Testing facility fudge: when running in the test harness and attempting
5715 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
5716 getting "No route to host". Convert this to a timeout.
5718 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
5721 6. Some OS don't have socklen_t but use size_t instead. This affects the
5722 fifth argument of getsockopt() amongst other things. This is now
5723 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
5724 can be set for individual OS. I have set it for SunOS5, OSF1, and
5725 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
5726 some earlier ones do not.
5728 7. Change 4.30/15 was not doing the test caselessly.
5730 8. The standard form for an IPv6 address literal was being rejected by address
5731 parsing in, for example, MAIL and RCPT commands. An example of this kind of
5732 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5733 this, as well as the form without the "IPv6" on the front (but only when
5734 address literals are enabled, of course).
5736 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
5738 10. Exim crashed if a message with an empty sender address specified by -f
5739 encountered a router with an errors_to setting. This could be provoked only
5740 by a command such as
5744 where an empty string was supplied; "<>" did not hit this bug.
5746 11. Installed PCRE release 4.5.
5748 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
5749 remained set. It is now erased.
5751 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
5752 times from message ids (which are base 36 rather than the normal 62).
5754 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
5755 were being counted as actual protocol errors, and logged if the log
5756 selector +smtp_protocol_error was set. One cannot be perfect in this test,
5757 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
5758 and DATA following a set of rejected RCPTs do not count as protocol errors.
5759 In other words, Exim assumes they were pipelined, though this may not
5760 actually be the case. Of course, in all cases the client gets an
5761 appropriate error code.
5763 15. If a lookup fails in an ACL condition, a message about the failure may
5764 be available; it is used if testing the ACL cannot continue, because most
5765 such messages specify what the cause of the deferral is. However, some
5766 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
5767 that caused an old message to be retained and used if a later statement
5768 caused a defer, replacing the real cause of the deferral.
5770 16. If an IP address had so many PTR records that the DNS lookup buffer
5771 was not large enough to hold them, Exim could crash while trying to process
5772 the truncated data. It now detects and logs this case.
5774 17. Further to 4.21/58, another change has been made: if (and only if) the
5775 first line of a message (the first header line) ends with CRLF, a bare LF
5776 in a subsequent header line has a space inserted after it, so as not to
5777 terminate the header.
5779 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
5780 unnecessarily, used atoi() instead of strtol(), and didn't check the
5781 termination when getting file sizes from file names by regex.
5783 19. Completely re-implemented the support for maildirsize files, in the light
5784 of a number of problems with the previous contributed implementation
5785 (4.30/29). In particular:
5787 . If the quota is zero, the maildirsize file is maintained, but no quota is
5790 . If the maildir directory does not exist, it is created before any attempt
5791 to write a maildirsize file.
5793 . The quota value in the file is just a cache; if the quota is changed in
5794 the transport, the new value overrides.
5796 . A regular expression is available for excluding directories from the
5799 20. The autoreply transport checks the characters in options that define the
5800 message's headers; it allows continued headers, but it was checking with
5801 isspace() after an embedded newline instead of explicitly looking for a
5804 21. If all the "regular" hosts to which an address was routed had passed their
5805 expiry times, and had not reached their retry times, the address was
5806 bounced, even if fallback hosts were defined. Now Exim should go on to try
5809 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
5810 equivalent code in the SMTP transport. Some hosts send humungous responses
5811 to HELO/EHLO, more than 1024 it seems.
5813 23. Refactoring: code in filter.c used (void *) for "any old type" but this
5814 gives compiler warnings in some environments. I've now done it "properly",
5817 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
5818 (because of problems with the built-in one) was declared to return uschar *
5819 instead of char *, causing compiler failure.
5821 25. Fixed a file descriptor leak when processing alias/forward files.
5823 26. Fixed a minor format string issue in dbfn.c.
5825 27. Typo in exim.c: ("dmbnz" for "dbmnz").
5827 28. If a filter file refered to $h_xxx or $message_headers, and the headers
5828 contained RFC 2047 "words", Exim's memory could, under certain conditions,
5831 29. When a sender address is verified, it is cached, to save repeating the test
5832 when there is more than one recipient in a message. However, when the
5833 verification involves a callout, it is possible for different callout
5834 options to be set for different recipients. It is too complicated to keep
5835 track of this in the cache, so now Exim always runs a verification when a
5836 callout is required, relying on the callout cache for the optimization.
5837 The overhead is duplication of the address routing, but this should not be
5840 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
5841 to be verified with callout=postmaster, and the main callout worked but the
5842 postmaster check failed, the verification correctly failed. However, if a
5843 subsequent RCPT command asked for sender verification *without* the
5844 postmaster check, incorrect caching caused this verification also to fail,
5847 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
5848 it was not caching the DNS options (qualify_single, search_parents) that
5849 were used when the lookup failed. A subsequent lookup with different
5850 options therefore always gave the same answer, though there were cases
5851 where it should not have. (Example: a "domains = !$mx_any" option on a
5852 dnslookup router: the "domains" option is always processed without any
5853 widening, but the router might have qualify_single set.) Now Exim uses the
5854 cached value only when the same options are set.
5856 32. Added John Jetmore's "exipick" utility to the distribution.
5858 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
5859 than a timeout (e.g. a certificate is required, and is not provided), an
5860 Exim server now closes the connection immediately. Previously it waited for
5861 the client to close - but if the client is SSL, it seems that they each
5862 wait for each other, leading to a delay before one of them times out.
5864 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
5865 maintained 0.8.x compatibility because I don't think many are using it, and
5866 it is clearly obsolete.
5868 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
5871 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
5872 client certificate was expired. A simple patch fixes this, though I don't
5873 understand the full logic of why the verify callback is called multiple
5876 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
5877 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
5878 which causes problems with some clients (such as the Certicom SSL Plus
5879 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
5880 disables the coutermeasure allowing Eudora to connect."
5882 38. Exim was not checking that a write() to a log file succeeded. This could
5883 lead to Bad Things if a log got too big, in particular if it hit a file
5884 size limit. Exim now panics and dies if it cannot write to a log file, just
5885 as it does if it cannot open a log file.
5887 39. Modified OS/Makefile-Linux so that it now contains
5889 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
5891 The two -D definitions ensure that Exim is compiled with large file
5892 support, which makes it possible to handle log files that are bigger than
5895 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
5896 instance) a domain was checked against a named list that involved a lookup,
5897 causing $domain_data to be set, then another domain was checked against the
5898 same list, then the first domain was re-checked, the value of $domain_data
5899 after the final check could be wrong. In particular, if the second check
5900 failed, it could be set empty. This bug probably also applied to
5903 41. The strip_trailing_dot option was not being applied to the address given
5904 with the -f command-line option.
5906 42. The code for reading a message's header from the spool was incrementing
5907 $received_count, but never initializing it. This meant that the value was
5908 incorrect (doubled) while delivering a message in the same process in which
5909 it was received. In the most common configuration of Exim, this never
5910 happens - a fresh exec is done - but it can happen when
5911 deliver_drop_privilege is set.
5913 43. When Exim logs an SMTP synchronization error - client data sent too soon -
5914 it now includes up to 150 characters of the unexpected data in the log
5917 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
5918 and building data strings. The size of both of these buffers was 10 000
5919 bytes - far larger than anybody would *ever* want, thought I. Needless to
5920 say, somebody hit the limit. I have increased the maximum line length to
5921 20 000 and the maximum data length of concatenated lines to 100 000. I have
5922 also fixed two bugs, because there was no checking on these buffers. Tsk,
5923 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
5924 buffer is too small.
5926 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
5927 lsearch lookups. Now it does.
5929 46. When parsing a route_list item in a manualroute router, a fixed-length
5930 buffer was used for the list of hosts. I made this 1024 bytes long,
5931 thinking that nobody would ever have a list of hosts that long. Wrong.
5932 Somebody had a whole pile of complicated expansion conditions, and the
5933 string was silently truncated, leading to an expansion error. It turns out
5934 that it is easier to change to an unlimited length (owing to other changes
5935 that have happened since this code was originally written) than to build
5936 structure for giving a limitation error. The length of the item that
5937 expands into the list of hosts is now unlimited.
5939 47. The lsearch lookup could not handle data where the length of text line was
5940 more than 4095 characters. Such lines were truncated, leading to shortened
5941 data being returned. It should now handle lines of any length.
5943 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
5944 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
5947 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
5948 replacing strings such as BIN_DIRECTORY in a source file: the replacement
5949 no longer happens in comment lines. A list of replacements is now placed
5950 at the head of all of the source files, except those whose only change is
5951 to replace PERL_COMMAND in the very first #! line.
5953 50. Replaced the slow insertion sort in queue.c, for sorting the list of
5954 messages on the queue, with a bottom-up merge sort, using code contributed
5955 by Michael Haardt. This should make operations like -bp somewhat faster on
5956 large queues. It won't affect queue runners, except when queue_run_in_order
5959 51. Installed eximstats 1.31 in the distribution.
5961 52. Added support for SRV lookups to the dnslookup router.
5963 53. If an ACL referred to $message_body or $message_body_end, the value was not
5964 reset for any messages that followed in the same SMTP session.
5966 54. The store-handling optimization for building very long strings was not
5967 differentiating between the different store pools. I don't think this
5968 actually made any difference in practice, but I've tidied it.
5970 55. While running the routers to verify a sender address, $sender_address
5971 was still set to the sender address. This is wrong, because when routing to
5972 send a bounce to the sender, it would be empty. Therefore, I have changed
5973 it so that, while verifying a sender address, $sender_address is set to <>.
5974 (There is no change to what happens when verifying a recipient address.)
5976 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
5977 target A or AAAA records (if not already returned) without resetting the
5978 qualify_single or search_parents options of the DNS resolver. These are
5979 inappropriate in this case because the targets of MX and SRV records must
5980 be FQDNs. A broken DNS record could cause trouble if it happened to have a
5981 target that, when qualified, matched something in the local domain. These
5982 two options are now turned off when doing these lookups.
5984 57. It seems that at least some releases of Reiserfs (which does not have the
5985 concept of a fixed number of inodes) returns zero and not -1 for the
5986 number of available inodes. This interacted badly with check_spool_inodes,
5987 which assumed that -1 was the "no such thing" setting. What I have done is
5988 to check that the total number of inodes is greater than zero before doing
5989 the test of how many are available.
5991 58. When a "warn" ACL statement has a log_message modifier, the message is
5992 remembered, and not repeated. This is to avoid a lot of repetition when a
5993 message has many recipients that cause the same warning to be written.
5994 However, Exim was preserving the list of already written lines for an
5995 entire SMTP session, which doesn't seem right. The memory is now reset if a
5996 new message is started.
5998 59. The "rewrite" debugging flag was not showing the result of rewriting in the
5999 debugging output unless log_rewrite was also set.
6001 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
6002 of (int)(handle) when we know that handle contains (void *)(-1).
6004 61. The Exim daemon panic-logs an error return when it closes the incoming
6005 connection. However "connection reset by peer" seems to be common, and
6006 isn't really an error worthy of noting specially, so that particular error
6009 62. When Exim is trying to find all the local interfaces, it used to panic and
6010 die if the ioctl to get the interface flags failed. However, it seems that
6011 on at least one OS (Solaris 9) it is possible to have an interface that is
6012 included in the list of interfaces, but for which you get a failure error
6013 for this call. This happens when the interface is not "plumbed" into a
6014 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
6015 failure of the "get flags" call assumes that the interface is down.
6017 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
6018 makes life easier for people who are doing arithmetic on fields extracted
6019 from dates, where you often get leading zeros that should not be
6020 interpreted as octal.
6022 64. Added qualify_domain to the redirect router, to override the global
6025 65. If a pathologically long header line contained very many addresses (the
6026 report of this problem mentioned 10 000) and each of them was rewritten,
6027 Exim could use up a very large amount of memory. (It kept on making new
6028 copies of the header line as it rewrote, and never released the old ones.)
6029 At the expense of a bit more processing, the header rewriting function has
6030 been changed so that it no longer eats memory in this way.
6032 66. The generation of the Received: header has been moved from the time that a
6033 message starts to be received, to the time that it finishes. The timestamp
6034 in the Received: header should now be very close to that of the <= log
6035 line. There are two side-effects of this change:
6037 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
6038 logged header lines no longer include the local Received: line, because
6039 it has not yet been created. The same applies to a copy of the message
6040 that is returned to a non-SMTP sender when a message is rejected.
6042 (b) When a filter file is tested using -bf, no additional Received: header
6043 is added to the test message. After some thought, I decided that this
6046 This change does not affect the value of $received_for. It is still set
6047 after address rewriting, but before local_scan() is called.
6049 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
6051 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
6052 gave an unhelpful panic error message, and a defer error. I have managed to
6053 change this behaviour so that it now rejects any supplied certificate,
6054 which seems right, as the list of acceptable certificates is empty.
6056 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
6057 gave an unhelpful defer error. I have not managed to make this reject any
6058 supplied certificates, but the error message it gives is "no certificate
6059 supplied", which is not helpful.
6061 70. exigrep's output now also includes lines that are not associated with any
6062 message, but which match the given pattern. Implemented by a patch from
6063 Martin Sluka, which also tidied up the Perl a bit.
6065 71. Recipient callout verification, like sender verification, was using <> in
6066 the MAIL FROM command. This isn't really the right thing, since the actual
6067 sender may affect whether the remote host accepts the recipient or not. I
6068 have changed it to use the actual sender in the callout; this means that
6069 the cache record is now keyed on a recipient/sender pair, not just the
6070 recipient address. There doesn't seem to be a real danger of callout loops,
6071 since a callout by the remote host to check the sender would use <>.
6072 [SEE ABOVE: changed after hitting problems.]
6074 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
6075 temporary errors. However, in the case of such a code being given after
6076 the end of a data transmission (i.e. after ".") Exim was failing to write
6077 a retry record for the message. (Yes, there was some broken host that was
6078 actually sending 8xx at this point.)
6080 73. An unknown lookup type in a host list could cause Exim to panic-die when
6081 the list was checked. (An example that provoked this was putting <; in the
6082 middle of a list instead of at the start.) If this happened during a DATA
6083 ACL check, a -D file could be left lying around. This kind of configuration
6084 error no longer causes Exim to die; instead it causes a defer error. The
6085 incident is still logged to the main and panic logs.
6087 74. Buglet left over from Exim 3 conversion. The message "too many messages
6088 in one connection" was written to the rejectlog but not the mainlog, except
6089 when address rewriting (yes!) was being logged.
6091 75. Added write_rejectlog option.
6093 76. When a system filter was run not as root (that is, when system_filter_user
6094 was set), the values of the $n variables were not being returned to the
6095 main process; thus, they were not subsequently available in the $sn
6098 77. Added +return_path_on_delivery log selector.
6100 78. A connection timeout was being treated differently from recipients deferred
6101 when testing hosts_max_try with a message that was older than the host's
6102 retry timeout. (The host should not be counted, thus allowing all hosts to
6103 be tried at least once before bouncing.) This may have been the cause of an
6104 occasionally reported bug whereby a message would remain on the queue
6105 longer than the retry timeout, but would be bounced if a delivery was
6106 forced. I say "may" because I never totally pinned down the problem;
6107 setting up timeout/retry tests is difficult. See also the next item.
6109 79. The ultimate address timeout was not being applied to errors that involved
6110 a combination of host plus message (for example, a timeout on a MAIL
6111 command). When an address resolved to a number of possible hosts, and they
6112 were not all tried for each delivery (e.g. because of hosts_max_try), a
6113 message could remain on the queue longer than the retry timeout.
6115 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
6118 81. Fixed an obscure SMTP outgoing bug which required at least the following
6119 conditions: (a) there was another message waiting for the same server;
6120 (b) the server returned 5xx to all RCPT commands in the first message so
6121 that the message was not completed; (c) the server dropped the connection
6122 or gave a negative response to the RSET that Exim sends to abort the
6123 transaction. The observed case was a dropped connection after DATA that had
6124 been sent in pipelining mode. That is, the server had advertised PIPELINING
6125 but was not implementing it correctly. The effect of the bug was incorrect
6126 behaviour, such as trying another host, and this could lead to a crash.
6132 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
6133 and daemon.c were passed as pointers to ints; they should have been
6134 pointers to socklen_t variables (which are typically unsigned ints).
6136 2. Some signed/unsigned type warnings in the os.c file for Linux have been
6139 3. Fixed a really odd bug that affected only the testing scheme; patching a
6140 certain fixed string in the binary changed the value of another string that
6141 happened to be identical to the end of the original first string.
6143 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
6144 name", it returns that address as the IP address. On some operating
6145 systems (e.g. Solaris), it also passes back the IP address string as the
6146 "host name". However, on others (e.g. Linux), it passes back an empty
6147 string. Exim wasn't checking for this, and was changing the host name to an
6148 empty string, assuming it had been canonicalized.
6150 5. Although rare, it is permitted to have more than one PTR record for a given
6151 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
6152 all the names associated with an address, because they do in Solaris.
6153 However, it seems that they do not in Linux for data that comes from the
6154 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
6155 I found this out when I moved to a new Linux workstation and tried to run
6156 the Exim test suite.
6158 To get round this problem I have changed the code so that it now does its
6159 own call to the DNS to look up PTR records when searching for a host name.
6160 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
6161 addresses that are only in /etc/hosts are still found.
6163 This behaviour is, however, controlled by an option called host_lookup_
6164 order, which defaults to "bydns:byaddr". If people want to use the other
6165 order, or indeed, just use one or the other means of lookup, they can
6166 specify it in this variable.
6168 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
6169 some operating systems, this comes back from gethostbyaddr() as an empty
6170 string, and this is what Exim used to test for. However, it seems that in
6171 other systems, "." is yielded. Exim now tests for this case too.
6173 7. The values of check_spool_space and check_log_space are now held internally
6174 as a number of kilobytes instead of an absolute number of bytes. If a
6175 numbers is specified without 'K' or 'M', it is rounded up to the nearest
6176 kilobyte. This means that much larger values can be stored.
6178 8. Exim monitor: an attempt to get the action menu when not actually pointing
6179 at a message produces an empty menu entitled "No message selected". This
6180 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
6181 no entries in it ("Shell widget menu has zero width and/or height"). So I
6182 have added a single, blank menu entry in this case.
6184 9. Added ${quote_local_part.
6186 10. MIME decoding is now applied to the contents of Subject: header lines when
6189 11. Now that a reference to $sender_host_address automatically causes a reverse
6190 lookup to occur if necessary (4.13/18), there is no need to arrange for a
6191 host lookup before query-style lookups in lists that might use this
6192 variable. This has therefore been abolished, and the "net-" prefix is no
6193 longer necessary for query-style lookups.
6195 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
6196 have been a typo for LFLAGS, so it has been changed.
6198 13. The install script calls Exim with "-C /dev/null" in order to find the
6199 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
6200 to be output. However, since Exim outputs its version number before the
6201 error, it didn't break the script. It just looked ugly. I fixed this by
6202 always allowing "-C /dev/null" if the caller is root.
6204 14. Ignore overlarge ACL variable number when reading spool file - insurance
6205 against a later release with more variables having written the file.
6207 15. The standard form for an IPv6 address literal was being rejected by EHLO.
6208 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
6209 this, as well as the form without the "IPv6" on the front.
6211 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
6212 OS/Makefile-Darwin file.
6214 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
6215 only to LDAP libraries that do not have LDAP_OPT_DEREF.
6217 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
6218 variable. However, some OS use ints for inodes. I've added cast to long int
6219 to get rid of the compiler warning.
6221 19. I had forgotten to lock out "/../" in configuration file names when
6222 ALT_CONFIG_PREFIX was set.
6224 20. Routers used for verification do not need to specify transports. However,
6225 if such a router generated a host list, and callout was configured, Exim
6226 crashed, because it could not find a port number from the (non-existent)
6227 transport. It now assumes port 25 in this circumstance.
6229 21. Added the -t option to exigrep.
6231 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
6232 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
6233 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
6234 them is set, LOOKUP_LSEARCH is forced.
6236 23. "exim -bV" now outputs a list of lookups that are included in the binary.
6238 24. Added sender and host information to the "rejected by local_scan()" log
6239 line; previously there was no indication of these.
6241 25. Added .include_if_exists.
6243 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
6244 Linux. It turns out that not all file systems support this. Apparently some
6245 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
6246 do it.) To cope with this, the error EINVAL, which means that sync-ing is
6247 not supported on the file descriptor, is now ignored when Exim is trying to
6248 sync a directory. This applies only to Linux.
6250 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
6252 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
6253 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
6256 29. Added support for maildirsize files from supplied patch (modified a bit).
6258 30. The use of :fail: followed by an empty string could lead Exim to respond to
6259 sender verification failures with (e.g.):
6261 550 Verification failed for <xxx>
6262 550 Sender verify failed
6264 where the first response line was missing the '-' that indicates it is not
6265 the final line of the response.
6267 31. The loop for finding the name of the user that called Exim had a hardwired
6268 limit of 10; it now uses the value of finduser_retries, which is used for
6269 all other user lookups.
6271 32. Added $received_count variable, available in data and not_smtp ACLs, and at
6274 33. Exim was neglecting to zero errno before one call of strtol() when
6275 expanding a string and expecting an integer value. On some systems this
6276 resulted in spurious "integer overflow" errors. Also, it was casting the
6277 result into an int without checking.
6279 34. Testing for a connection timeout using "timeout_connect" in the retry rules
6280 did not work. The code looks as if it has *never* worked, though it appears
6281 to have been documented since at least release 1.62. I have made it work.
6283 35. The "timeout_DNS" error in retry rules, also documented since at least
6284 1.62, also never worked. As it isn't clear exactly what this means, and
6285 clearly it isn't a major issue, I have abolished the feature by treating it
6286 as "timeout", and writing a warning to the main and panic logs.
6288 36. The display of retry rules for -brt wasn't always showing the error code
6291 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
6292 timeout_connect_A, timeout_connect_MX.
6294 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
6295 to the empty sender.
6297 39. The daemon was not analysing the content of -oX till after it had closed
6298 stderr and disconnected from the controlling terminal. This meant that any
6299 syntax errors were only noted on the panic log, and the return code from
6300 the command was 0. By re-arranging the code a little, I've made the
6301 decoding happen first, so such errors now appear on stderr, and the return
6302 code is 1. However, the actual setting up of the sockets still happens in
6303 the disconnected process, so errors there are still only recorded on the
6306 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
6307 connections (as happens on some IP stacks) was logged at start up time as
6308 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
6309 it from "IPv6 and IPv4", which means that two separate sockets are being
6312 41. The debug output for gethostbyname2() or getipnodebyname() failures now
6313 says whether AF_INET or AF_INET6 was passed as an argument.
6315 42. Exiwhat output was messed up when time zones were included in log
6318 43. Exiwhat now gives more information about the daemon's listening ports,
6319 and whether -tls-on-connect was used.
6321 44. The "port" option of the smtp transport is now expanded.
6323 45. A "message" modifier in a "warn" statement in a non-message ACL was being
6324 silently ignored. Now an error message is written to the main and panic
6327 46. There's a new ACL modifier called "logwrite" which writes to a log file
6328 as soon as it is encountered.
6330 47. Added $local_user_uid and $local_user_gid at routing time.
6332 48. Exim crashed when trying to verify a sender address that was being
6335 49. Exim was recognizing only a space character after ".include". It now also
6336 recognizes a tab character.
6338 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
6339 extracting the relevant information from the specification. The man page no
6340 longer contains scrambled data for the -d option, and I've added a section
6341 at the front about calling Exim under different names.
6343 51. Added "extra_headers" argument to the "mail" command in filter files.
6345 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
6348 53. Installed eximstats 1.29.
6350 54. Added transport_filter_timeout as a generic transport option.
6352 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
6353 Cc: header lines. This was required by RFC 822, but it not required by RFC
6356 56. Exim used to add From:, Date:, and Message-Id: header lines to any
6357 incoming messages that did not have them. Now it does so only if the
6358 message originates locally, that is, if there is no associated remote host
6359 address. When Resent- header lines are present, this applies to the Resent-
6360 lines rather than the non-Resent- lines.
6362 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
6363 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
6365 58. Messages for configuration errors now include the name of the main
6366 configuration file - useful now that there may be more than one file in a
6367 list (.included file names were always shown).
6369 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
6370 for those rare installations that do not start the daemon as root or run it
6371 setuid root. I've cut out the call to initgroups() if the daemon is not
6374 60. The Exim user and group can now be bound into the binary as text strings
6375 that are looked up at the start of Exim's processing.
6377 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
6379 62. Added $mailstore_basename variable.
6381 63. Installed patch to sieve.c from Michael Haardt.
6383 64. When Exim failed to open the panic log after failing to open the main log,
6384 the original message it was trying to log was written to stderr and debug
6385 output, but if they were not available (the usual case in production), it
6386 was lost. Now it is written to syslog before the two lines that record the
6387 failures to open the logs.
6389 65. Users' Exim filters run in subprocesses under the user's uid. It is
6390 possible for a "deliver" command or an alias in a "personal" command to
6391 provoke an address rewrite. If logging of address rewriting is configured,
6392 this fails because the process is not running as root or exim. There may be
6393 a better way of dealing with this, but for the moment (because 4.30 needs
6394 to be released), I have disabled address rewrite logging when running a
6395 filter in a non-root, non-exim process.
6401 1. The buildconfig auxiliary program wasn't quoting the value set for
6402 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
6403 not defined. This bug was masked in 4.22 by the effect that was fixed in
6406 2. Some messages that were rejected after a message id was allocated were
6407 shown as "incomplete" by exigrep. It no longer does this for messages that
6408 are rejected by local_scan() or the DATA or non-SMTP ACLs.
6410 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
6411 have allow_domain_literals set, the ID did not get logged in the <= line.
6412 Domain literals are now always recognized in Message-ID: header lines.
6414 4. The first argument for a ${extract expansion item is the key name or field
6415 number. Leading and trailing spaces in this item were not being ignored,
6416 causing some misleading effects.
6418 5. When deliver_drop_privilege was set, single queue runner processes started
6419 manually (i.e. by the command "exim -q") or by the daemon (which uses the
6420 same command in the process it spins off) were not dropping privilege.
6422 6. When the daemon running as "exim" started a queue runner, it always
6423 re-executed Exim in the spun-off process. This is a waste of effort when
6424 deliver_drop_privilege is set. The new process now just calls the
6425 queue-runner function directly.
6431 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
6434 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
6435 ignored. Though the use of -oP was forcing the writing of a pid file, it
6436 was always written to the default place.
6438 3. If the message "no IP address found for host xxxx" is generated during
6439 incoming verification, it is now followed by identification of the incoming
6440 connection (so you can more easily find what provoked it).
6442 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
6444 5. Added some features to "harden" Exim a bit more against certain attacks:
6446 (a) There is now a build-time option called FIXED_NEVER_USERS that can
6447 be put in Local/Makefile. This is like the never_users runtime option,
6448 but it cannot be overridden. The default setting is "root".
6450 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
6451 prefix string with which any file named in a -C command line option
6454 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
6455 is retained for -C and -D only if the caller of Exim is root. Without
6456 it, the exim user may also use -C and -D and retain privilege.
6458 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
6459 command line option is disabled.
6461 6. Macro names set by the -D option must start with an upper case letter, just
6462 like macro names defined in the configuration file.
6464 7. Added "dereference=" facility to LDAP.
6466 8. Two instances of the typo "uknown" in the source files are fixed.
6468 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
6469 the Configure-Makefile script screwed up while processing it.
6471 10. Incorporated PCRE 4.4.
6473 11. The SMTP synchronization check was not operating right at the start of an
6474 SMTP session. For example, it could not catch a HELO sent before the client
6475 waited for the greeting. There is now a check for outstanding input at the
6476 point when the greeting is written. Because of the duplex, asynchronous
6477 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
6478 way, but not yet received, when the check is performed.
6480 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
6481 on TCP/IP sockets, because this apparently causes some broken clients to
6484 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
6485 unchanged) from the Cygwin maintainer.
6487 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
6488 was supported instead of testing for mbx. Effectively a typo.
6490 15. The spa authenticator server code was not checking that the input it
6491 received was valid base64.
6493 16. The debug output line for the "set" modifier in ACLs was not showing the
6494 name of the variable that was being set.
6496 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
6498 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
6499 lookup on its own. Something else was needed to trigger the lookup. For
6500 example, a match in host_lookup or the need for a host name in a host list.
6501 Now, if $sender_host_name is referenced and the host name has not yet been
6502 looked up, a lookup is performed. If the lookup fails, the variable remains
6503 empty, and $host_lookup_failed is set to "1".
6505 19. Added "eqi" as a case-independent comparison operator.
6507 20. The saslauthd authentication condition could segfault if neither service
6508 nor realm was specified.
6510 21. If an overflowing value such as "2048M" was set for message_size_limit, the
6511 error message that was logged was misleading, and incoming SMTP
6512 connections were dropped. The message is now more accurate, and temporary
6513 errors are given to SMTP connections.
6515 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
6516 (except RSET) with a 421 error, until QUIT is received. However, it was
6517 failing to send a response to QUIT.
6519 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
6520 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
6521 ACL is now run after the helo_try_verify_hosts code.
6523 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
6524 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
6525 case-independent, so other case variants are also recognized). Apparently
6526 some systems use these upper case variants.
6528 25. If more than two messages were waiting for the same host, and a transport
6529 filter was specified for the transport, Exim sent two messages over the
6530 same TCP/IP connection, and then failed with "socket operation on non-
6531 socket" when it tried to send the third.
6533 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
6535 27. The extern definition of crypt16() in expand.c was not being excluded when
6536 the OS had its own crypt16() function.
6538 28. Added bounce_return_body as a new option, and bounce_return_size_limit
6539 as a preferred synonym for return_size_limit, both as an option and as an
6542 29. Added LIBS=-liconv to OS/Makefile-OSF1.
6544 30. Changed the default configuration ACL to relax the local part checking rule
6545 for addresses that are not in any local domains. For these addresses,
6546 slashes and pipe symbols are allowed within local parts, but the sequence
6547 /../ is explicitly forbidden.
6549 31. SPA server authentication was not clearing the challenge buffer before
6552 32. log_message in a "warn" ACL statement was writing to the reject log as
6553 well as to the main log, which contradicts the documentation and doesn't
6554 seem right (because no rejection is happening). So I have stopped it.
6556 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
6557 However, I am unable to do any testing of this.
6559 34. Fixed an infelicity in the appendfile transport. When checking directories
6560 for a mailbox, to see if any needed to be created, it was accidentally
6561 using path names with one or more superfluous leading slashes; tracing
6562 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
6564 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
6565 contents are added to the log line that is written for every discarded
6566 recipient. (Previously a log_message setting was ignored.)
6568 36. The ${quote: operator now quotes the string if it is empty.
6570 37. The install script runs exim in order to find its version number. If for
6571 some reason other than non-existence or emptiness, which it checks, it
6572 could not run './exim', it was installing it with an empty version number,
6573 i.e. as "exim-". This error state is now caught, and the installation is
6576 38. An argument was missing from the function that creates an error message
6577 when Exim fails to connect to the socket for saslauthd authentication.
6578 This could cause Exim to crash, or give a corrupted message.
6580 39. Added isip, isip4, and isip6 to ${if conditions.
6582 40. The ACL variables $acl_xx are now saved with the message, and can be
6583 accessed later in routers, transports, and filters.
6585 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
6586 strings in the file are not string-expanded.
6588 42. If a MAIL command specified a SIZE value that was too large to fit into an
6589 int variable, the check against message_size_limit failed. Such values are
6590 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
6591 day this will have to be increased, but I don't think I want to be around
6592 when emails are that large.
6599 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
6600 iconv() is not standard in FreeBSD.
6602 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
6603 IPv6 enabled. The observed symptom was a segmentation fault on return from
6604 the function os_common_find_running_interfaces() in src/os.c.
6606 3. In the check_special_case() function in daemon.c I had used "errno" as an
6607 argument name, which causes warnings on some systems. This was basically a
6608 typo, since it was named "eno" in the comments!
6610 4. The code that waits for the clock to tick (at a resolution of some fraction
6611 of a second) so as to ensure message-id uniqueness was always waiting for
6612 at least one whole tick, when it could have waited for less. [This is
6613 almost certainly not relevant at current processor speeds, where it is
6614 unlikely to ever wait at all. But we try to future-proof.]
6616 5. The function that sleeps for a time interval that includes fractions of a
6617 second contained a race. It did not block SIGALRM between setting the
6618 timer, and suspending (a couple of lines later). If the interval was short
6619 and the sigsuspend() was delayed until after it had expired, the suspension
6620 never ended. On busy systems this could lead to processes getting stuck for
6623 6. Some uncommon configurations may cause a lookup to happen in a queue runner
6624 process, before it forks any delivery processes. The open lookup caching
6625 mechanism meant that the open file or database connection was passed into
6626 the delivery process. The problem was that delivery processes always tidy
6627 up cached lookup data. This could cause a problem for the next delivery
6628 process started by the queue runner, because the external queue runner
6629 process does not know about the closure. So the next delivery process
6630 still has data in the lookup cache. In the case of a file lookup, there was
6631 no problem because closing a file descriptor in a subprocess doesn't affect
6632 the parent. However, if the lookup was caching a connection to a database,
6633 the connection was closed, and the second delivery process was likely to
6634 see errors such as "PGSQL: query failed: server closed the connection
6635 unexpectedly". The problem has been fixed by closing all cached lookups
6636 in a queue runner before running a delivery process.
6638 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
6639 seem to have the "const" qualifier which it has on other OS. I've
6642 8. Change 4.21/2 was too strict. It is only if there are two authenticators
6643 *of the same type* (client or server) with the same public name that an
6644 error should be diagnosed.
6646 9. When Exim looked up a host name for an IP address, but failed to find the
6647 original IP address when looking up the host name (a safety check), it
6648 output the message "<ip address> does not match any IP for NULL", which was
6649 confusing, to say the least. The bug was that the host name should have
6650 appeared instead of "NULL".
6652 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
6653 user that is built into the binary, and the -C or -D options is used, root
6654 privilege is dropped before the configuration file is read. In addition,
6655 logging is switched to stderr instead of the normal log files. If the
6656 configuration then re-defines the Exim user, the unprivileged environment
6657 is probably not what is expected, so Exim logs a panic warning message (but
6660 However, if deliver_drop_privilege is set, the unprivileged state may well
6661 be exactly what is intended, so the warning has been cut out in that case,
6662 and Exim is allowed to try to write to its normal log files.
6668 1. smtp_return_error_details was not giving details for temporary sender
6669 or receiver verification errors.
6671 2. Diagnose a configuration error if two authenticators have the same public
6674 3. Exim used not to create the message log file for a message until the first
6675 delivery attempt. This could be confusing when incoming messages were held
6676 for policy or load reasons. The message log file is now created at the time
6677 the message is received, and an initial "Received" line is written to it.
6679 4. The automatically generated man page for command line options had a minor
6680 bug that caused no ill effects; however, a more serious problem was that
6681 the procedure for building the man page automatically didn't always
6682 operate. Consequently, release 4.20 contains an out-of-date version. This
6683 shouldn't happen again.
6685 5. When building Exim with embedded Perl support, the script that builds the
6686 Makefile was calling 'perl' to find its compile-time parameters, ignoring
6687 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
6689 6. The freeze_tell option was not being used for messages that were frozen on
6690 arrival, either by an ACL or by local_scan().
6692 7. Added the smtp_incomplete_transaction log selector.
6694 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
6695 was accepting AUTH without a new EHLO.
6697 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
6698 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
6701 10. Logging of TCP/IP connections (when configured) now happens in the main
6702 daemon process instead of the child process, so that the TCP/IP connection
6703 count is more accurate (but it can never be perfect).
6705 11. The use of "drop" in a nested ACL was not being handled correctly in the
6706 outer ACL. Now, if condition failure induced by the nested "drop" causes
6707 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
6708 or "require"), the connection is dropped.
6710 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
6711 that yield "discard" can now be used with an "accept" or a "discard" verb,
6712 but an error is generated for any others (because I can't see a useful way
6713 to define what should happen).
6715 13. When an ACL is read dynamically from a file (or anywhere else), the lines
6716 are now processed in the same way as lines in the Exim configuration file.
6717 In particular, continuation lines are supported.
6719 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
6721 15. Added -ti meaning -t -i.
6723 16. Check for letters, digits, hyphens, and dots in the names of dnslist
6724 domains, and warn by logging if others are found.
6726 17. At least on BSD, alignment is not guaranteed for the array of ifreq's
6727 returned from GIFCONF when Exim is trying to find the list of interfaces on
6728 a host. The code in os.c has been modified to copy each ifreq to an aligned
6729 structure in all cases.
6731 Also, in some cases, the returned ifreq's were being copied to a 'struct
6732 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
6733 means the last couple of bytes of an IPv6 address could be chopped if the
6734 ifreq contained only a normal sockaddr (14 bytes storage).
6736 18. Named domain lists were not supported in the hosts_treat_as_local option.
6737 An entry such as +xxxx was not recognized, and was treated as a literal
6740 19. Ensure that header lines added by a DATA ACL are included in the reject log
6741 if the ACL subsequently rejects the message.
6743 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
6744 MD5 (which is deprecated).
6746 21. When testing a filter file using -bf, Exim was writing a message when it
6747 took the sender from a "From " line in the message, but it was not doing so
6748 when it took $return_path from a Return-Path: header line. It now does.
6750 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
6751 with a valid header line field name (a series of printing characters
6752 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
6754 23. Changed "disc" in the source to "disk" to conform to the documentation and
6755 the book and for uniformity.
6757 24. Ignore Sendmail's -Ooption=value command line item.
6759 25. When execve() failed while trying to run a command in a pipe transport,
6760 Exim was returning EX_UNAVAILABLE (69) from the subprocess. However, this
6761 could be confused with a return value of 69 from the command itself. This
6762 has been changed to 127, the value the shell returns if it is asked to run
6763 a non-existent command. The wording for the related log line suggests a
6764 non-existent command as the problem.
6766 26. If received_header_text expands to an empty string, do not add a Received:
6767 header line to the message. (Well, it adds a token one on the spool, but
6768 marks it "old" so that it doesn't get used or transmitted.)
6770 27. Installed eximstats 1.28 (addition of -nt option).
6772 28. There was no check for failure on the call to getsockname() in the daemon
6773 code. This can fail if there is a shortage of resources on the system, with
6774 ENOMEM, for example. A temporary error is now given on failure.
6776 29. Contrary to the C standard, it seems that in some environments, the
6777 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
6778 program. Exim now does this explicitly; it affects the formatting of
6779 timestamps using strftime().
6781 30. If exiqsumm was given junk data, it threw up some uninitialized variable
6782 complaints. I've now initialized all the variables, to avoid this.
6784 32. Header lines added by a system filter were not being "seen" during
6785 transport-time rewrites.
6787 33. The info_callback() function passed to OpenSSL is set up with type void
6788 (*)(SSL *, int, int), as described somewhere. However, when calling the
6789 function (actually a macro) that sets it up, the type void(*)() is
6790 expected. I've put in a cast to prevent warnings from picky compilers.
6792 34. If a DNS black list lookup found a CNAME record, but there were no A
6793 records associated with the domain it pointed at, Exim crashed.
6795 35. If a DNS black list lookup returned more than one A record, Exim ignored
6796 all but the first. It now scans all returned addresses if a particular IP
6797 value is being sought. In this situation, the contents of the
6798 $dnslist_value variable are a list of all the addresses, separated by a
6801 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
6802 to accept a host name and all its aliases if the forward lookup for any of
6803 them yielded the IP address of the incoming connection. Now it accepts only
6804 those names whose forward lookup yields the correct IP address. Any other
6805 names are discarded. This closes a loophole whereby a rogue DNS
6806 administrator could create reverse DNS records to break through a
6807 wildcarded host restriction in an ACL.
6809 37. If a user filter or a system filter that ran in a subprocess used any of
6810 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
6811 the wrong values were passed to the pipe command ($thisaddress had the
6812 value of $0, $0 had the value of $1, etc). This bug was introduced by
6813 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
6815 38. Improved the line breaking for long SMTP error messages from ACLs.
6816 Previously, if there was no break point between 40 and 75 characters, Exim
6817 left the rest of the message alone. Two changes have been made: (a) I've
6818 reduced the minimum length to 35 characters; (b) if it can't find a break
6819 point between 35 and 75 characters, it looks ahead and uses the first one
6820 that it finds. This may give the occasional overlong line, but at least the
6821 remaining text gets split now.
6823 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
6824 file descriptors might be low, and that setting 1000 would always raise it.
6825 It turns out that in some environments, the limit is already over 1000 and
6826 that lowering it causes trouble. So now Exim takes care not to decrease it.
6828 40. When delivering a message, the value of $return_path is set to $sender_
6829 address at the start of routing (routers may change the value). By an
6830 oversight, this default was not being set up when an address was tested by
6831 -bt or -bv, which affected the outcome if any router or filter referred to
6834 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
6835 log only when "message" or "log_message" are set. However, if one of the
6836 conditions was an address verification, or a call to a nested ACL, the
6837 messages generated by the underlying test were being passed through. This
6838 no longer happens. The underlying message is available in $acl_verify_
6839 message for both "message" and "log_message" expansions, so it can be
6840 passed through if needed.
6842 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
6843 new expansion $bh_ to give the encoded byte string without charset
6844 translation. Translation happens only if iconv() is available; HAVE_ICONV
6845 indicates this at build time. HEADERS_CHARSET gives the charset to
6846 translate to; headers_charset can change it in the configuration, and
6847 "headers charset" can change it in an individual filter file.
6849 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
6850 that creates RFC 2047 encoded "words" labels them as that charset instead
6851 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
6852 expansion operator; (ii) when Exim creates a From: line for a local
6853 message; (iii) when a header line is rewritten to include a "phrase" part.
6855 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
6856 buggy, causing it to skip the first lines of messages whose message ID
6857 ended in 'D'. This would not have bitten before Exim release 4.14, because
6858 message IDs were unlikely to end in 'D' before then. The effect was to have
6859 incorrect size information for certain domains.
6861 45. #include "config.h" was missing at the start of the crypt16.c module. This
6862 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
6865 46. If there was a timeout during a "random" callout check, Exim treated it as
6866 a failure of the random address, and carried on sending RSET and the real
6867 address. If the delay was just some slowness somewhere, the response to the
6868 original RCPT would be taken as a response to RSET and so on, causing
6869 mayhem of various kinds.
6871 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
6872 when I implemented it. It didn't allow for the fact that some option values
6873 may legitimately be negative (e.g. size_addition), and it didn't even do
6874 the right test for positive values.
6876 48. Domain names in DNS records are case-independent. Exim always looks them up
6877 in lower case. Some resolvers return domain names in exactly the case they
6878 appear in the zone file, that is, they may contain uppercase letters. Not
6879 all resolvers do this - some return always lower case. Exim was treating a
6880 change of case by a resolver as a change of domain, similar to a widening
6881 of a domain abbreviation. This triggered its re-routing code and so it was
6882 trying to route what was effectively the same domain again. This normally
6883 caused routing to fail (because the router wouldn't handle the domain
6884 twice). Now Exim checks for this case specially, and just changes the
6885 casing of the domain that it ultimately uses when it transmits the message
6888 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
6891 50. If a filter generated a file delivery with a non-absolute name (possible if
6892 no home directory exists for the router), the forbid_file option was not
6895 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
6896 the existing equality matching.
6898 52. Exim was using ints instead of ino_t variables in some places where it was
6899 dealing with inode numbers.
6901 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
6902 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
6903 called TMPDIR, and if it finds it is different, it changes its value.
6905 54. The smtp_printf() function is now made available to local_scan() so
6906 additional output lines can be written before returning. There is also an
6907 smtp_fflush() function to enable the detection of a dropped connection.
6908 The variables smtp_input and smtp_batched_input are exported to
6911 55. Changed the default runtime configuration: the message "Unknown user"
6912 has been removed from the ACL, and instead placed on the localuser router,
6913 using the cannot_route_message feature. This means that any verification
6914 failures that generate their own messages won't get overridden. Similarly,
6915 the "Unrouteable address" message that was in the ACL for unverifiable
6916 relay addresses has also been removed.
6918 56. Added hosts_avoid_esmtp to the smtp transport.
6920 57. The exicyclog script was not checking for the esoteric option
6921 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
6922 will work only if exicyclog is run under the appropriate euid.
6924 58. Following a discussion on the list, the rules by which Exim recognises line
6925 endings on incoming messages have been changed. The -dropcr and drop_cr
6926 options are now no-ops, retained only for backwards compatibility. The
6927 following line terminators are recognized: LF CRLF CR. However, special
6928 processing applies to CR:
6930 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
6931 nor a local message in the state where . is a terminator.
6933 (ii) If a bare CR is encountered in a header line, an extra space is added
6934 after the line terminator so as not to end the header. The reasoning
6935 behind this is that bare CRs in header lines are most likely either
6936 to be mistakes, or people trying to play silly games.
6938 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
6939 was being incorrectly given as 18 bytes larger than it should have been.
6940 This is a VOB (very old bug).
6942 60. This may never have affected anything current, but just in case it has:
6943 When the local host is found other than at the start of a list of hosts,
6944 the local host, those with the same MX, and any that follow, are discarded.
6945 When the list in question was part of a longer list of hosts, the following
6946 hosts (not currently being processed) were also being discarded. This no
6947 longer happens. I'm not sure if this situation could ever has previously
6950 61. Added the "/MX" feature to lists of hosts in the manualroute and query
6953 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
6954 header. This is something that is recommended in a new Internet Draft, and
6955 is something that is documented as being done by Sendmail. There are two
6956 possible values. For messages generated by the autoreply transport, Exim
6959 Auto-Submitted: auto-replied
6961 whereas for all other generated messages (e.g. bounces) it adds
6963 Auto-Submitted: auto-generated
6965 63. The "personal" condition in filters now includes a test for the
6966 Auto-Submitted: header. If it contains the string "auto-" the message it
6967 not considered personal.
6969 64. Added rcpt_include_affixes as a generic transport option.
6971 65. Added queue_only_override (default true).
6973 66. Added the syslog_duplication option.
6975 67. If what should have been the first header line of a message consisted of
6976 a space followed by a colon, Exim was mis-interpreting it as a header line.
6977 It isn't of course - it is syntactically invalid and should therefore be
6978 treated as the start of the message body. The misbehaviour could have
6979 caused a number of strange effects, including loss of data in subsequent
6980 header lines, and spool format errors.
6982 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
6983 client host had authenticated. This control can now be exercised by an ACL
6984 for more flexibility.
6986 69. By default, callouts do not happen when testing with -bh. There is now a
6987 variant, -bhc, which does actually run the callout code, including
6988 consulting and updating the callout cache.
6990 70. Added support for saslauthd authentication, courtesy of Alexander
6993 71. If statvfs() failed on the spool or log directories while checking their
6994 size for availability, Exim confusingly gave the error "space shortage".
6995 Furthermore, in debugging mode it crashed with a floating point exception.
6996 These checks are done if check_{spool,log}_{space,inodes} are set, and when
6997 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
6998 serious problem, Exim now writes to the main and panic logs when this
6999 happens, with details of the failure. It then refuses to accept the
7000 incoming message, giving the message "spool directory problem" or "log
7001 directory problem" with a 421 code for SMTP messages.
7003 72. When Exim is about to re-exec itself, it ensures that the file descriptors
7004 0, 1, and 2 exist, because some OS complain for execs without them (see
7005 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
7006 descriptors. However, the code omitted to check that the open succeeded,
7007 causing mysterious errors if for some reason the permissions on /dev/null
7008 got screwed. Now Exim writes a message to the main and panic logs, and
7009 bombs out if it can't open /dev/null.
7011 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
7012 interact so that it is all more flexible. It is supposed to remain
7013 backwards compatible. Also added extra_local_interfaces.
7015 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
7016 to bomb out with an assertion failure - to the client this appears as a
7017 connection drop. This problem occurs in the part of the code that was taken
7018 from the Samba project. Fortunately, the assertion is in a very simple
7019 function, so I have fixed this by reproducing the function inline in the
7020 one place where it is called, and arranging for authentication to fail
7021 instead of killing the process with assert().
7023 75. The SPA client code was not working when the server requested OEM rather
7024 than Unicode encoding.
7026 76. Added code to make require_files with a specific uid setting more usable in
7027 the case where statting the file as root fails - usually a non-root-mounted
7028 NFS file system. When this happens and the failure is EACCES, Exim now
7029 forks a subprocess and does the per-uid checking as the relevant uid.
7031 77. Added process_log_path.
7033 78. If log_file_path was not explicitly set, a setting of check_log_space or
7034 check_log_inodes was ignored.
7036 79. If a space check for the spool or log partitions fails, the incident is now
7037 logged. Of course, in the latter case the data may get lost...
7039 80. Added the %p formatting code to string_format() so that it can be used to
7040 print addresses in debug_print(). Adjusted all the address printing in the
7041 debugging in store.c to use %p rather than %d.
7043 81. There was a concern that a line of code in smtp_in.c could overflow a
7044 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
7045 initially expressed, the concern was not well-founded, because trailing
7046 spaces are removed early. However, if the trailing spaces were followed by
7047 a NULL, they did not get removed, so the overflow was possible. Two fixes
7050 (a) I re-wrote the offending code in a cleaner fashion.
7051 (b) If an incoming SMTP command contains a NULL character, it is rejected
7054 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
7055 runs initgroups(), so that if the Exim user is in any additional groups,
7056 they will be used during message reception.
7062 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / exim.git / blob