1 Change log file for Exim from version 4.21
2 ------------------------------------------
3 This document describes *changes* to previous versions, that might
4 affect Exim's operation, with an unchanged configuration file. For new
5 options, and new features, see the NewStuff file next to this ChangeLog.
8 Since Exim version 4.90
9 -----------------------
11 JH/01 Replace the store_release() internal interface with store_newblock(),
12 which internalises the check required to safely use the old one, plus
13 the allocate and data copy operations duplicated in both (!) of the
16 JH/02 Disallow '/' characters in queue names specified for the "queue=" ACL
17 modifier. This matches the restriction on the commandline.
19 JH/03 Fix pgsql lookup for multiple result-tuples with a single column.
20 Previously only the last row was returned.
22 JH/04 Bug 2217: Tighten up the parsing of DKIM signature headers. Previously
23 we assumed that tags in the header were well-formed, and parsed the
24 element content after inspecting only the first char of the tag.
25 Assumptions at that stage could crash the receive process on malformed
28 JH/05 Bug 2215: Fix crash associated with dnsdb lookup done from DKIM ACL.
29 While running the DKIM ACL we operate on the Permanent memory pool so that
30 variables created with "set" persist to the DATA ACL. Also (at any time)
31 DNS lookups that fail create cache records using the Permanent pool. But
32 expansions release any allocations made on the current pool - so a dnsdb
33 lookup expansion done in the DKIM ACL releases the memory used for the
34 DNS negative-cache, and bad things result. Solution is to switch to the
35 Main pool for expansions.
36 While we're in that code, add checks on the DNS cache during store_reset,
37 active in the testsuite.
38 Problem spotted, and debugging aided, by Wolfgang Breyha.
40 JH/06 Fix issue with continued-connections when the DNS shifts unreliably.
41 When none of the hosts presented to a transport match an already-open
42 connection, close it and proceed with the list. Previously we would
43 queue the message. Spotted by Lena with Yahoo, probably involving
46 JH/07 Bug 2214: Fix SMTP responses resulting from non-accept result of MIME ACL.
47 Previously a spurious "250 OK id=" response was appended to the proper
50 JH/08 The "support for" informational output now, which built with Content
51 Scanning support, has a line for the malware scanner interfaces compiled
52 in. Interface can be individually included or not at build time.
54 JH/09 The "aveserver", "kavdaemon" and "mksd" interfaces are now not included
55 by the template makefile "src/EDITME". The "STREAM" support for an older
56 ClamAV interface method is removed.
58 JH/10 Bug 2223: Fix mysql lookup returns for the no-data case (when the number of
59 rows affected is given instead).
61 JH/11 The runtime Berkeley DB library version is now additionally output by
62 "exim -d -bV". Previously only the compile-time version was shown.
64 JH/12 Bug 2230: Fix cutthrough routing for nonfirst messages in an initiating
65 SMTP connection. Previously, when one had more receipients than the
66 first, an abortive onward connection was made. Move to full support for
67 multiple onward connections in sequence, handling cutthrough connection
68 for all multi-message initiating connections.
70 JH/13 Bug 2229: Fix cutthrough routing for nonstandard port numbers defined by
71 routers. Previously, a multi-recipient message would fail to match the
72 onward-connection opened for the first recipient, and cause its closure.
74 JH/14 Bug 2174: A timeout on connect for a callout was also erroneously seen as
75 a timeout on read on a GnuTLS initiating connection, resulting in the
76 initiating connection being dropped. This mattered most when the callout
77 was marked defer_ok. Fix to keep the two timeout-detection methods
80 JH/15 Relax results from ACL control request to enable cutthrough, in
81 unsupported situations, from error to silently (except under debug)
82 ignoring. This covers use with PRDR, frozen messages, queue-only and
85 HS/01 Fix Buffer overflow in base64d() (CVE-2018-6789)
87 JH/16 Fix bug in DKIM verify: a buffer overflow could corrupt the malloc
88 metadata, resulting in a crash in free().
90 PP/01 Fix broken Heimdal GSSAPI authenticator integration.
91 Broken in f2ed27cf5, missing an equals sign for specified-initialisers.
92 Broken also in d185889f4, with init system revamp.
94 JH/17 Bug 2113: Fix conversation closedown with the Avast malware scanner.
95 Previously we abruptly closed the connection after reading a malware-
96 found indication; now we go on to read the "scan ok" response line,
99 JH/18 Bug 2239: Enforce non-usability of control=utf8_downconvert in the mail
100 ACL. Previously, a crash would result.
102 JH/19 Speed up macro lookups during configuration file read, by skipping non-
103 macro text after a replacement (previously it was only once per line) and
104 by skipping builtin macros when searching for an uppercase lead character.
106 JH/20 DANE support moved from Experimental to mainline. The Makefile control
107 for the build is renamed.
113 JH/01 Rework error string handling in TLS interface so that the caller in
114 more cases is responsible for logging. This permits library-sourced
115 string to be attached to addresses during delivery, and collapses
116 pairs of long lines into single ones.
118 PP/01 Allow PKG_CONFIG_PATH to be set in Local/Makefile and use it correctly
119 during configuration. Wildcards are allowed and expanded.
121 JH/02 Rework error string handling in DKIM to pass more info back to callers.
122 This permits better logging.
124 JH/03 Rework the transport continued-connection mechanism: when TLS is active,
125 do not close it down and have the child transport start it up again on
126 the passed-on TCP connection. Instead, proxy the child (and any
127 subsequent ones) for TLS via a unix-domain socket channel. Logging is
128 affected: the continued delivery log lines do not have any DNSSEC, TLS
129 Certificate or OCSP information. TLS cipher information is still logged.
131 JH/04 Shorten the log line for daemon startup by collapsing adjacent sets of
132 identical IP addresses on different listening ports. Will also affect
135 PP/02 Bug 2070: uClibc defines __GLIBC__ without providing glibc headers;
136 add noisy ifdef guards to special-case this sillyness.
137 Patch from Bernd Kuhls.
139 JH/05 Tighten up the checking in isip4 (et al): dotted-quad components larger
140 than 255 are no longer allowed.
142 JH/06 Default openssl_options to include +no_ticket, to reduce load on peers.
143 Disable the session-cache too, which might reduce our load. Since we
144 currrectly use a new context for every connection, both as server and
145 client, there is no benefit for these.
146 GnuTLS appears to not support tickets server-side by default (we don't
147 call gnutls_session_ticket_enable_server()) but client side is enabled
148 by default on recent versions (3.1.3 +) unless the PFS priority string
151 PP/03 Add $SOURCE_DATE_EPOCH support for reproducible builds, per spec at
152 <https://reproducible-builds.org/specs/source-date-epoch/>.
154 JH/07 Fix smtp transport use of limited max_rcpt under mua_wrapper. Previously
155 the check for any unsuccessful recipients did not notice the limit, and
156 erroneously found still-pending ones.
158 JH/08 Pipeline CHUNKING command and data together, on kernels that support
159 MSG_MORE. Only in-clear (not on TLS connections).
161 JH/09 Avoid using a temporary file during transport using dkim. Unless a
162 transport-filter is involved we can buffer the headers in memory for
163 creating the signature, and read the spool data file once for the
164 signature and again for transmission.
166 JH/10 Enable use of sendfile in Linux builds as default. It was disabled in
167 4.77 as the kernel support then wasn't solid, having issues in 64bit
168 mode. Now, it's been long enough. Add support for FreeBSD also.
170 JH/11 Bug 2104: Fix continued use of a transport connection with TLS. In the
171 case where the routing stage had gathered several addresses to send to
172 a host before calling the transport for the first, we previously failed
173 to close down TLS in the old transport process before passing the TCP
174 connection to the new process. The new one sent a STARTTLS command
175 which naturally failed, giving a failed delivery and bloating the retry
176 database. Investigation and fix prototype from Wolfgang Breyha.
178 JH/12 Fix check on SMTP command input synchronisation. Previously there were
179 false-negatives in the check that the sender had not preempted a response
180 or prompt from Exim (running as a server), due to that code's lack of
181 awareness of the SMTP input buffering.
183 PP/04 Add commandline_checks_require_admin option.
184 Exim drops privileges sanely, various checks such as -be aren't a
185 security problem, as long as you trust local users with access to their
186 own account. When invoked by services which pass untrusted data to
187 Exim, this might be an issue. Set this option in main configuration
188 AND make fixes to the calling application, such as using `--` to stop
191 JH/13 Do pipelining under TLS. Previously, although safe, no advantage was
192 taken. Now take care to pack both (client) MAIL,RCPT,DATA, and (server)
193 responses to those, into a single TLS record each way (this usually means
194 a single packet). As a side issue, smtp_enforce_sync now works on TLS
197 PP/05 OpenSSL/1.1: use DH_bits() for more accurate DH param sizes. This
198 affects you only if you're dancing at the edge of the param size limits.
199 If you are, and this message makes sense to you, then: raise the
200 configured limit or use OpenSSL 1.1. Nothing we can do for older
203 JH/14 For the "sock" variant of the malware scanner interface, accept an empty
204 cmdline element to get the documented default one. Previously it was
207 JH/15 Fix a crash in the smtp transport caused when two hosts in succession
208 are unsuable for non-message-specific reasons - eg. connection timeout,
209 banner-time rejection.
211 JH/16 Fix logging of delivery remote port, when specified by router, under
214 PP/06 Repair manualroute's ability to take options in any order, even if one
215 is the name of a transport.
218 HS/01 Cleanup, prevent repeated use of -p/-oMr (CVE-2017-1000369)
220 JH/17 Change the list-building routines interface to use the expanding-string
221 triplet model, for better allocation and copying behaviour.
223 JH/18 Prebuild the data-structure for "builtin" macros, for faster startup.
224 Previously it was constructed the first time a possibly-matching string
225 was met in the configuration file input during startup; now it is done
228 JH/19 Bug 2141: Use the full-complex API for Berkeley DB rather than the legacy-
229 compatible one, to avoid the (poorly documented) possibility of a config
230 file in the working directory redirecting the DB files, possibly correpting
231 some existing file. CVE-2017-10140 assigned for BDB.
233 JH/20 Bug 2147: Do not defer for a verify-with-callout-and-random which is not
234 cache-hot. Previously, although the result was properly cached, the
235 initial verify call returned a defer.
237 JH/21 Bug 2151: Avoid using SIZE on the MAIL for a callout verify, on any but
238 the main verify for receipient in uncached-mode.
240 JH/22 Retire historical build files to an "unsupported" subdir. These are
241 defined as "ones for which we have no current evidence of testing".
243 JH/23 DKIM: enforce the DNS pubkey record "h" permitted-hashes optional field,
244 if present. Previously it was ignored.
246 JH/24 Start using specified-initialisers in C structure init coding. This is
247 a C99 feature (it's 2017, so now considered safe).
249 JH/25 Use one-bit bitfields for flags in the "addr" data structure. Previously
250 if was a fixed-sized field and bitmask ops via macros; it is now more
253 PP/07 GitHub PR 56: Apply MariaDB build fix.
254 Patch provided by Jaroslav Škarvada.
256 PP/08 Bug 2161: Fix regression in sieve quoted-printable handling introduced
257 during Coverity cleanups [4.87 JH/47]
258 Diagnosis and fix provided by Michael Fischer v. Mollard.
260 JH/26 Fix DKIM bug: when the pseudoheader generated for signing was exactly
261 the right size to place the terminating semicolon on its own folded
262 line, the header hash was calculated to an incorrect value thanks to
263 the (relaxed) space the fold became.
265 HS/02 Fix Bug 2130: large writes from the transport subprocess where chunked
266 and confused the parent.
268 JH/27 Fix SOCKS bug: an unitialized pointer was deref'd by the transport process
269 which could crash as a result. This could lead to undeliverable messages.
271 JH/28 Logging: "next input sent too soon" now shows where input was truncated
274 JH/29 Fix queue_run_in_order to ignore the PID portion of the message ID. This
275 matters on fast-turnover and PID-randomising systems, which were getting
276 out-of-order delivery.
278 JH/30 Fix a logging bug on aarch64: an unsafe routine was previously used for
279 a possibly-overlapping copy. The symptom was that "Remote host closed
280 connection in response to HELO" was logged instead of the actual 4xx
283 JH/31 Fix CHUNKING code to properly flush the unwanted chunk after an error.
284 Previously only that bufferd was discarded, resulting in SYMTP command
287 JH/32 DKIM: when a message has multiple signatures matching an identity given
288 in dkim_verify_signers, run the dkim acl once for each. Previously only
289 one run was done. Bug 2189.
291 JH/33 Downgrade an unfound-list name (usually a typo in the config file) from
292 "panic the current process" to "deliberately defer". The panic log is
293 still written with the problem list name; the mail and reject logs now
294 get a temp-reject line for the message that was being handled, saying
295 something like "domains check lookup or other defer". The SMTP 451
296 message is still "Temporary local problem".
298 JH/34 Bug 2199: Fix a use-after-free while reading smtp input for header lines.
299 A crafted sequence of BDAT commands could result in in-use memory beeing
300 freed. CVE-2017-16943.
302 HS/03 Bug 2201: Fix checking for leading-dot on a line during headers reading
303 from SMTP input. Previously it was always done; now only done for DATA
304 and not BDAT commands. CVE-2017-16944.
306 JH/35 Bug 2201: Flush received data in BDAT mode after detecting an error fatal
307 to the message (such as an overlong header line). Previously this was
308 not done and we did not exit BDAT mode. Followon from the previous item
309 though a different problem.
315 JH/01 Bug 1922: Support IDNA2008. This has slightly different conversion rules
316 than -2003 did; needs libidn2 in addition to libidn.
318 JH/02 The path option on a pipe transport is now expanded before use.
320 PP/01 GitHub PR 50: Do not call ldap_start_tls_s on ldapi:// connections.
321 Patch provided by "Björn", documentation fix added too.
323 JH/03 Bug 2003: fix Proxy Protocol v2 handling: the address size field was
324 missing a wire-to-host endian conversion.
326 JH/04 Bug 2004: fix CHUNKING in non-PIPELINEING mode. Chunk data following
327 close after a BDAT command line could be taken as a following command,
328 giving a synch failure. Fix by only checking for synch immediately
329 before acknowledging the chunk.
331 PP/02 GitHub PR 52: many spelling fixes, which include fixing parsing of
332 no_require_dnssec option and creation of _HAVE_TRANSPORT_APPEND_MAILDIR
333 macro. Patches provided by Josh Soref.
335 JH/05 Have the EHLO response advertise VRFY, if there is a vrfy ACL defined.
336 Previously we did not; the RFC seems ambiguous and VRFY is not listed
337 by IANA as a service extension. However, John Klensin suggests that we
340 JH/06 Bug 2017: Fix DKIM verification in -bh test mode. The data feed into
341 the dkim code may be unix-mode line endings rather than smtp wire-format
342 CRLF, so prepend a CR to any bare LF.
344 JH/07 Rationalise the coding for callout smtp conversations and transport ones.
345 As a side-benfit, callouts can now use PIPELINING hence fewer round-trips.
347 JH/08 Bug 2016: Fix DKIM verification vs. CHUNKING. Any BDAT commands after
348 the first were themselves being wrongly included in the feed into dkim
349 processing; with most chunk sizes in use this resulted in an incorrect
350 body hash calculated value.
352 JH/09 Bug 2014: permit inclusion of a DKIM-Signature header in a received
353 DKIM signature block, for verification. Although advised against by
354 standards it is specifically not ruled illegal.
356 JH/10 Bug 2025: Fix reception of (quoted) local-parts with embedded spaces.
358 JH/11 Bug 2029: Fix crash in DKIM verification when a message signature block is
359 missing a body hash (the bh= tag).
361 JH/12 Bug 2018: Re-order Proxy Protocol startup versus TLS-on-connect startup.
362 It seems that HAProxy sends the Proxy Protocol information in clear and
363 only then does a TLS startup, so do the same.
365 JH/13 Bug 2027: Avoid attempting to use TCP Fast Open for non-transport client
366 TCP connections (such as for Spamd) unless the daemon successfully set
367 Fast Open mode on its listening sockets. This fixes breakage seen on
368 too-old kernels or those not configured for Fast Open, at the cost of
369 requiring both directions being enabled for TFO, and TFO never being used
370 by non-daemon-related Exim processes.
372 JH/14 Bug 2000: Reject messages recieved with CHUNKING but with malformed line
373 endings, at least on the first header line. Try to canonify any that get
374 past that check, despite the cost.
376 JH/15 Angle-bracket nesting (an error inserted by broken sendmails) levels are
377 now limited to an arbitrary five deep, while parsing addresses with the
378 strip_excess_angle_brackets option enabled.
380 PP/03 Bug 2018: For Proxy Protocol and TLS-on-connect, do not over-read and
381 instead leave the unprompted TLS handshake in socket buffer for the
382 TLS library to consume.
384 PP/04 Bug 2018: Also handle Proxy Protocol v2 safely.
386 PP/05 FreeBSD compat: handle that Ports no longer create /usr/bin/perl
388 JH/16 Drop variables when they go out of scope. Memory management drops a whole
389 region in one operation, for speed, and this leaves assigned pointers
390 dangling. Add checks run only under the testsuite which checks all
391 variables at a store-reset and panics on a dangling pointer; add code
392 explicitly nulling out all the variables discovered. Fixes one known
393 bug: a transport crash, where a dangling pointer for $sending_ip_address
394 originally assigned in a verify callout, is re-used.
396 PP/06 Drop '.' from @INC in various Perl scripts.
398 PP/07 Switch FreeBSD iconv to always use the base-system libc functions.
400 PP/08 Reduce a number of compilation warnings under clang; building with
401 CC=clang CFLAGS+=-Wno-dangling-else -Wno-logical-op-parentheses
402 should be warning-free.
404 JH/17 Fix inbound CHUNKING when DKIM disabled at runtime.
406 HS/01 Fix portability problems introduced by PP/08 for platforms where
407 realloc(NULL) is not equivalent to malloc() [SunOS et al].
409 HS/02 Bug 1974: Fix missing line terminator on the last received BDAT
410 chunk. This allows us to accept broken chunked messages. We need a more
411 general solution here.
413 PP/09 Wrote util/chunking_fixqueue_finalnewlines.pl to help recover
414 already-broken messages in the queue.
416 JH/18 Bug 2061: Fix ${extract } corrupting an enclosing ${reduce } $value.
418 JH/19 Fix reference counting bug in routing-generated-address tracking.
424 JH/01 Use SIZE on MAIL FROM in a cutthrough connection, if the destination
425 supports it and a size is available (ie. the sending peer gave us one).
427 JH/02 The obsolete acl condition "demime" is removed (finally, after ten
428 years of being deprecated). The replacements are the ACLs
429 acl_smtp_mime and acl_not_smtp_mime.
431 JH/03 Upgrade security requirements imposed for hosts_try_dane: previously
432 a downgraded non-dane trust-anchor for the TLS connection (CA-style)
433 or even an in-clear connection were permitted. Now, if the host lookup
434 was dnssec and dane was requested then the host is only used if the
435 TLSA lookup succeeds and is dnssec. Further hosts (eg. lower priority
436 MXs) will be tried (for hosts_try_dane though not for hosts_require_dane)
437 if one fails this test.
438 This means that a poorly-configured remote DNS will make it incommunicado;
439 but it protects against a DNS-interception attack on it.
441 JH/04 Bug 1810: make continued-use of an open smtp transport connection
442 non-noisy when a race steals the message being considered.
444 JH/05 If main configuration option tls_certificate is unset, generate a
445 self-signed certificate for inbound TLS connections.
447 JH/06 Bug 165: hide more cases of password exposure - this time in expansions
448 in rewrites and routers.
450 JH/07 Retire gnutls_require_mac et.al. These were nonfunctional since 4.80
451 and logged a warning sing 4.83; now they are a configuration file error.
453 JH/08 Bug 1836: Fix crash in VRFY handling when handed an unqualified name
454 (lacking @domain). Apply the same qualification processing as RCPT.
456 JH/09 Bug 1804: Avoid writing msglog files when in -bh or -bhc mode.
458 JH/10 Support ${sha256:} applied to a string (as well as the previous
461 JH/11 Cutthrough: avoid using the callout hints db on a verify callout when
462 a cutthrough deliver is pending, as we always want to make a connection.
463 This also avoids re-routing the message when later placing the cutthrough
464 connection after a verify cache hit.
465 Do not update it with the verify result either.
467 JH/12 Cutthrough: disable when verify option success_on_redirect is used, and
468 when routing results in more than one destination address.
470 JH/13 Cutthrough: expand transport dkim_domain option when testing for dkim
471 signing (which inhibits the cutthrough capability). Previously only
472 the presence of an option was tested; now an expansion evaluating as
473 empty is permissible (obviously it should depend only on data available
474 when the cutthrough connection is made).
476 JH/14 Fix logging of errors under PIPELINING. Previously the log line giving
477 the relevant preceding SMTP command did not note the pipelining mode.
479 JH/15 Fix counting of empty lines in $body_linecount and $message_linecount.
480 Previously they were not counted.
482 JH/16 DANE: treat a TLSA lookup response having all non-TLSA RRs, the same
483 as one having no matching records. Previously we deferred the message
484 that needed the lookup.
486 JH/17 Fakereject: previously logged as a normal message arrival "<="; now
487 distinguished as "(=".
489 JH/18 Bug 1867: make the fail_defer_domains option on a dnslookup router work
490 for missing MX records. Previously it only worked for missing A records.
492 JH/19 Bug 1850: support Radius libraries that return REJECT_RC.
494 JH/20 Bug 1872: Ensure that acl_smtp_notquit is run when the connection drops
495 after the data-go-ahead and data-ack. Patch from Jason Betts.
497 JH/21 Bug 1846: Send DMARC forensic reports for reject and quarantine results,
498 even for a "none" policy. Patch from Tony Meyer.
500 JH/22 Fix continued use of a connection for further deliveries. If a port was
501 specified by a router, it must also match for the delivery to be
504 JH/23 Bug 1874: fix continued use of a connection for further deliveries.
505 When one of the recipients of a message was unsuitable for the connection
506 (has no matching addresses), we lost track of needing to mark it
507 deferred. As a result mail would be lost.
509 JH/24 Bug 1832: Log EHLO response on getting conn-close response for HELO.
511 JH/25 Decoding ACL controls is now done using a binary search; the source code
512 takes up less space and should be simpler to maintain. Merge the ACL
513 condition decode tables also, with similar effect.
515 JH/26 Fix problem with one_time used on a redirect router which returned the
516 parent address unchanged. A retry would see the parent address marked as
517 delivered, so not attempt the (identical) child. As a result mail would
520 JH/27 Fix a possible security hole, wherein a process operating with the Exim
521 UID can gain a root shell. Credit to http://www.halfdog.net/ for
522 discovery and writeup. Ubuntu bug 1580454; no bug raised against Exim
525 JH/28 Enable {spool,log} filesystem space and inode checks as default.
526 Main config options check_{log,spool}_{inodes,space} are now
527 100 inodes, 10MB unless set otherwise in the configuration.
529 JH/29 Fix the connection_reject log selector to apply to the connect ACL.
530 Previously it only applied to the main-section connection policy
533 JH/30 Bug 1897: fix callouts connection fallback from TLS to cleartext.
535 PP/01 Changed default Diffie-Hellman parameters to be Exim-specific, created
536 by me. Added RFC7919 DH primes as an alternative.
538 PP/02 Unbreak build via pkg-config with new hash support when crypto headers
539 are not in the system include path.
541 JH/31 Fix longstanding bug with aborted TLS server connection handling. Under
542 GnuTLS, when a session startup failed (eg because the client disconnected)
543 Exim did stdio operations after fclose. This was exposed by a recent
544 change which nulled out the file handle after the fclose.
546 JH/32 Bug 1909: Fix OCSP proof verification for cases where the proof is
547 signed directly by the cert-signing cert, rather than an intermediate
548 OCSP-signing cert. This is the model used by LetsEncrypt.
550 JH/33 Bug 1914: Ensure socket is nonblocking before draining after SMTP QUIT.
552 HS/01 Fix leak in verify callout under GnuTLS, about 3MB per recipient on
553 an incoming connection.
555 HS/02 Bug 1802: Do not half-close the connection after sending a request
558 HS/03 Use "auto" as the default EC curve parameter. For OpenSSL < 1.0.2
559 fallback to "prime256v1".
561 JH/34 SECURITY: Use proper copy of DATA command in error message.
562 Could leak key material. Remotely exploitable. CVE-2016-9963.
568 JH/01 Bug 1664: Disable OCSP for GnuTLS library versions at/before 3.3.16
569 and 3.4.4 - once the server is enabled to respond to an OCSP request
570 it does even when not requested, resulting in a stapling non-aware
571 client dropping the TLS connection.
573 TF/01 Code cleanup: Overhaul the debug_selector and log_selector machinery to
574 support variable-length bit vectors. No functional change.
576 TF/02 Improve the consistency of logging incoming and outgoing interfaces.
577 The I= interface field on outgoing lines is now after the H= remote
578 host field, same as incoming lines. There is a separate
579 outgoing_interface log selector which allows you to disable the
582 JH/02 Bug 728: Close logfiles after a daemon-process "exceptional" log write.
583 If not running log_selector +smtp_connection the mainlog would be held
584 open indefinitely after a "too many connections" event, including to a
585 deleted file after a log rotate. Leave the per net connection logging
586 leaving it open for efficiency as that will be quickly detected by the
587 check on the next write.
589 HS/01 Bug 1671: Fix post transport crash.
590 Processing the wait-<transport> messages could crash the delivery
591 process if the message IDs didn't exist for some reason. When
592 using 'split_spool_directory=yes' the construction of the spool
593 file name failed already, exposing the same netto behaviour.
595 JH/03 Bug 425: Capture substrings in $regex1, $regex2 etc from regex &
596 mime_regex ACL conditions.
598 JH/04 Bug 1686: When compiled with EXPERIMENTAL_DSN_INFO: Add extra information
599 to DSN fail messages (bounces): remote IP, remote greeting, remote response
600 to HELO, local diagnostic string.
602 JH/05 Downgrade message for a TLS-certificate-based authentication fail from
603 log line to debug. Even when configured with a tls authenticator many
604 client connections are expected to not authenticate in this way, so
605 an authenticate fail is not an error.
607 HS/02 Add the Exim version string to the process info. This way exiwhat
608 gives some more detail about the running daemon.
610 JH/06 Bug 1395: time-limit caching of DNS lookups, to the TTL value. This may
611 matter for fast-change records such as DNSBLs.
613 JH/07 Bug 1678: Always record an interface option value, if set, as part of a
614 retry record, even if constant. There may be multiple transports with
615 different interface settings and the retry behaviour needs to be kept
618 JH/08 Bug 1586: exiqgrep now refuses to run if there are unexpected arguments.
620 JH/09 Bug 1700: ignore space & tab embedded in base64 during decode.
622 JH/10 Bug 840: fix log_defer_output option of pipe transport
624 JH/11 Bug 830: use same host for all RCPTS of a message, even under
625 hosts_randomize. This matters a lot when combined with mua_wrapper.
627 JH/12 Bug 1706: percent and underbar characters are no longer escaped by the
628 ${quote_pgsql:<string>} operator.
630 JH/13 Bug 1708: avoid misaligned access in cached lookup.
632 JH/14 Change header file name for freeradius-client. Relevant if compiling
633 with Radius support; from the Gentoo tree and checked under Fedora.
635 JH/15 Bug 1712: Introduce $prdr_requested flag variable
637 JH/16 Bug 1714: Permit an empty string as expansion result for transport
638 option transport_filter, meaning no filtering.
640 JH/17 Bug 1713: Fix non-PDKIM_DEBUG build. Patch from Jasen Betts.
642 JH/18 Bug 1709: When built with TLS support, the tls_advertise_hosts option now
643 defaults to "*" (all hosts). The variable is now available when not built
644 with TLS, default unset, mainly to enable keeping the testsuite sane.
645 If a server certificate is not supplied (via tls_certificate) an error is
646 logged, and clients will find TLS connections fail on startup. Presumably
647 they will retry in-clear.
648 Packagers of Exim are strongly encouraged to create a server certificate
649 at installation time.
651 HS/03 Add -bP config_file as a synonym for -bP configure_file, for consistency
652 with the $config_file variable.
654 JH/19 Two additional event types: msg:rcpt:defer and msg:rcpt:host:defer. Both
655 in transport context, after the attempt, and per-recipient. The latter type
656 is per host attempted. The event data is the error message, and the errno
657 information encodes the lookup type (A vs. MX) used for the (first) host,
658 and the trailing two digits of the smtp 4xx response.
660 GF/01 Bug 1715: Fix for race condition in exicyclog, where exim could attempt
661 to write to mainlog (or rejectlog, paniclog) in the window between file
662 creation and permissions/ownership being changed. Particularly affects
663 installations where exicyclog is run as root, rather than exim user;
664 result is that the running daemon panics and dies.
666 JH/20 Bug 1701: For MySQL lookups, support MySQL config file option group names.
668 JH/21 Bug 1720: Add support for priority groups and weighted-random proxy
669 selection for the EXPERIMENTAL_SOCKS feature, via new per-proxy options
670 "pri" and "weight". Note that the previous implicit priority given by the
671 list order is no longer honoured.
673 JH/22 Bugs 963, 1721: Fix some corner cases in message body canonicalization
676 JH/23 Move SOCKS5 support from Experimental to mainline, enabled for a build
677 by defining SUPPORT_SOCKS.
679 JH/26 Move PROXY support from Experimental to mainline, enabled for a build
680 by defining SUPPORT_PROXY. Note that the proxy_required_hosts option
681 is renamed to hosts_proxy, and the proxy_{host,target}_{address,port}.
682 variables are renamed to proxy_{local,external}_{address,port}.
684 JH/27 Move Internationalisation support from Experimental to mainline, enabled
685 for a build by defining SUPPORT_I18N
687 JH/28 Bug 1745: Fix redis lookups to handle (quoted) spaces embedded in parts
688 of the query string, and make ${quote_redis:} do that quoting.
690 JH/29 Move Events support from Experimental to mainline, enabled by default
691 and removable for a build by defining DISABLE_EVENT.
693 JH/30 Updated DANE implementation code to current from Viktor Dukhovni.
695 JH/31 Fix bug with hosts_connection_nolog and named-lists which were wrongly
696 cached by the daemon.
698 JH/32 Move Redis support from Experimental to mainline, enabled for a build
699 by defining LOOKUP_REDIS. The libhiredis library is required.
701 JH/33 Bug 1748: Permit ACL dnslists= condition in non-smtp ACLs if explicit
702 keys are given for lookup.
704 JH/34 Bug 1192: replace the embedded copy of PolarSSL RSA routines in the DKIM
705 support, by using OpenSSL or GnuTLS library ones. This means DKIM is
706 only supported when built with TLS support. The PolarSSL SHA routines
707 are still used when the TLS library is too old for convenient support.
709 JH/35 Require SINGLE_DH_USE by default in OpenSSL (main config option
710 openssl_options), for security. OpenSSL forces this from version 1.1.0
711 server-side so match that on older versions.
713 JH/36 Bug 1778: longstanding bug in memory use by the ${run } expansion: A fresh
714 allocation for $value could be released as the expansion processing
715 concluded, but leaving the global pointer active for it.
717 JH/37 Bug 1769: Permit a VRFY ACL to override the default 252 response,
718 and to use the domains and local_parts ACL conditions.
720 JH/38 Fix cutthrough bug with body lines having a single dot. The dot was
721 incorrectly not doubled on cutthrough transmission, hence seen as a
722 body-termination at the receiving system - resulting in truncated mails.
723 Commonly the sender saw a TCP-level error, and retransmitted the message
724 via the normal store-and-forward channel. This could result in duplicates
725 received - but deduplicating mailstores were liable to retain only the
726 initial truncated version.
728 JH/39 Bug 1781: Fix use of DKIM private-keys having trailing '=' in the base-64.
730 JH/40 Fix crash in queryprogram router when compiled with EXPERIMENTAL_SRS.
732 JH/41 Bug 1792: Fix selection of headers to sign for DKIM: bottom-up. While
733 we're in there, support oversigning also; bug 1309.
735 JH/42 Bug 1796: Fix error logged on a malware scanner connection failure.
737 HS/04 Add support for keep_environment and add_environment options.
739 JH/43 Tidy coding issues detected by gcc --fsanitize=undefined. Some remain;
740 either intentional arithmetic overflow during PRNG, or testing config-
743 JH/44 Bug 1800: The combination of a -bhc commandline option and cutthrough
744 delivery resulted in actual delivery. Cancel cutthrough before DATA
747 JH/45 Fix cutthrough, when connection not opened by verify and target hard-
748 rejects a recipient: pass the reject to the originator.
750 JH/46 Multiple issues raised by Coverity. Some were obvious or plausible bugs.
751 Many were false-positives and ignorable, but it's worth fixing the
754 JH/47 Fix build on HP-UX and older Solaris, which need (un)setenv now also
755 for the new environment-manipulation done at startup. Move the routines
756 from being local to tls.c to being global via the os.c file.
758 JH/48 Bug 1807: Fix ${extract } for the numeric/3-string case. While preparsing
759 an extract embedded as result-arg for a map, the first arg for extract
760 is unavailable so we cannot tell if this is a numbered or keyed
761 extraction. Accept either.
767 JH/01 Bug 1545: The smtp transport option "retry_include_ip_address" is now
770 JH/02 The smtp transport option "multi_domain" is now expanded.
772 JH/03 The smtp transport now requests PRDR by default, if the server offers
775 JH/04 Certificate name checking on server certificates, when exim is a client,
776 is now done by default. The transport option tls_verify_cert_hostnames
777 can be used to disable this per-host. The build option
778 EXPERIMENTAL_CERTNAMES is withdrawn.
780 JH/05 The value of the tls_verify_certificates smtp transport and main options
781 default to the word "system" to access the system default CA bundle.
782 For GnuTLS, only version 3.0.20 or later.
784 JH/06 Verification of the server certificate for a TLS connection is now tried
785 (but not required) by default. The verification status is now logged by
786 default, for both outbound TLS and client-certificate supplying inbound
789 JH/07 Changed the default rfc1413 lookup settings to disable calls. Few
792 JH/08 The EXPERIMENTAL_DSN compile option is no longer needed; all Delivery
793 Status Notification (bounce) messages are now MIME format per RFC 3464.
794 Support for RFC 3461 DSN options NOTIFY,ENVID,RET,ORCPT can be advertised
795 under the control of the dsn_advertise_hosts option, and routers may
796 have a dsn_lasthop option.
798 JH/09 A timeout of 2 minutes is now applied to all malware scanner types by
799 default, modifiable by a malware= option. The list separator for
800 the options can now be changed in the usual way. Bug 68.
802 JH/10 The smtp_receive_timeout main option is now expanded before use.
804 JH/11 The incoming_interface log option now also enables logging of the
805 local interface on delivery outgoing connections.
807 JH/12 The cutthrough-routing facility now supports multi-recipient mails,
808 if the interface and destination host and port all match.
810 JH/13 Bug 344: The verify = reverse_host_lookup ACL condition now accepts a
813 JH/14 Bug 1573: The spam= ACL condition now additionally supports Rspamd.
814 Patch from Andrew Lewis.
816 JH/15 Bug 670: The spamd_address main option (for the spam= ACL condition)
817 now supports optional time-restrictions, weighting, and priority
818 modifiers per server. Patch originally by <rommer@active.by>.
820 JH/16 The spamd_address main option now supports a mixed list of local
821 and remote servers. Remote servers can be IPv6 addresses, and
822 specify a port-range.
824 JH/17 Bug 68: The spamd_address main option now supports an optional
825 timeout value per server.
827 JH/18 Bug 1581: Router and transport options headers_add/remove can
828 now have the list separator specified.
830 JH/19 Bug 392: spamd_address, and clamd av_scanner, now support retry
833 JH/20 Bug 1571: Ensure that $tls_in_peerdn is set, when verification fails
836 JH/21 Support for the A6 type of dns record is withdrawn.
838 JH/22 Bug 608: The result of a QUIT or not-QUIT toplevel ACL now matters
839 rather than the verbs used.
841 JH/23 Bug 1572: Increase limit on SMTP confirmation message copy size
842 from 255 to 1024 chars.
844 JH/24 Verification callouts now attempt to use TLS by default.
846 HS/01 DNSSEC options (dnssec_require_domains, dnssec_request_domains)
847 are generic router options now. The defaults didn't change.
849 JH/25 Bug 466: Add RFC2322 support for MIME attachment filenames.
850 Original patch from Alexander Shikoff, worked over by JH.
852 HS/02 Bug 1575: exigrep falls back to autodetection of compressed
853 files if ZCAT_COMMAND is not executable.
855 JH/26 Bug 1539: Add timeout/retry options on dnsdb lookups.
857 JH/27 Bug 286: Support SOA lookup in dnsdb lookups.
859 JH/28 Bug 1588: Do not use the A lookup following an AAAA for setting the FQDN.
860 Normally benign, it bites when the pair was led to by a CNAME;
861 modern usage is to not canonicalize the domain to a CNAME target
862 (and we were inconsistent anyway for A-only vs AAAA+A).
864 JH/29 Bug 1632: Removed the word "rejected" from line logged for ACL discards.
866 JH/30 Check the forward DNS lookup for DNSSEC, in addition to the reverse,
867 when evaluating $sender_host_dnssec.
869 JH/31 Check the HELO verification lookup for DNSSEC, adding new
870 $sender_helo_dnssec variable.
872 JH/32 Bug 1397: Enable ECDHE on OpenSSL, just the NIST P-256 curve.
874 JH/33 Bug 1346: Note MAIL cmd seen in -bS batch, to avoid smtp_no_mail log.
876 JH/34 Bug 1648: Fix a memory leak seen with "mailq" and large queues.
878 JH/35 Bug 1642: Fix support of $spam_ variables at delivery time. Was
879 documented as working, but never had. Support all but $spam_report.
881 JH/36 Bug 1659: Guard checking of input smtp commands again pseudo-command
882 added for tls authenticator.
884 HS/03 Add perl_taintmode main config option
890 TL/01 When running the test suite, the README says that variables such as
891 no_msglog_check are global and can be placed anywhere in a specific
892 test's script, however it was observed that placement needed to be near
893 the beginning for it to behave that way. Changed the runtest perl
894 script to read through the entire script once to detect and set these
895 variables, reset to the beginning of the script, and then run through
896 the script parsing/test process like normal.
898 TL/02 The BSD's have an arc4random API. One of the functions to induce
899 adding randomness was arc4random_stir(), but it has been removed in
900 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
901 function when detected.
903 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
904 cause callback expansion.
906 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
907 syntax errors in an expansion can be treated as a string instead of
908 logging or causing an error, due to the internal use of bool_lax
909 instead of bool when processing it.
911 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
912 server certificates when making smtp deliveries.
914 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
916 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
918 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
920 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
921 Merged patch from Sebastian Wiedenroth.
923 JH/05 Fix results-pipe from transport process. Several recipients, combined
924 with certificate use, exposed issues where response data items split
925 over buffer boundaries were not parsed properly. This eventually
926 resulted in duplicates being sent. This issue only became common enough
927 to notice due to the introduction of connection certificate information,
928 the item size being so much larger. Found and fixed by Wolfgang Breyha.
930 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
931 size buffer was used, resulting in syntax errors when an expansion
934 JH/07 Add support for directories of certificates when compiled with a GnuTLS
935 version 3.3.6 or later.
937 JH/08 Rename the TPDA experimental facility to Event Actions. The #ifdef
938 is EXPERIMENTAL_EVENT, the main-configuration and transport options
939 both become "event_action", the variables become $event_name, $event_data
940 and $event_defer_errno. There is a new variable $verify_mode, usable in
941 routers, transports and related events. The tls:cert event is now also
942 raised for inbound connections, if the main configuration event_action
945 TL/06 In test suite, disable OCSP for old versions of openssl which contained
946 early OCSP support, but no stapling (appears to be less than 1.0.0).
948 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
949 server certificate names available under the smtp transport option
950 "tls_verify_cert_hostname" now do not permit multi-component wildcard
953 JH/10 Time-related extraction expansions from certificates now use the main
954 option "timezone" setting for output formatting, and are consistent
955 between OpenSSL and GnuTLS compilations. Bug 1541.
957 JH/11 Fix a crash in mime ACL when meeting a zero-length, quoted or RFC2047-
958 encoded parameter in the incoming message. Bug 1558.
960 JH/12 Bug 1527: Autogrow buffer used in reading spool files. Since they now
961 include certificate info, eximon was claiming there were spoolfile
964 JH/13 Bug 1521: Fix ldap lookup for single-attr request, multiple-attr return.
966 JH/14 Log delivery-related information more consistently, using the sequence
967 "H=<name> [<ip>]" wherever possible.
969 TL/07 Bug 1547: Omit RFCs from release. Draft and RFCs have licenses which
970 are problematic for Debian distribution, omit them from the release
973 JH/15 Updates and fixes to the EXPERIMENTAL_DSN feature.
975 JH/16 Fix string representation of time values on 64bit time_t architectures.
978 JH/17 Fix a null-indirection in certextract expansions when a nondefault
979 output list separator was used.
984 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
985 checkers that were complaining about end of non-void function with no
988 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
989 This was a regression introduced in 4.83 by another bugfix.
991 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
993 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
994 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
1000 TF/01 Correctly close the server side of TLS when forking for delivery.
1002 When a message was received over SMTP with TLS, Exim failed to clear up
1003 the incoming connection properly after forking off the child process to
1004 deliver the message. In some situations the subsequent outgoing
1005 delivery connection happened to have the same fd number as the incoming
1006 connection previously had. Exim would try to use TLS and fail, logging
1007 a "Bad file descriptor" error.
1009 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
1010 utilities have not been installed.
1012 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
1013 temporary space as the ACL may create new global variables.
1015 TL/01 LDAP support uses per connection or global context settings, depending
1016 upon the detected version of the libraries at build time.
1018 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
1019 to extract and use the src ip:port in logging and expansions as if it
1020 were a direct connection from the outside internet. PPv2 support was
1021 updated based on HAProxy spec change in May 2014.
1023 JH/02 Add ${listextract {number}{list}{success}{fail}}.
1025 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
1026 Properly escape header and check for NULL return.
1028 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
1031 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
1033 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
1034 characters in header names, implemented as a verify condition.
1035 Contributed by Michael Fischer v. Mollard.
1037 TL/05 Rename SPF condition results err_perm and err_temp to standardized
1038 results permerror and temperror. Previous values are deprecated but
1039 still accepted. In a future release, err_perm and err_temp will be
1040 completely removed, which will be a backward incompatibility if the
1041 ACL tests for either of these two old results. Patch contributed by
1042 user bes-internal on the mailing list.
1044 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
1046 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
1047 selectors, in both main and reject logs.
1049 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
1052 JH/07 Add malware type "sock" for talking to simple daemon.
1054 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
1056 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
1057 routers/transports under cutthrough routing.
1059 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
1060 numbers. Touch up "bool" conditional to keep the same definition.
1062 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
1064 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
1066 JH/12 Expand items in router/transport headers_add or headers_remove lists
1067 individually rather than the list as a whole. Bug 1452.
1069 Required for reasonable handling of multiple headers_ options when
1070 they may be empty; requires that headers_remove items with embedded
1071 colons must have them doubled (or the list-separator changed).
1073 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
1074 view the policy declared in the DMARC record. Currently, $dmarc_status
1075 is a combined value of both the record presence and the result of the
1078 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
1080 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
1081 dnslookup router and the smtp transport (applying to the forward
1084 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
1085 of ldap servers used for a specific lookup. Patch provided by Heiko
1088 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
1089 New variable $lookup_dnssec_authenticated for observability.
1091 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
1092 Patch submitted by Lars Timman.
1094 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
1096 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
1097 Requires trusted mode and valid format message id, aborts otherwise.
1098 Patch contributed by Heiko Schlichting.
1100 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
1101 certextract with support for various fields. Bug 1358.
1103 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
1104 is requested by default, modifiable by smtp transport option
1107 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
1108 operate on certificate variables to give certificate fingerprints
1109 Also new ${sha256:cert_variable}.
1111 JH/23 The PRDR feature is moved from being Experimental into the mainline.
1113 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
1114 Christian Aistleitner.
1116 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
1118 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
1119 file. Patch from Wolfgang Breyha.
1121 JH/25 Expand the coverage of the delivery $host and $host_address to
1122 client authenticators run in verify callout. Bug 1476.
1124 JH/26 Port service names are now accepted for tls_on_connect_ports, to
1125 align with daemon_smtp_ports. Bug 72.
1127 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
1128 support and error reporting did not work properly.
1130 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
1131 and is readable. Patch from Andrew Colin Kissa.
1133 TL/14 Enhance documentation of ${run expansion and how it parses the
1134 commandline after expansion, particularly in the case when an
1135 unquoted variable expansion results in an empty value.
1137 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
1139 PP/02 Fix internal collision of T_APL on systems which support RFC3123
1140 by renaming away from it. Addresses GH issue 15, reported by
1143 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
1145 TL/15 SECURITY: prevent double expansion in math comparison functions
1146 (can expand unsanitized data). Not remotely exploitable.
1153 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
1155 PP/02 Make -n do something, by making it not do something.
1156 When combined with -bP, the name of an option is not output.
1158 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
1161 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
1162 $sender_host_name and config options to manage this, and basic check
1165 PP/05 DSCP support for outbound connections and control modifier for inbound.
1167 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
1168 (Only plugin which currently uses this is kerberos4, which nobody should
1169 be using, but we should make it available and other future plugins might
1170 conceivably use it, even though it would break NAT; stuff *should* be
1171 using channel bindings instead).
1173 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
1174 name; added for Sendmail compatibility; requires admin caller.
1175 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
1176 just ignore it); requires trusted caller.
1177 Also parse but ignore: -Ac -Am -X<logfile>
1180 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
1182 TL/02 Add +smtp_confirmation as a default logging option.
1184 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
1185 Patch by Magnus Holmgren from 2007-02-20.
1187 TL/04 Bugzilla 1281 - Spec typo.
1188 Bugzilla 1283 - Spec typo.
1189 Bugzilla 1290 - Spec grammar fixes.
1191 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
1193 TL/06 Add Experimental DMARC support using libopendmarc libraries.
1195 TL/07 Fix an out of order global option causing a segfault. Reported to dev
1196 mailing list by by Dmitry Isaikin.
1198 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
1200 JH/02 Support "G" suffix to numbers in ${if comparisons.
1202 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
1204 NM/01 Bugzilla 1197 - Spec typo
1205 Bugzilla 1196 - Spec examples corrections
1207 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
1209 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
1210 gnutls_enable_pkcs11, but renamed to more accurately indicate its
1213 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
1214 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
1216 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
1217 "acl {{name}{arg}...}", and optional args on acl condition
1220 JH/05 Permit multiple router/transport headers_add/remove lines.
1222 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
1224 JH/07 Avoid using a waiting database for a single-message-only transport.
1225 Performance patch from Paul Fisher. Bugzilla 1262.
1227 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
1230 JH/09 Add $headers_added variable, with content from use of ACL modifier
1231 add_header (but not yet added to the message). Bugzilla 199.
1233 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
1234 Pulled from Bugzilla 817 by Wolfgang Breyha.
1236 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
1238 (nb: this is the same fix as in Exim 4.80.1)
1240 JH/11 Add A= logging on delivery lines, and a client_set_id option on
1243 JH/12 Add optional authenticated_sender logging to A= and a log_selector
1246 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
1248 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
1249 advertise SMTP AUTH mechanism to us, instead of a generic
1250 protocol violation error. Also, make Exim more robust to bad
1251 data from the Dovecot auth socket.
1253 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
1255 When a queue runner is handling a message, Exim first routes the
1256 recipient addresses, during which it prunes them based on the retry
1257 hints database. After that it attempts to deliver the message to
1258 any remaining recipients. It then updates the hints database using
1261 So if a recipient address works intermittently, it can get repeatedly
1262 deferred at routing time. The retry hints record remains fresh so the
1263 address never reaches the final cutoff time.
1265 This is a fairly common occurrence when a user is bumping up against
1266 their storage quota. Exim had some logic in its local delivery code
1267 to deal with this. However it did not apply to per-recipient defers
1268 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
1270 This change adds a proper retry rule check during routing so that the
1271 final cutoff time is checked against the message's age. We only do
1272 this check if there is an address retry record and there is not a
1273 domain retry record; this implies that previous attempts to handle
1274 the address had the retry_use_local_parts option turned on. We use
1275 this as an approximation for the destination being like a local
1276 delivery, as in LMTP.
1278 I suspect this new check makes the old local delivery cutoff check
1279 redundant, but I have not verified this so I left the code in place.
1281 TF/02 Correct gecos expansion when From: is a prefix of the username.
1283 Test 0254 submits a message to Exim with the header
1287 When I ran the test suite under the user fanf2, Exim expanded
1288 the header to contain my full name, whereas it should have added
1289 a Resent-Sender: header. It erroneously treats any prefix of the
1290 username as equal to the username.
1292 This change corrects that bug.
1294 GF/01 DCC debug and logging tidyup
1295 Error conditions log to paniclog rather than rejectlog.
1296 Debug lines prefixed by "DCC: " to remove any ambiguity.
1298 TF/03 Avoid unnecessary rebuilds of lookup-related code.
1300 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
1301 Bug spotted by Jeremy Harris; was flawed since initial commit.
1302 Would have resulted in OCSP responses post-SNI triggering an Exim
1303 NULL dereference and crash.
1305 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
1307 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
1308 Bug detection, analysis and fix by Samuel Thibault.
1309 Bugzilla 1331, Debian bug #698092.
1311 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
1313 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
1314 Server implementation by Todd Lyons, client by JH.
1315 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
1316 config variable "prdr_enable" controls whether the server
1317 advertises the facility. If the client requests PRDR a new
1318 acl_data_smtp_prdr ACL is called once for each recipient, after
1319 the body content is received and before the acl_smtp_data ACL.
1320 The client is controlled by both of: a hosts_try_prdr option
1321 on the smtp transport, and the server advertisement.
1322 Default client logging of deliveries and rejections involving
1323 PRDR are flagged with the string "PRDR".
1325 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
1326 fclose(). Diagnosis by Todd Lyons.
1328 PP/17 Update configure.default to handle IPv6 localhost better.
1329 Patch by Alain Williams (plus minor tweaks).
1332 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
1333 This is now consistent with GnuTLS, and is now documented: the
1334 previous undocumented portable approach to treating the option as
1335 unset was to force an expansion failure. That still works, and
1336 an empty string is now equivalent.
1338 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
1339 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
1340 not performing validation itself.
1342 PP/20 Added force_command boolean option to pipe transport.
1343 Patch from Nick Koston, of cPanel Inc.
1345 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
1348 TF/04 Added udpsend ACL modifier and hexquote expansion operator
1350 PP/21 Fix eximon continuous updating with timestamped log-files.
1351 Broken in a format-string cleanup in 4.80, missed when I repaired the
1352 other false fix of the same issue.
1353 Report and fix from Heiko Schlichting.
1356 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
1357 Report from Prashanth Katuri.
1359 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
1360 It's SecureTransport, so affects any MacOS clients which use the
1361 system-integrated TLS libraries, including email clients.
1363 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
1364 using a MIME ACL for non-SMTP local injection.
1365 Report and assistance in diagnosis by Warren Baker.
1367 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
1369 JH/16 Fix comparisons for 64b. Bugzilla 1385.
1371 TL/09 Add expansion variable $authenticated_fail_id to keep track of
1372 last id that failed so it may be referenced in subsequent ACL's.
1374 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
1377 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
1378 ldap library initialization, allowing self-signed CA's to be
1379 used. Also properly sets require_cert option later in code by
1380 using NULL (global ldap config) instead of ldap handle (per
1381 session). Bug diagnosis and testing by alxgomz.
1383 TL/12 Enhanced documentation in the ratelimit.pl script provided in
1384 the src/util/ subdirectory.
1386 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
1387 renamed to Transport Post Delivery Action by Jeremy Harris, as
1390 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
1391 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
1392 redis_servers = needs to be configured which will be used by the redis
1393 lookup. Patch from Warren Baker, of The Packet Hub.
1395 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
1397 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
1398 hostname or reverse DNS when processing a host list. Used suggestions
1399 from multiple comments on this bug.
1401 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
1403 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
1404 Missed a few lines, added it to make the runtest require no keyboard
1407 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
1408 contains upper case chars. Make router use caseful_local_part.
1410 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
1411 support when GnuTLS has been built with p11-kit.
1417 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
1419 This, or similar/improved, will also be change PP/11 of 4.82.
1425 PP/01 Handle short writes when writing local log-files.
1426 In practice, only affects FreeBSD (8 onwards).
1427 Bugzilla 1053, with thanks to Dmitry Isaikin.
1429 NM/01 Bugzilla 949 - Documentation tweak
1431 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
1434 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
1436 PP/02 Implemented gsasl authenticator.
1438 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
1440 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
1441 `pkg-config foo` for cflags/libs.
1443 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
1444 with rest of GSASL and with heimdal_gssapi.
1446 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
1447 `pkg-config foo` for cflags/libs for the TLS implementation.
1449 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
1450 properties get this fed in as external SSF. A number of robustness
1451 and debugging improvements to the cyrus_sasl authenticator.
1453 PP/08 cyrus_sasl server now expands the server_realm option.
1455 PP/09 Bugzilla 1214 - Log authentication information in reject log.
1456 Patch by Jeremy Harris.
1458 PP/10 Added dbmjz lookup type.
1460 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
1462 PP/12 MAIL args handles TAB as well as SP, for better interop with
1463 non-compliant senders.
1464 Analysis and variant patch by Todd Lyons.
1466 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
1467 Bug report from Lars Müller <lars@samba.org> (via SUSE),
1468 Patch from Dirk Mueller <dmueller@suse.com>
1470 PP/13 tls_peerdn now print-escaped for spool files.
1471 Observed some $tls_peerdn in wild which contained \n, which resulted
1472 in spool file corruption.
1474 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
1475 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
1476 or write after TLS renegotiation, which otherwise led to messages
1479 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
1480 as a tracking header (ie: a signed header comes before the signature).
1481 Patch from Wolfgang Breyha.
1483 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
1484 comma-sep list; embedded commas doubled.
1486 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
1488 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
1490 Report and patch from Dmitry Banschikov.
1492 PP/16 Removed "dont_insert_empty_fragments" from "openssl_options".
1493 Removed SSL_clear() after SSL_new() which led to protocol negotiation
1494 failures. We appear to now support TLS1.1+ with Exim.
1496 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
1497 lets Exim select keys and certificates based upon TLS SNI from client.
1498 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
1499 before an outbound SMTP session. New log_selector, +tls_sni.
1501 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
1502 NULL dereference. Report and patch from Alun Jones.
1504 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
1505 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
1506 Not seeing resolver debug output on NetBSD, but suspect this is a
1507 resolver implementation change.
1509 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
1510 Left warnings. Added "eximon gdb" invocation mode.
1512 PP/21 Defaulting "accept_8bitmime" to true, not false.
1514 PP/22 Added -bw for inetd wait mode support.
1516 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
1517 locate the relevant includes and libraries. Made this the default.
1519 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
1520 Bugzilla 1246, report and most of solution from Tomasz Kusy.
1522 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
1523 This may cause build issues on older platforms.
1525 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
1526 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
1527 gnutls_require_mac & gnutls_require_protocols (no longer supported).
1528 Added SNI support via GnuTLS too.
1529 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
1531 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
1533 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
1534 Applied second patch from Janne, implementing suggestion to default
1535 multiple-strings-in-record handling to match SPF spec.
1537 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
1539 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
1540 read-only, out of scope).
1541 Patch from Wolfgang Breyha, report from Stuart Northfield.
1543 PP/29 Fix three issues highlighted by clang analyser static analysis.
1544 Only crash-plausible issue would require the Cambridge-specific
1545 iplookup router and a misconfiguration.
1546 Report from Marcin Mirosław.
1548 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
1550 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
1551 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
1552 As part of this, removing so much warning spew let me fix some minor
1553 real issues in debug logging.
1555 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
1556 assignment on my part. Fixed.
1558 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
1559 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
1560 Janne Snabb (who went above and beyond: thank you).
1562 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
1563 string otherwise requires a connection and a bunch more work and it's
1564 relatively easy to get wrong. Should also expose TLS library linkage
1567 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
1568 64-bit ${eval} (JH/03).
1570 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
1571 GNU libc to support some of the 64-bit stuff, should not lead to
1572 conflicts. Defined before os.h is pulled in, so if a given platform
1573 needs to override this, it can.
1575 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
1576 protection layer was required, which is not implemented.
1577 Bugzilla 1254, patch from Wolfgang Breyha.
1579 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
1580 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
1581 tls_dhparam take prime identifiers. Also unbreak combination of
1582 OpenSSL+DH_params+TLSSNI.
1584 PP/39 Disable SSLv2 by default in OpenSSL support.
1590 PP/01 Solaris build fix for Oracle's LDAP libraries.
1591 Bugzilla 1109, patch from Stephen Usher.
1593 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
1595 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
1598 TF/02 Fix a couple more cases where we did not log the error message
1599 when unlink() failed. See also change 4.74-TF/03.
1601 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
1602 lock up or crash if it happened to be inside a call to libc when it
1603 got a SIGUSR1 from exiwhat.
1605 The SIGUSR1 handler appends the current process status to the process
1606 log which is later printed by exiwhat. It used to use the general
1607 purpose logging code to do this, but several functions it calls are
1608 not safe for signals.
1610 The new output code in the SIGUSR1 handler is specific to the process
1611 log, and simple enough that it's easy to inspect for signal safety.
1612 Removing some special cases also simplifies the general logging code.
1613 Removing the spurious timestamps from the process log simplifies
1616 TF/04 Improved ratelimit ACL condition.
1618 The /noupdate option has been deprecated in favour of /readonly which
1619 has clearer semantics. The /leaky, /strict, and /readonly update modes
1620 are mutually exclusive. The update mode is no longer included in the
1621 database key; it just determines when the database is updated. (This
1622 means that when you upgrade Exim will forget old rate measurements.)
1624 Exim now checks that the per_* options are used with an update mode that
1625 makes sense for the current ACL. For example, when Exim is processing a
1626 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
1627 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
1628 must specify per_mail/readonly. If you omit the update mode it defaults to
1629 /leaky where that makes sense (as before) or /readonly where required.
1631 The /noupdate option is now undocumented but still supported for
1632 backwards compatibility. It is equivalent to /readonly except that in
1633 ACLs where /readonly is required you may specify /leaky/noupdate or
1634 /strict/noupdate which are treated the same as /readonly.
1636 A useful new feature is the /count= option. This is a generalization
1637 of the per_byte option, so that you can measure the throughput of other
1638 aggregate values. For example, the per_byte option is now equivalent
1639 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
1641 The per_rcpt option has been generalized using the /count= mechanism
1642 (though it's more complicated than the per_byte equivalence). When it is
1643 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
1644 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
1645 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
1646 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
1647 Note that using per_rcpt with a non-readonly update mode in more than
1648 one ACL will cause the recipients to be double-counted. (The per_mail
1649 and per_byte options don't have this problem.)
1651 The handling of very low rates has changed slightly. If the computed rate
1652 is less than the event's count (usually one) then this event is the first
1653 after a long gap. In this case the rate is set to the same as this event's
1654 count, so that the first message of a spam run is counted properly.
1656 The major new feature is a mechanism for counting the rate of unique
1657 events. The new per_addr option counts the number of different
1658 recipients that someone has sent messages to in the last time period. It
1659 behaves like per_rcpt if all the recipient addresses are different, but
1660 duplicate recipient addresses do not increase the measured rate. Like
1661 the /count= option this is a general mechanism, so the per_addr option
1662 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
1663 example, measure the rate that a client uses different sender addresses
1664 with the options per_mail/unique=$sender_address. There are further
1665 details in the main documentation.
1667 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
1669 TF/06 Removed a few PCRE remnants.
1671 TF/07 Automatically extract Exim's version number from tags in the git
1672 repository when doing development or release builds.
1674 PP/02 Raise smtp_cmd_buffer_size to 16kB.
1675 Bugzilla 879. Patch from Paul Fisher.
1677 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
1678 Heavily based on revision 40f9a89a from Simon Arlott's tree.
1681 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
1683 PP/05 Variable $av_failed, true if the AV scanner deferred.
1684 Bugzilla 1078. Patch from John Horne.
1686 PP/06 Stop make process more reliably on build failure.
1687 Bugzilla 1087. Patch from Heiko Schlittermann.
1689 PP/07 Make maildir_use_size_file an _expandable_ boolean.
1690 Bugzilla 1089. Patch from Heiko Schlittermann.
1692 PP/08 Handle ${run} returning more data than OS pipe buffer size.
1693 Bugzilla 1131. Patch from Holger Weiß.
1695 PP/09 Handle IPv6 addresses with SPF.
1696 Bugzilla 860. Patch from Wolfgang Breyha.
1698 PP/10 GnuTLS: support TLS 1.2 & 1.1.
1700 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
1703 PP/11 match_* no longer expand right-hand-side by default.
1704 New compile-time build option, EXPAND_LISTMATCH_RHS.
1705 New expansion conditions, "inlist", "inlisti".
1707 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
1709 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
1711 PP/14 fix log_write() format string regression from TF/03.
1712 Bugzilla 1152. Patch from Dmitry Isaikin.
1718 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
1720 PP/02 Harmonised TLS library version reporting; only show if debugging.
1721 Layout now matches that introduced for other libraries in 4.74 PP/03.
1723 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
1725 PP/04 New "dns_use_edns0" global option.
1727 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
1730 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
1731 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
1733 TK/01 Updated PolarSSL code to 0.14.2.
1734 Bugzilla 1097. Patch from Andreas Metzler.
1736 PP/07 Catch divide-by-zero in ${eval:...}.
1737 Fixes bugzilla 1102.
1739 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
1742 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
1743 format-string attack -- SECURITY: remote arbitrary code execution.
1745 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
1746 time unintentionally subject to list matching rules, letting the header
1747 cause arbitrary Exim lookups (of items which can occur in lists, *not*
1748 arbitrary string expansion). This allowed for information disclosure.
1750 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
1751 INT_MIN/-1 -- value coerced to INT_MAX.
1757 NM/01 Workaround for PCRE version dependency in version reporting
1760 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
1761 This fixes portability to compilers other than gcc, notably
1762 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
1764 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
1765 makefiles for portability to HP-UX and POSIX correctness.
1767 PP/01 Permit LOOKUP_foo enabling on the make command-line.
1768 Also via indented variable definition in the Makefile.
1769 (Debugging by Oliver Heesakkers).
1771 PP/02 Restore caching of spamd results with expanded spamd_address.
1772 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
1774 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
1775 Improves build reliability. Fix from: Frank Elsner
1777 NM/02 Fix wide character breakage in the rfc2047 coding
1778 Fixes bug 1064. Patch from Andrey N. Oktyabrski
1780 NM/03 Allow underscore in dnslist lookups
1781 Fixes bug 1026. Patch from Graeme Fowler
1783 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
1784 Code patches from Adam Ciarcinski of NetBSD.
1786 NM/04 Fixed exiqgrep to cope with mailq missing size issue
1789 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
1790 is logged, to avoid truncation. Patch from John Horne.
1792 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
1793 Patch from Jakob Hirsch.
1795 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
1796 SQL string expansion failure details.
1797 Patch from Andrey Oktyabrski.
1799 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
1800 Patch from Simon Arlott.
1802 PP/09 New lookups functionality failed to compile on old gcc which rejects
1803 extern declarations in function scope.
1804 Patch from Oliver Fleischmann
1806 PP/10 Use sig_atomic_t for flags set from signal handlers.
1807 Check getgroups() return and improve debugging.
1808 Fixed developed for diagnosis in bug 927 (which turned out to be
1811 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
1812 Patch from Mark Zealey.
1814 PP/12 Bugzilla 1056: Improved spamd server selection.
1815 Patch from Mark Zealey.
1817 PP/13 Bugzilla 1086: Deal with maildir quota file races.
1818 Based on patch from Heiko Schlittermann.
1820 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
1821 Patch from Uwe Doering, sign-off by Michael Haardt.
1823 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
1824 variable declaration deep within a block. Bug and patch from
1827 PP/15 lookups-Makefile IRIX compatibility coercion.
1829 PP/16 Make DISABLE_DKIM build knob functional.
1831 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
1832 Patch by Simon Arlott
1834 TF/03 Fix valgrind.h portability to C89 compilers that do not support
1835 variable argument macros. Our copy now differs from upstream.
1841 TF/01 Failure to get a lock on a hints database can have serious
1842 consequences so log it to the panic log.
1844 TF/02 Log LMTP confirmation messages in the same way as SMTP,
1845 controlled using the smtp_confirmation log selector.
1847 TF/03 Include the error message when we fail to unlink a spool file.
1849 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
1850 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
1851 for maintaining out-of-tree patches for some time.
1853 PP/01 Bugzilla 139: Documentation and portability issues.
1854 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
1855 Handle per-OS dynamic-module compilation flags.
1857 PP/02 Let /dev/null have normal permissions.
1858 The 4.73 fixes were a little too stringent and complained about the
1859 permissions on /dev/null. Exempt it from some checks.
1860 Reported by Andreas M. Kirchwitz.
1862 PP/03 Report version information for many libraries, including
1863 Exim version information for dynamically loaded libraries. Created
1864 version.h, now support a version extension string for distributors
1865 who patch heavily. Dynamic module ABI change.
1867 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
1868 privilege escalation vulnerability whereby the Exim run-time user
1869 can cause root to append content of the attacker's choosing to
1872 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
1875 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
1876 If dropping privileges for untrusted macros, we disabled normal logging
1877 on the basis that it would fail; for the Exim run-time user, this is not
1878 the case, and it resulted in successful deliveries going unlogged.
1879 Fixed. Reported by Andreas Metzler.
1885 PP/01 Date: & Message-Id: revert to normally being appended to a message,
1886 only prepend for the Resent-* case. Fixes regression introduced in
1887 Exim 4.70 by NM/22 for Bugzilla 607.
1889 PP/02 Include check_rfc2047_length in configure.default because we're seeing
1890 increasing numbers of administrators be bitten by this.
1892 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
1894 PP/03 Bugzilla 994: added openssl_options main configuration option.
1896 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
1898 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
1900 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
1902 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
1903 without a peer certificate, leading to a segfault because of an
1904 assumption that peers always have certificates. Be a little more
1905 paranoid. Problem reported by Martin Tscholak.
1907 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
1908 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
1909 NB: ClamAV planning to remove STREAM in "middle of 2010".
1910 CL also introduces -bmalware, various -d+acl logging additions and
1911 more caution in buffer sizes.
1913 PP/09 Implemented reverse_ip expansion operator.
1915 PP/10 Bugzilla 937: provide a "debug" ACL control.
1917 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
1919 PP/12 Bugzilla 973: Implement --version.
1921 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
1923 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
1925 PP/15 Bugzilla 816: support multiple condition rules on Routers.
1927 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
1928 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
1929 ignore trailing whitespace.
1931 JJ/02 prevent non-panic DKIM error from being sent to paniclog
1933 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
1936 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
1937 Notification from Dr Andrew Aitchison.
1939 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
1940 ExtendedDetectionInfo response format.
1941 Notification from John Horne.
1943 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
1946 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
1947 XSL and documented dependency on system catalogs, with examples of how
1950 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
1953 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
1954 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
1955 configuration file which is writeable by the Exim user or group.
1957 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
1958 of configuration files to cover files specified with the -C option if
1959 they are going to be used with root privileges, not just the default
1962 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
1963 option (effectively making it always true).
1965 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
1966 files to be used while preserving root privileges.
1968 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
1969 that rogue child processes cannot use them.
1971 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
1972 run-time user, instead of root.
1974 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
1975 Exim run-time user without dropping privileges.
1977 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
1978 result string, instead of calling string_vformat() twice with the same
1981 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
1982 for other users. Others should always drop root privileges if they use
1983 -C on the command line, even for a whitelisted configure file.
1985 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
1987 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
1993 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
1994 $data_path, and $header_path variables; fixed documentation bugs and
1997 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
1998 exipick to access non-standard spools, including the "frozen" queue
2001 NM/01 Bugzilla 965: Support mysql stored procedures.
2002 Patch from Alain Williams
2004 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
2006 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
2007 Patch from Andreas Metzler
2009 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
2010 Patch from Kirill Miazine
2012 NM/05 Bugzilla 671: Added umask to procmail example.
2014 JJ/03 installed exipick 20100323.0, fixing doc bug
2016 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
2017 directory. Notification and patch from Dan Rosenberg.
2019 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
2021 TK/02 Improve log output when DKIM signing operation fails.
2023 MH/01 Treat the transport option dkim_domain as a colon separated
2024 list, not as a single string, and sign the message with each element,
2025 omitting multiple occurences of the same signer.
2027 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
2028 Bugzilla 985, 986. Patch by Simon Arlott
2030 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
2031 Patch by Simon Arlott
2033 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
2034 MBX locking. Notification from Dan Rosenberg.
2040 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
2042 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
2044 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
2046 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
2048 NM/04 Bugzilla 915: Flag broken perl installation during build.
2054 TK/01 Added patch by Johannes Berg that expands the main option
2055 "spamd_address" if it starts with a dollar sign.
2057 TK/02 Write list of recipients to X-Envelope-Sender header when building
2058 the mbox-format spool file for content scanning (suggested by Jakob
2061 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
2062 (http://www.dcc-servers.net/) support via dccifd. Activated by
2063 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
2065 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
2066 by Mark Daniel Reidel <mr@df.eu>.
2068 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
2069 When building exim an external PCRE library is now needed -
2070 PCRE is a system library on the majority of modern systems.
2071 See entry on PCRE_LIBS in EDITME file.
2073 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
2074 conversation. Added nologin parameter to request.
2075 Patch contributed by Kirill Miazine.
2077 TF/01 Do not log submission mode rewrites if they do not change the address.
2079 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
2081 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
2082 log files in place. Contributed by Roberto Lima.
2084 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
2086 TF/03 Bugzilla 615: When checking the local_parts router precondition
2087 after a local_part_suffix or local_part_prefix option, Exim now
2088 does not use the address's named list lookup cache, since this
2089 contains cached lookups for the whole local part.
2091 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
2092 Robert Millan. Documentation is in experimental-spec.txt.
2094 TF/04 Bugzilla 668: Fix parallel build (make -j).
2096 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
2098 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
2099 Patch provided by Jan Srzednicki.
2101 TF/05 Leading white space used to be stripped from $spam_report which
2102 wrecked the formatting. Now it is preserved.
2104 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
2105 that they are available at delivery time.
2107 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
2109 TF/08 TLS error reporting now respects the incoming_interface and
2110 incoming_port log selectors.
2112 TF/09 Produce a more useful error message if an SMTP transport's hosts
2113 setting expands to an empty string.
2115 NM/06 Bugzilla 744: EXPN did not work under TLS.
2116 Patch provided by Phil Pennock.
2118 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
2119 Patch provided by Richard Godbee.
2121 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
2122 acl_smtp_notquit, added index entry.
2124 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
2125 Patch provided by Eugene Bujak.
2127 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
2128 accept(). Patch provided by Maxim Dounin.
2130 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
2131 Patch provided by Phil Pennock.
2133 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
2135 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
2136 Patch provided by Brad "anomie" Jorsch.
2138 NM/14 Bugzilla 622: Added timeout setting to transport filter.
2139 Patch provided by Dean Brooks.
2141 TK/05 Add native DKIM support (does not depend on external libraries).
2143 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
2144 Patch provided by Graeme Fowler.
2146 NM/16 Bugzilla 851: Documentation example syntax fix.
2148 NM/17 Changed NOTICE file to remove references to embedded PCRE.
2150 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
2153 NM/19 Bugzilla 745: TLS version reporting.
2154 Patch provided by Phil Pennock.
2156 NM/20 Bugzilla 167: bool: condition support.
2157 Patch provided by Phil Pennock.
2159 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
2160 clients. Patch provided by Phil Pennock.
2162 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
2163 Patch provided by Brad "anomie" Jorsch.
2165 NM/23 Bugzilla 687: Fix misparses in eximstats.
2166 Patch provided by Heiko Schlittermann.
2168 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
2169 Patch provided by Heiko Schlittermann.
2171 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
2172 plus update to original patch.
2174 NM/26 Bugzilla 799: Documentation correction for ratelimit.
2176 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
2177 Patch provided by David Brownlee.
2179 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
2181 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
2183 NM/30 Bugzilla 888: TLS documentation bugfixes.
2185 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
2187 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
2188 Unlike the original bugzilla I have changed all shell scripts in src tree.
2190 NM/33 Bugzilla 898: Transport filter timeout fix.
2191 Patch by Todd Rinaldo.
2193 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
2194 Patch by Serge Demonchaux.
2196 NM/35 Bugzilla 39: Base64 decode bug fixes.
2197 Patch by Jakob Hirsch.
2199 NM/36 Bugzilla 909: Correct connect() call in dcc code.
2201 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
2203 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
2205 NM/39 Bugzilla 911: Fixed MakeLinks build script.
2211 TK/01 Add preliminary DKIM support. Currently requires a forked version of
2212 ALT-N's libdkim that I have put here:
2213 http://duncanthrax.net/exim-experimental/
2215 Note to Michael Haardt: I had to rename some vars in sieve.c. They
2216 were called 'true' and it seems that C99 defines that as a reserved
2217 keyword to be used with 'bool' variable types. That means you could
2218 not include C99-style headers which use bools without triggering
2219 build errors in sieve.c.
2221 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
2222 as mailq or other aliases. Changed the --help handling significantly
2223 to do whats expected. exim_usage() emits usage/help information.
2225 SC/01 Added the -bylocaldomain option to eximstats.
2227 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
2229 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
2231 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
2237 PH/01 Another patch from the Sieve maintainer.
2239 PH/02 When an IPv6 address is converted to a string for single-key lookup
2240 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
2241 dots are used instead of colons so that keys in lsearch files need not
2242 contain colons. This was done some time before quoting was made available
2243 in lsearch files. However, iplsearch files do require colons in IPv6 keys
2244 (notated using the quote facility) so as to distinguish them from IPv4
2245 keys. This meant that lookups for IP addresses in host lists did not work
2246 for iplsearch lookups.
2248 This has been fixed by arranging for IPv6 addresses to be expressed with
2249 colons if the lookup type is iplsearch. This is not incompatible, because
2250 previously such lookups could never work.
2252 The situation is now rather anomalous, since one *can* have colons in
2253 ordinary lsearch keys. However, making the change in all cases is
2254 incompatible and would probably break a number of configurations.
2256 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
2259 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
2260 conversion specification without a maximum field width, thereby enabling
2261 a rogue spamd server to cause a buffer overflow. While nobody in their
2262 right mind would setup Exim to query an untrusted spamd server, an
2263 attacker that gains access to a server running spamd could potentially
2264 exploit this vulnerability to run arbitrary code as the Exim user.
2266 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
2267 $primary_hostname instead of what libspf2 thinks the hosts name is.
2269 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
2270 a directory entry by the name of the lookup key. Previously, if a
2271 symlink pointed to a non-existing file or a file in a directory that
2272 Exim lacked permissions to read, a lookup for a key matching that
2273 symlink would fail. Now it is enough that a matching directory entry
2274 exists, symlink or not. (Bugzilla 503.)
2276 PH/03 The body_linecount and body_zerocount variables are now exported in the
2279 PH/04 Added the $dnslist_matched variable.
2281 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
2282 This means they are set thereafter only if the connection becomes
2285 PH/06 Added the client_condition to authenticators so that some can be skipped
2286 by clients under certain conditions.
2288 PH/07 The error message for a badly-placed control=no_multiline_responses left
2289 "_responses" off the end of the name.
2291 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
2293 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
2294 (without spaces) instead of just copying the configuration text.
2296 PH/10 Added the /noupdate option to the ratelimit ACL condition.
2298 PH/11 Added $max_received_linelength.
2300 PH/12 Added +ignore_defer and +include_defer to host lists.
2302 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
2303 way in which PCRE > 7.0 is built.
2305 PH/14 Implemented queue_only_load_latch.
2307 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
2308 MAIL command. The effect was to mangle the value on 64-bit systems.
2310 PH/16 Another patch from the Sieve maintainer.
2312 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
2314 PH/18 If a system quota error occurred while trying to create the file for
2315 a maildir delivery, the message "Mailbox is full" was not appended to the
2316 bounce if the delivery eventually timed out. Change 4.67/27 below applied
2317 only to a quota excession during the actual writing of the file.
2319 PH/19 It seems that peer DN values may contain newlines (and other non-printing
2320 characters?) which causes problems in log lines. The DN values are now
2321 passed through string_printing() before being added to log lines.
2323 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
2324 and InterBase are left for another time.)
2326 PH/21 Added message_body_newlines option.
2328 PH/22 Guard against possible overflow in moan_check_errorcopy().
2330 PH/23 POSIX allows open() to be a macro; guard against that.
2332 PH/24 If the recipient of an error message contained an @ in the local part
2333 (suitably quoted, of course), incorrect values were put in $domain and
2334 $local_part during the evaluation of errors_copy.
2340 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
2341 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
2344 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
2345 issue a MAIL command.
2347 PH/02 In an ACL statement such as
2349 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
2351 if a client was not listed at all, or was listed with a value other than
2352 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
2353 the condition was not true (as it should be), so access was not denied.
2354 The bug was that the ! inversion was incorrectly passed on to the second
2355 item. This has been fixed.
2357 PH/03 Added additional dnslists conditions == and =& which are different from
2358 = and & when the dns lookup returns more than one IP address.
2360 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
2361 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
2363 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
2364 FSYNC, which compiles an option called disable_fsync that allows for
2365 bypassing fsync(). The documentation is heavily laced with warnings.
2367 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
2369 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
2370 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
2371 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
2372 including adding "make clean"; (3) Added -fPIC when compiling the test
2373 dynamically loaded module, to get rid of a warning.
2375 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
2376 message fails, move_frozen_messages = true and ignore_bounce_errors_after
2377 = 0s. The bug is otherwise harmless.
2379 PH/07 There was a bug in the dovecot authenticator such that the value of
2380 $auth1 could be overwritten, and so not correctly preserved, after a
2381 successful authentication. This usually meant that the value preserved by
2382 the server_setid option was incorrect.
2384 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
2386 PH/09 Installed PCRE release 7.0.
2388 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
2389 run for batched SMTP input. It is now run at the start of every message
2390 in the batch. While fixing this I discovered that the process information
2391 (output by running exiwhat) was not always getting set for -bs and -bS
2392 input. This is fixed, and it now also says "batched" for BSMTP.
2394 PH/11 Added control=no_pipelining.
2396 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
2397 patch, slightly modified), and move the expansion of helo_data till after
2398 the connection is made in the smtp transport (so it can use these
2401 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
2403 PH/14 Added log_selector = +pid.
2405 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
2407 PH/16 Add ${if forany and ${if forall.
2409 PH/17 Added dsn_from option to vary the From: line in DSNs.
2411 PH/18 Flush SMTP output before performing a callout, unless control =
2412 no_callout_flush is set.
2414 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
2415 was true (the default) a successful delivery failed to delete the retry
2416 item, thus causing premature timeout of the address. The bug is now
2419 PH/20 Added hosts_avoid_pipelining to the smtp transport.
2421 PH/21 Long custom messages for fakedefer and fakereject are now split up
2422 into multiline responses in the same way that messages for "deny" and
2423 other ACL rejections are.
2425 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
2426 with slight modification.
2428 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
2429 draft, changing the syntax and factoring some duplicate code".
2431 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
2432 for deliveries of the second and subsequent messages over the same SMTP
2435 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
2436 ${reduce, with only minor "tidies".
2438 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
2440 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
2441 expansion side effects.
2443 PH/27 When a message times out after an over-quota error from an Exim-imposed
2444 quota, the bounce message says "mailbox is full". This message was not
2445 being given when it was a system quota that was exceeded. It now should
2448 MH/03 Made $recipients available in local_scan(). local_scan() already has
2449 better access to the recipient list through recipients_list[], but
2450 $recipients can be useful in postmaster-provided expansion strings.
2452 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
2453 in the case of a MAIL command with additional options following the
2454 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
2455 were accidentally chopped off.
2457 PH/29 SMTP synchronization checks are implemented when a command is read -
2458 there is a check that no more input is waiting when there shouldn't be
2459 any. However, for some commands, a delay in an ACL can mean that it is
2460 some time before the response is written. In this time, more input might
2461 arrive, invalidly. So now there are extra checks after an ACL has run for
2462 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
2463 pipelining has not been advertised.
2465 PH/30 MH's patch to allow iscntrl() characters to be list separators.
2467 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
2468 returned in the SMTP response when smtp_return_error_details was false.
2469 This has been fixed.
2471 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
2472 instead of the C I/O that was originally supplied, because problems were
2473 reported on Solaris.
2475 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
2476 Exim which did not show up earlier: it was assuming that a call to
2477 SSL_CTX_set_info_callback() might give an error value. In fact, there is
2478 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
2479 was a macro that became an assignment, so it seemed to work. This has
2480 changed to a proper function call with a void return, hence the compile
2481 error. Exim's code has been fixed.
2483 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
2486 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
2488 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
2490 PH/37 If a message is not accepted after it has had an id assigned (e.g.
2491 because it turns out to be too big or there is a timeout) there is no
2492 "Completed" line in the log. When some messages of this type were
2493 selected by exigrep, they were listed as "not completed". Others were
2494 picked up by some special patterns. I have improved the selection
2495 criteria to be more general.
2497 PH/38 The host_find_failed option in the manualroute router can now be set
2498 to "ignore", to completely ignore a host whose IP address cannot be
2499 found. If all hosts are ignored, the behaviour is controlled by the new
2500 host_all_ignored option.
2502 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
2503 homing or because of multiple MX records with /mx) generated more than
2504 one IP address, and the following item turned out to be the local host,
2505 all the secondary addresses of the first item were incorrectly removed
2506 from the list, along with the local host and any following hosts (which
2507 is what is supposed to happen).
2509 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
2510 whoever called Exim into the -H file. In the case of the daemon it was
2511 behaving confusingly. When first started, it used values for whoever
2512 started the daemon, but after a SIGHUP it used the Exim user (because it
2513 calls itself on a restart). I have changed the code so that it now always
2516 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
2517 message are rejected with the same error (e.g. no authentication or bad
2518 sender address), and a DATA command is nevertheless sent (as can happen
2519 with PIPELINING or a stupid MUA), the error message that was given to the
2520 RCPT commands is included in the rejection of the DATA command. This is
2521 intended to be helpful for MUAs that show only the final error to their
2524 PH/42 Another patch from the Sieve maintainer.
2526 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
2527 Eximstats - Fixed some broken HTML links and added missing column headers
2529 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
2530 columns for Rejects, Temp Rejects, Ham, and Spam rows.
2532 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
2534 PH/43 Yet another patch from the Sieve maintainer.
2536 PH/44 I found a way to check for a TCP/IP connection going away before sending
2537 the response to the final '.' that terminates a message, but only in the
2538 case where the client has not sent further data following the '.'
2539 (unfortunately, this is allowed). However, in many cases there won't be
2540 any further data because there won't be any more messages to send. A call
2541 to select() can be used: if it shows that the input is "ready", there is
2542 either input waiting, or the socket has been closed. An attempt to read
2543 the next input character can distinguish the two cases. Previously, Exim
2544 would have sent an OK response which the client would never have see.
2545 This could lead to message repetition. This fix should cure that, at
2546 least in a lot of common cases.
2548 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
2549 advertised in response to EHLO.
2555 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
2556 fixed by 4.65/MH/01 (is this a record?) are fixed:
2558 (i) An empty string was always treated as zero by the numeric comparison
2559 operators. This behaviour has been restored.
2561 (ii) It is documented that the numeric comparison operators always treat
2562 their arguments as decimal numbers. This was broken in that numbers
2563 starting with 0 were being interpreted as octal.
2565 While fixing these problems I realized that there was another issue that
2566 hadn't been noticed. Values of message_size_limit (both the global option
2567 and the transport option) were treated as octal if they started with 0.
2568 The documentation was vague. These values are now always treated as
2569 decimal, and I will make that clear in the documentation.
2575 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
2576 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
2579 MH/01 Don't check that the operands of numeric comparison operators are
2580 integers when their expansion is in "skipping" mode (fixes bug
2581 introduced by 4.64-PH/07).
2583 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
2584 child addresses, Exim now panics and dies. Previously, because the count
2585 is held in a short int, deliveries were likely to be lost. As such a
2586 large number of recipients for a single message is ridiculous
2587 (performance will be very, very poor), I have chosen to impose a limit
2588 rather than extend the field.
2594 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
2595 leftover -K file (the existence of which was triggered by #402).
2596 While we were at it, introduced process PID as part of the -K
2597 filename. This should rule out race conditions when creating
2600 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
2601 processing considerably. Previous code took too long for large mails,
2602 triggering a timeout which in turn triggers #401.
2604 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
2605 in the DK code in transports.c. sendfile() is not really portable,
2606 hence the _LINUX specificness.
2608 TF/01 In the add_headers option to the mail command in an Exim filter,
2609 there was a bug that Exim would claim a syntax error in any
2610 header after the first one which had an odd number of characters
2613 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
2614 callout verification, Exim cached a "reject" for the entire domain. This
2615 is correct for most verifications, but it is not correct for a recipient
2616 verification with use_sender or use_postmaster set, because in that case
2617 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
2618 case of MAIL FROM:<> rejection from other early rejections (e.g.
2619 rejection of HELO). When verifying a recipient using a non-null MAIL
2620 address, the cache is ignored if it shows MAIL FROM:<> rejection.
2621 Whatever the result of the callout, the value of the domain cache is
2622 left unchanged (for any other kind of callout, getting as far as trying
2623 RCPT means that the domain itself is ok).
2625 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
2628 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
2629 manifest itself as EPIPE rather than ECONNECT. When tidying away a
2630 session, the daemon ignores ECONNECT errors and logs others; it now
2631 ignores EPIPE as well.
2633 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
2634 (quoted-printable decoding).
2636 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
2637 later the small subsequent patch to fix an introduced bug.
2639 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
2641 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
2643 PH/08 An error is now given if message_size_limit is specified negative.
2645 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
2646 to be given (somewhat) arbitrary names.
2648 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
2651 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
2652 miscellaneous code fixes
2654 PH/10 Added the log_reject_target ACL modifier to specify where to log
2657 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
2658 hostname. This is wrong, because it relates to the incoming message (and
2659 probably the interface on which it is arriving) and not to the outgoing
2660 callout (which could be using a different interface). This has been
2661 changed to use the value of the helo_data option from the smtp transport
2662 instead - this is what is used when a message is actually being sent. If
2663 there is no remote transport (possible with a router that sets up host
2664 addresses), $smtp_active_hostname is used.
2666 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
2667 tweaks were necessary in order to get it to work (see also 21 below):
2668 (a) The code assumed that strncpy() returns a negative number on buffer
2669 overflow, which isn't the case. Replaced with Exim's string_format()
2671 (b) There were several signed/unsigned issues. I just did the minimum
2672 hacking in of casts. There is scope for a larger refactoring.
2673 (c) The code used strcasecmp() which is not a standard C function.
2674 Replaced with Exim's strcmpic() function.
2675 (d) The code set only $1; it now sets $auth1 as well.
2676 (e) A simple test gave the error "authentication client didn't specify
2677 service in request". It would seem that Dovecot has changed its
2678 interface. Fortunately there's a specification; I followed it and
2679 changed what the client sends and it appears to be working now.
2681 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
2684 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
2685 address A is aliased to B and C, where B exists and C does not. Without
2686 -v the output is "A verified" because verification stops after a
2687 successful redirection if more than one address is generated. However,
2688 with -v the child addresses are also verified. Exim was outputting "A
2689 failed to verify" and then showing the successful verification for C,
2690 with its parentage. It now outputs "B failed to verify", showing B's
2691 parentage before showing the successful verification of C.
2693 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
2694 look up a TXT record in a specific list after matching in a combined
2697 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
2698 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
2699 they consult the DNS. I had assumed they would set it the way they
2700 wanted; and indeed my experiments on Linux seem to show that in some
2701 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
2702 To be on the safe side, however, I have now made the interface to
2703 host_find_byname() similar to host_find_bydns(), with an argument
2704 containing the DNS resolver options. The host_find_byname() function now
2705 sets these options at its start, just as host_find_bydns() does. The smtp
2706 transport options dns_qualify_single and dns_search_parents are passed to
2707 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
2708 of host_find_byname() use the default settings of RES_DEFNAMES
2709 (qualify_single) but not RES_DNSRCH (search_parents).
2711 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
2712 spool_read_header() do less string testing, by means of a preliminary
2713 switch on the second character of optional "-foo" lines. (This is
2714 overdue, caused by the large number of possibilities that now exist.
2715 Originally there were few.) While I was there, I also converted the
2716 str(n)cmp tests so they don't re-test the leading "-" and the first
2717 character, in the hope this might squeeze out yet more improvement.
2719 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
2720 flag allowing group syntax was set by the header_syntax check but not
2721 turned off, possible causing trouble later; (2) The flag was not being
2722 set at all for the header_verify test, causing "group"-style headers to
2723 be rejected. I have now set it in this case, and also caused header_
2724 verify to ignore an empty address taken from a group. While doing this, I
2725 came across some other cases where the code for allowing group syntax
2726 while scanning a header line wasn't quite right (mostly, not resetting
2727 the flag correctly in the right place). These bugs could have caused
2728 trouble for malformed header lines. I hope it is now all correct.
2730 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
2731 with the "reply" argument non-NULL. The code, however (which originally
2732 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
2733 but it didn't always do it. This confused somebody who was copying the
2734 code for some other use. I have removed all the tests.
2736 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
2737 feature that was used to support insecure browsers during the U.S. crypto
2738 embargo. It requires special client support, and Exim is probably the
2739 only MTA that supported it -- and would never use it because real RSA is
2740 always available. This code has been removed, because it had the bad
2741 effect of slowing Exim down by computing (never used) parameters for the
2742 RSA_EXPORT functionality.
2744 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
2745 authenticator to fail if there's a tab character in the incoming data
2746 (there should never be unless someone is messing about, as it's supposed
2747 to be base64-encoded). Also added, on Timo's advice, the "secured" option
2748 if the connection is using TLS or if the remote IP is the same as the
2749 local IP, and the "valid-client-cert option" if a client certificate has
2752 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
2753 authenticators. This can be used for authorization after authentication
2754 succeeds. (In the case of plaintext, it servers for both authentication
2757 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
2758 if any retry times were supplied.
2760 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
2761 connection, where there is no client IP address to check. In this
2762 situation, the verify now always succeeds.
2764 PH/25 Applied John Jetmore's -Mset patch.
2766 PH/26 Added -bem to be like -Mset, but loading a message from a file.
2768 PH/27 In a string expansion for a processed (not raw) header when multiple
2769 headers of the same name were present, leading whitespace was being
2770 removed from all of them, but trailing whitespace was being removed only
2771 from the last one. Now trailing whitespace is removed from each header
2772 before concatenation. Completely empty headers in a concatenation (as
2773 before) are ignored.
2775 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
2776 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
2778 PH/29 [Removed. This was a change that I later backed out, and forgot to
2779 correct the ChangeLog entry (that I had efficiently created) before
2780 committing the later change.]
2782 PH/30 Exim was sometimes attempting to deliver messages that had suffered
2783 address errors (4xx response to RCPT) over the same connection as other
2784 messages routed to the same hosts. Such deliveries are always "forced",
2785 so retry times are not inspected. This resulted in far too many retries
2786 for the affected addresses. The effect occurred only when there were more
2787 hosts than the hosts_max_try setting in the smtp transport when it had
2788 the 4xx errors. Those hosts that it had tried were not added to the list
2789 of hosts for which the message was waiting, so if all were tried, there
2790 was no problem. Two fixes have been applied:
2792 (i) If there are any address or message errors in an SMTP delivery, none
2793 of the hosts (tried or untried) are now added to the list of hosts
2794 for which the message is waiting, so the message should not be a
2795 candidate for sending over the same connection that was used for a
2796 successful delivery of some other message. This seems entirely
2797 reasonable: after all the message is NOT "waiting for some host".
2798 This is so "obvious" that I'm not sure why it wasn't done
2799 previously. Hope I haven't missed anything, but it can't do any
2800 harm, as the worst effect is to miss an optimization.
2802 (ii) If, despite (i), such a delivery is accidentally attempted, the
2803 routing retry time is respected, so at least it doesn't keep
2804 hammering the server.
2806 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
2807 in ${readsocket because some servers need this prod.
2809 PH/32 Added some extra debug output when updating a wait-xxx database.
2811 PH/33 The hint "could be header name not terminated by colon", which has been
2812 given for certain expansion errors for a long time, was not being given
2813 for the ${if def:h_colon_omitted{... case.
2815 PH/34 The spec says: "With one important exception, whenever a domain list is
2816 being scanned, $domain contains the subject domain." There was at least
2817 one case where this was not true.
2819 PH/35 The error "getsockname() failed: connection reset by peer" was being
2820 written to the panic log as well as the main log, but it isn't really
2821 panic-worthy as it just means the connection died rather early on. I have
2822 removed the panic log writing for the ECONNRESET error when getsockname()
2825 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
2826 runs only) independently of the message's sender address. This meant
2827 that, if the 4xx error was in fact related to the sender, a different
2828 message to the same recipient with a different sender could confuse
2829 things. In particular, this can happen when sending to a greylisting
2830 server, but other circumstances could also provoke similar problems.
2831 I have changed the default so that the retry time for these errors is now
2832 based a combination of the sender and recipient addresses. This change
2833 can be overridden by setting address_retry_include_sender=false in the
2836 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
2837 remote server are returned as part of bounce messages. This was not
2838 happening for LMTP over a pipe (the lmtp transport), but now it is the
2839 same for both kinds of LMTP.
2841 PH/38 Despite being documented as not happening, Exim was rewriting addresses
2842 in header lines that were in fact CNAMEs. This is no longer the case.
2844 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
2845 and queue runs started by the daemon processed all messages. This has
2846 been fixed so that -R and -S can now usefully be given with -q<time>.
2848 PH/40 Import PCRE release 6.7 (fixes some bugs).
2850 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
2852 PH/42 Give an error if -q is specified more than once.
2854 PH/43 Renamed the variables $interface_address and $interface_port as
2855 $received_ip_address and $received_port, to make it clear that these
2856 values apply to message reception, and not to the outgoing interface when
2857 a message is delivered. (The old names remain recognized, of course.)
2859 PH/44 There was no timeout on the connect() call when using a Unix domain
2860 socket in the ${readsocket expansion. There now is.
2862 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
2863 be meaningful with "accept".
2865 SC/01 Eximstats V1.43
2866 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
2868 SC/02 Eximstats V1.44
2869 Use a glob alias rather than an array ref in the generated
2870 parser. This improves both readability and performance.
2872 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
2873 Collect SpamAssassin and rejection statistics.
2874 Don't display local sender or destination tables unless
2875 there is data to show.
2876 Added average volumes into the top table text output.
2878 SC/04 Eximstats V1.46
2879 Collect data on the number of addresses (recipients)
2880 as well as the number of messages.
2882 SC/05 Eximstats V1.47
2883 Added 'Message too big' to the list of mail rejection
2884 reasons (thanks to Marco Gaiarin).
2886 SC/06 Eximstats V1.48
2887 Mainlog lines which have GMT offsets and are too short to
2888 have a flag are now skipped.
2890 SC/07 Eximstats V1.49 (Alain Williams)
2891 Added the -emptyok flag.
2893 SC/08 Eximstats V1.50
2894 Fixes for obtaining the IP address from reject messages.
2896 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
2897 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
2898 whitespace changes from 4.64-PH/27
2900 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
2903 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
2904 are found, allow negative numbers in numeric criteria)
2906 JJ/06 exipick.20061117.2, added new $message_body_missing variable
2908 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
2909 to match changes made in 4.64-PH/43
2911 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
2913 PH/47 Put in an explicit test for a DNS lookup of an address record where the
2914 "domain" is actually an IP address, and force a failure. This locks out
2915 those revolvers/nameservers that support "A-for-A" lookups, in
2916 contravention of the specifications.
2918 PH/48 When a host name was looked up from an IP address, and the subsequent
2919 forward lookup of the name timed out, the host name was left in
2920 $sender_host_name, contrary to the specification.
2922 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
2923 restricted to single-key lookups, Exim was not diagnosing an error if
2924 * or *@ was used with a query-style lookup.
2926 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
2928 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
2929 long ago, but noone interested enough thought of it. Let's just say that
2930 the "1.1" means that there are some new functions that weren't there at
2931 some point in the past.
2933 PH/51 Error processing for expansion failure of helo_data from an smtp
2934 transport during callout processing was broken.
2936 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
2937 tested/used via the -bh/-bhc/-bs options.
2939 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
2940 bug, fixed in subsequent PCRE releases).
2942 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
2943 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
2945 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
2951 SC/01 Use a glob alias rather than an array ref in eximstats generated
2952 parser. This improves both readability and performance.
2954 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
2955 Don't display local sender or destination tables in eximstats unless
2956 there is data to show.
2957 Added average volumes into the eximstats top table text output.
2959 SC/03 Collect data on the number of addresses (recipients) as well
2960 as the number of messages in eximstats.
2962 TF/01 Correct an error in the documentation for the redirect router. Exim
2963 does (usually) call initgroups() when daemonizing.
2965 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
2966 with consistent privilege compared to when running as a daemon.
2968 TF/03 Note in the spec that $authenticated_id is not set for local
2969 submissions from trusted users.
2971 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
2972 Thanks to Dean Brooks <dean@iglou.com> for the patch.
2974 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
2975 by adding some example configuration directives to the default
2976 configuration file. A little bit of work is required to uncomment the
2977 directives and define how usernames and passwords are checked, but
2978 there is now a framework to start from.
2980 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
2981 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
2982 without this. I don't know how relevant this is to other LDAP libraries.
2984 PH/02 Add the verb name to the "unknown ACL verb" error.
2986 PH/03 Magnus Holmgren's patch for filter_prepend_home.
2988 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
2990 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
2991 directory not expanded when it should be if an expanded home directory
2992 was set for the address (which is overridden by the transport).
2994 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
2997 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
2998 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
2999 because it is too late at that time, and has no effect.
3001 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
3002 security issue with \' (bugzilla #107). I could not use the
3003 PQescapeStringConn() function, because it needs a PGconn value as one of
3006 PH/08 When testing addresses using -bt, indicate those final addresses that
3007 are duplicates that would not cause an additional delivery. At least one
3008 person was confused, thinking that -bt output corresponded to deliveries.
3009 (Suppressing duplicates isn't a good idea as you lose the information
3010 about possibly different redirections that led to the duplicates.)
3012 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
3013 systems where poll() doesn't work, in particular OS X.
3015 PH/10 Added more information to debugging output for retry time not reached.
3017 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
3018 operations in malware.c.
3020 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
3023 PH/13 If write_rejectlog was set false when logging was sent to syslog with
3024 syslog_duplication set false, log lines that would normally be written
3025 both the the main log and to the reject log were not written to syslog at
3028 PH/14 In the default configuration, change the use of "message" in ACL warn
3029 statements to "add_header".
3031 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
3032 not followed by a command (e.g. "seen endif").
3034 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
3035 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
3038 PH/17 Added extra conditions to the default value of delay_warning_condition
3042 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
3043 { match{$h_precedence:}{(?i)bulk|list|junk} } \
3044 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
3047 The Auto-Submitted: and various List- headers are standardised, whereas I
3048 don't think Precedence: ever was.
3050 PH/18 Refactored debugging code in route_finduser() to show more information,
3051 in particular, the error code if getpwnam() issues one.
3053 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
3054 This is apparently needed in addition to the PH/07 change above to avoid
3055 any possible encoding problems.
3057 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
3058 but not after initializing Perl.
3060 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
3061 output them only if debugging. By default they are written stderr,
3062 apparently, which is not desirable.
3064 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
3067 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
3070 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
3072 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
3073 authenticated or an ident call has been made. Suppress the default
3074 values for $authenticated_id and $authenticated_sender (but permit -oMai
3075 and -oMas) when testing with -bh.
3077 PH/24 Re-jigged the order of the tests in the default configuration so that the
3078 tests for valid domains and recipients precede the DNS black list and CSA
3079 tests, on the grounds that those ones are more expensive.
3081 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
3082 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
3083 This bug exists in every version of Exim that I still have, right back to
3086 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
3087 However, an attempt to turn on case-sensitivity in a regex key by
3088 including (?-i) didn't work because the subject string was already
3089 lowercased, and the effects were non-intuitive. It turns out that a
3090 one-line patch can be used to allow (?-i) to work as expected.
3096 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
3097 other effects) broke the use of negated acl sub-conditions.
3099 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
3102 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
3103 "Deny" causes Exim to reject the incoming connection with a 554 error.
3104 Unfortunately, if there is a major crisis, such as a disk failure,
3105 tcp-wrappers gives "deny", whereas what one would like would be some
3106 kind of temporary error. A kludge has been added to help with this.
3107 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
3108 554 error is used if errno is still zero or contains ENOENT (which occurs
3109 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
3112 PH/03 Add -lutil to the default FreeBSD LIBS setting.
3114 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
3115 errors. Otherwise a message that provokes a temporary error (when other
3116 messages do not) can cause a whole host to time out.
3118 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
3119 addresses were routed directly to files or pipes from a redirect router.
3120 File deliveries just didn't batch; pipe deliveries might have suffered
3123 PH/06 A failure to get a lock for a hints database would erroneously always say
3124 "Failed to get write lock", even when it was really a read lock.
3126 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
3127 of 0600. This has been changed to use the value of the lockfile_mode
3128 option (which defaults to 0600).
3130 PH/08 Applied small patch from the Sieve maintainer.
3132 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
3133 folder from quota calculations, a direct delivery into this folder messed
3134 up the contents of the maildirsize file. This was because the regex was
3135 used only to exclude .Trash (or whatever) when the size of the mailbox
3136 was calculated. There was no check that a delivery was happening into an
3137 excluded directory. This bug has been fixed by ignoring all quota
3138 processing for deliveries into excluded directories.
3140 PH/10 Added the maildirfolder_create_regex option to appendfile.
3146 PH/01 The code for finding all the local interface addresses on a FreeBSD
3147 system running IPv6 was broken. This may well have applied to all BSD
3148 systems, as well as to others that have similar system calls. The broken
3149 code found IPv4 interfaces correctly, but gave incorrect values for the
3150 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
3151 that it would not match correctly against @[] and not recognize the IPv6
3154 PH/02 The ipliteral router was not recognizing addresses of the form user@
3155 [ipv6:....] because it didn't know about the "ipv6:" prefix.
3157 PH/03 Added disable_ipv6.
3159 PH/04 Changed $reply_address to use the raw form of the headers instead of the
3160 decoded form, because it is most often used to construct To: headers
3161 lines in autoreplies, and the decoded form may well be syntactically
3162 invalid. However, $reply_address has leading white space removed, and all
3163 newlines turned into spaces so that the autoreply transport does not
3166 PH/05 If group was specified without a user on a router, and no group or user
3167 was specified on a transport, the group from the router was ignored.
3169 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
3170 for visible compile-time settings that can be used to change these
3171 numbers, for those that want even more. Backwards compatibility with old
3172 spool files has been maintained. However, going back to a previous Exim
3173 release will lost any variables that are in spool files.
3175 PH/07 Two small changes when running in the test harness: increase delay when
3176 passing a TCP/IP connection to a new process, in case the original
3177 process has to generate a bounce, and remove special handling of
3178 127.0.0.2 (sic), which is no longer necessary.
3180 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
3181 be the same on different OS.
3183 PH/09 Moved a debug statement in filter processing to avoid a race problem when
3186 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
3187 whether --show-vars was specified or not
3189 JJ/02 exipick: Added support for new ACL variable spool format introduced
3192 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
3193 syntactically invalid From: or Reply-to: line, and a filter used this to
3194 generate an autoreply, and therefore failed to obtain an address for the
3195 autoreply, Exim could try to deliver to a non-existent relative file
3196 name, causing unrelated and misleading errors. What now happens is that
3197 it logs this as a hard delivery error, but does not attempt to create a
3200 PH/11 The exinext utility has a -C option for testing purposes, but although
3201 the given file was scanned by exinext itself; it wasn't being passed on
3202 when Exim was called.
3204 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
3205 an end-of-file indication when reading a command response.
3207 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
3208 compiled. In many other places in Exim, IPv6 addresses are always
3209 recognized, so I have changed this. It also means that IPv4 domain
3210 literals of the form [IPV4:n.n.n.n] are now always recognized.
3212 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
3213 used if the router is not running as root, for example, when verifying at
3214 ACL time, or when using -bh. The debugging output from this situation was
3215 non-existent - all you got was a failure to exec. I have made two
3218 (a) Failures to set uid/gid, the current directory, or a process leader
3219 in a subprocess such as that created by queryprogram now generate
3220 suitable debugging output when -d is set.
3222 (b) The queryprogram router detects when it is not running as root,
3223 outputs suitable debugging information if -d is set, and then runs
3224 the subprocess without attempting to change uid/gid.
3226 PH/15 Minor change to Makefile for building test_host (undocumented testing
3229 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
3230 additional section of a DNS packet that returns MX or SRV records.
3231 Instead, it always explicitly searches for A/AAAA records. This avoids
3232 major problems that occur when a DNS server includes only records of one
3233 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
3234 fixed another bug: if SRV records were looked up and the corresponding
3235 address records were *not* found in the additional section, the port
3236 values from the SRV records were lost.
3238 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
3239 using the correct key (the original address) when searching the retry
3240 rules in order to find which one to use for generating the retry hint.
3242 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
3243 adding the default one. Similarly, if it contains a Reply-To: header, the
3244 errors_reply_to option, if set, is not used.
3246 PH/19 When calculating a retry time, Exim used to measure the "time since
3247 failure" by looking at the "first failed" field in the retry record. Now
3248 it does not use this if it is later than than the arrival time of the
3249 message. Instead it uses the arrival time. This makes for better
3250 behaviour in cases where some deliveries succeed, thus re-setting the
3251 "first failed" field. An example is a quota failure for a huge message
3252 when small messages continue to be delivered. Without this change, the
3253 "time since failure" will always be short, possible causing more frequent
3254 delivery attempts for the huge message than are intended.
3255 [Note: This change was subsequently modified - see PH/04 for 4.62.]
3257 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
3258 $1, $2, $3) because the numerical variables can be reset during some
3259 expansion items (e.g. "match"), thereby losing the authentication data.
3261 PH/21 Make -bV show the size of off_t variables so that the test suite can
3262 decide whether to run tests for quotas > 2G.
3264 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
3265 mailbox_size, and mailbox_filecount in the appendfile transport. If a
3266 filecount value is greater than 2G or if a quota value is greater than 2G
3267 on a system where the size of off_t is not greater than 4, a panic error
3270 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
3271 never match. The debug and -bh output now contains an explicit error
3272 message indicating a malformed IPv4 address or mask.
3274 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
3275 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
3276 PH/23 above applies.
3278 PH/25 Do not write to syslog when running in the test harness. The only
3279 occasion when this arises is a failure to open the main or panic logs
3280 (for which there is an explicit test).
3282 PH/26 Added the /no_tell option to "control=freeze".
3284 PH/27 If a host name lookup failed very early in a connection, for example, if
3285 the IP address matched host_lookup and the reverse lookup yielded a name
3286 that did not have a forward lookup, an error message of the form "no IP
3287 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
3288 could be logged. Now it outputs the IP address instead of "NULL".
3290 PH/28 An enabling patch from MH: add new function child_open_exim2() which
3291 allows the sender and the authenticated sender to be set when
3292 submitting a message from within Exim. Since child_open_exim() is
3293 documented for local_scan(), the new function should be too.
3295 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
3296 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
3297 results in an empty string is now treated as unset.
3299 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
3301 PH/31 Added sender_verify_fail logging option.
3303 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
3304 needed by RFC 822 but not by RFC 2822 was commented out. I have now
3305 tidied the source and removed it altogether.
3307 PH/33 When a queue run was abandoned because the load average was too high, a
3308 log line was always written; now it is written only if the queue_run log
3309 selector is set. In addition, the log line for abandonment now contains
3310 information about the queue run such as the pid. This is always present
3311 in "start" and "stop" lines but was omitted from the "abandon" line.
3313 PH/34 Omit spaces between a header name and the colon in the error message that
3314 is given when verify = headers_syntax fails (if there are lots of them,
3315 the message gets confusing).
3317 PH/35 Change the default for dns_check_names_pattern to allow slashes within
3318 names, as there are now some PTR records that contain slashes. This check
3319 is only to protect against broken name servers that fall over on strange
3320 characters, so the fact that it applies to all lookups doesn't matter.
3322 PH/36 Now that the new test suite is complete, we can remove some of the
3323 special code in Exim that was needed for the old test suite. For example,
3324 sorting DNS records because real resolvers return them in an arbitrary
3325 order. The new test suite's fake resolver always returns records in the
3328 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
3329 bounces) except when queue_only is set, to avoid logging races between
3330 the different processes.
3332 PH/38 Panic-die if .include specifies a non-absolute path.
3334 PH/39 A tweak to the "H" retry rule from its user.
3336 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
3337 a label. They prevented compilation on older perls.
3339 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
3340 a warning to be raised on newish perls.
3342 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
3343 on queue. Changes to match documented behaviour of showing count of
3344 messages matching specified criteria.
3346 PH/40 Changed the default ident timeout from 30s to 5s.
3348 PH/41 Added support for the use of login_cap features, on those BSD systems
3349 that have them, for controlling the resources used by pipe deliveries.
3351 PH/42 The content-scanning code uses fopen() to create files in which to put
3352 message data. Previously it was not paying any attention to the mode of
3353 the files. Exim runs with umask(0) because the rest of the code creates
3354 files with open(), and sets the required mode explicitly. Thus, these
3355 files were ending up world-writeable. This was not a big issue, because,
3356 being within the spool directory, they were not world-accessible. I have
3357 created a function called modefopen, which takes an additional mode
3358 argument. It sets umask(777), creates the file, chmods it to the required
3359 mode, then resets the umask. All the relevant calls to fopen() in the
3360 content scanning code have been changed to use this function.
3362 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
3363 to 24 hours. This avoids potential overflow problems when processing G
3364 and H retry rules. I suspect nobody ever tinkers with this value.
3366 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
3368 PH/45 When the plaintext authenticator is running as a client, the server's
3369 challenges are checked to ensure they are valid base64 strings. By
3370 default, the authentication attempt is cancelled if an invalid string is
3371 received. Setting client_ignore_invalid_base64 true ignores these errors.
3372 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
3373 they are received. Thus, the responses can be made to depend on the
3374 challenges. If an invalid string is ignored, an empty string is placed in
3377 PH/46 Messages that are created by the autoreply transport now contains a
3378 References: header, in accordance with RFCs 2822 and 3834.
3380 PH/47 Added authenticated_sender_force to the smtp transport.
3382 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
3384 PH/49 Installed latest patch from the Sieve maintainer.
3386 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
3387 was also set, the appendfile transport was unnecessarily scanning a
3388 directory of message files (e.g. for maildir delivery) to find the count
3389 of files (along with the size), even though it did not need this
3390 information. It now does the scan only if it needs to find either the
3391 size of the count of files.
3393 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
3395 PH/52 Two bugs concerned with error handling when the smtp transport is
3398 (i) Exim was not creating retry information for temporary errors given
3399 for individual recipients after the DATA command when the smtp transport
3400 was used in LMTP mode. This meant that they could be retried too
3401 frequently, and not timed out correctly.
3403 (ii) Exim was setting the flag that allows error details to be returned
3404 for LMTP errors on RCPT commands, but not for LMTP errors for individual
3405 recipients that were returned after the DATA command.
3407 PH/53 This is related to PH/52, but is more general: for any failing address,
3408 when detailed error information was permitted to be returned to the
3409 sender, but the error was temporary, then after the final timeout, only
3410 "retry timeout exceeded" was returned. Now it returns the full error as
3411 well as "retry timeout exceeded".
3413 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
3414 do this, and (what is worse) MTAs that accept it.
3416 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
3417 will now be deprecated.
3419 PH/56 New os.c-cygwin from the Cygwin maintainer.
3421 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
3422 formats (previously only available in exim formats via -bpr, -bpru,
3423 and -bpra. Now also available in native and exiqgrep formats)
3425 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
3426 with very large, slow to parse queues
3428 JJ/08 exipick: added ! as generic prefix to negate any criteria format
3430 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
3432 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
3433 responses to authentication challenges, though it was showing the
3434 challenges; (ii) I've removed the CR characters from the debug output for
3437 PH/58 Allow for the insertion of a newline as well as a space when a string
3438 is turned into more than one encoded-word during RFC 2047 encoding. The
3439 Sieve code now uses this.
3441 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
3442 data_4xx, lost_connection, tls_required.
3444 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
3445 message was being sent as an SMTP response.
3447 PH/61 Add -l and -k options to exicyclog.
3449 PH/62 When verifying, if an address was redirected to one new address, so that
3450 verification continued, and the new address failed or deferred after
3451 having set something in $address_data, the value of $address_data was not
3452 passed back to the ACL. This was different to the case when no
3453 redirection occurred. The value is now passed back in both cases.
3455 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
3456 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
3457 use login_cap.h, so on its own it isn't the distinguishing feature. The
3458 new name refers directly to the setclassresources() function.
3460 PH/65 Added configuration files for NetBSD3.
3462 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
3464 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
3465 is preferred over IPv4.
3467 PH/68 The bounce_return_message and bounce_return_body options were not being
3468 honoured for bounces generated during the reception of non-SMTP messages.
3469 In particular, this applied to messages rejected by the ACL. This bug has
3470 been fixed. However, if bounce_return_message is true and bounce_return_
3471 body is false, the headers that are returned for a non-SMTP message
3472 include only those that have been read before the error was detected.
3473 (In the case of an ACL rejection, they have all been read.)
3475 PH/69 The HTML version of the specification is now built in a directory called
3476 spec_html instead of spec.html, because the latter looks like a path with
3477 a MIME-type, and this confuses some software.
3479 PH/70 Catch two compiler warnings in sieve.c.
3481 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
3482 function verify_get_ident() calls ip_connect() to connect a socket, but
3483 if the "connect()" function timed out, ip_connect() used to close the
3484 socket. However, verify_get_ident() also closes the socket later, and in
3485 between Exim writes to the log, which may get opened at this point. When
3486 the socket was closed in ip_connect(), the log could get the same file
3487 descriptor number as the socket. This naturally causes chaos. The fix is
3488 not to close the socket in ip_connect(); the socket should be closed by
3489 the function that creates it. There was only one place in the code where
3490 this was missing, in the iplookup router, which I don't think anybody now
3491 uses, but I've fixed it anyway.
3493 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
3494 well as to direct DNS lookups. Otherwise the handling of names in host
3495 lists is inconsistent and therefore confusing.
3501 PH/01 Two changes to the default runtime configuration:
3503 (1) Move the checks for relay_from_hosts and authenticated clients from
3504 after to before the (commented out) DNS black list checks.
3506 (2) Add control=submission to the relay_from_hosts and authenticated
3507 clients checks, on the grounds that messages accepted by these
3508 statements are most likely to be submissions.
3510 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
3512 (1) Generate an error if the third argument for the ${prvs expansion is
3515 (2) Treat a missing third argument of ${prvscheck as if it were an empty
3518 (3) Reset the variables that are obtained from the first argument of
3519 ${prvscheck and used in the second argument before leaving the code,
3520 because their memory is reclaimed, so using them afterwards may do
3523 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
3524 one (it's much easier than Tom thought :-).
3526 (5) Because of (4), we can now allow for the use of $prvscheck_result
3527 inside the third argument.
3529 PH/03 For some reason, the default setting of PATH when running a command from
3530 a pipe transport was just "/usr/bin". I have changed it to
3533 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
3534 anything to be listed in the output from -bV.
3536 PH/05 When a filter generated an autoreply, the entire To: header line was
3537 quoted in the delivery log line, like this:
3539 => >A.N.Other <ano@some.domain> <original@ddress> ...
3541 This has been changed so that it extracts the operative address. There
3542 may be more than one such address. If so, they are comma-separated, like
3545 => >ano@some.domain,ona@other.domain <original@ddress> ...
3547 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
3548 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
3549 not being looked up in the rDNS to get a host name, Exim was showing the
3550 IP address twice in Received: lines, even though the IP addresses were
3551 identical. For example:
3553 Received: from [1.2.3.4] (helo=[1.2.3.4])
3555 However, if the real host name was known, it was omitting the HELO data
3556 if it matched the actual IP address. This has been tidied up so that it
3557 doesn't show the same IP address twice.
3559 PH/07 When both +timestamp and +memory debugging was on, the value given by
3560 $tod_xxx expansions could be wrong, because the tod_stamp() function was
3561 called by the debug printing, thereby overwriting the timestamp buffer.
3562 Debugging no longer uses the tod_stamp() function when +timestamp is set.
3564 PH/08 When the original message was included in an autoreply transport, it
3565 always said "this is a copy of the message, including all the headers",
3566 even if body_only or headers_only was set. It now gives an appropriate
3569 PH/09 Applied a patch from the Sieve maintainer which:
3571 o fixes some comments
3572 o adds the (disabled) notify extension core
3573 o adds some debug output for the result of if/elsif tests
3574 o points to the current vacation draft in the documentation
3575 and documents the missing references header update
3579 o fixes a bug in processing the envelope test (when testing
3580 multiple envelope elements, the last element determined the
3583 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
3584 Electronic Mail") by including:
3586 Auto-submitted: auto-generated
3588 in the messages that it generates (bounce messages and others, such as
3589 warnings). In the case of bounce messages for non-SMTP messages, there was
3590 also a typo: it was using "Auto_submitted" (underscore instead of
3591 hyphen). Since every message generated by Exim is necessarily in response
3592 to another message, thes have all been changed to:
3594 Auto-Submitted: auto-replied
3596 in accordance with these statements in the RFC:
3598 The auto-replied keyword:
3600 - SHOULD be used on messages sent in direct response to another
3601 message by an automatic process,
3603 - MUST NOT be used on manually-generated messages,
3605 - MAY be used on Delivery Status Notifications (DSNs) and Message
3606 Disposition Notifications (MDNs),
3608 - MUST NOT be used on messages generated by automatic or periodic
3609 processes, except for messages which are automatic responses to
3612 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
3613 to the default Received: header definition.
3615 PH/12 Added log selector acl_warn_skipped (default on).
3617 PH/13 After a successful wildlsearch lookup, discard the values of numeric
3618 variables because (a) they are in the wrong storage pool and (b) even if
3619 they were copied, it wouldn't work properly because of the caching.
3621 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
3622 checking when decoding. Apparently there are clients that generate
3623 overlong encoded strings. Why am I not surprised?
3625 PH/15 If the first argument of "${if match_address" was not empty, but did not
3626 contain an "@" character, Exim crashed. Now it writes a panic log message
3627 and treats the condition as false.
3629 PH/16 In autoreply, treat an empty string for "once" the same as unset.
3631 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
3632 extension "envelope-auth". The code is finished and in agreement with
3633 other implementations, but there is no documentation so far and in fact,
3634 nobody wrote the draft yet. This extension is currently #undef'ed, thus
3635 not changing the active code.
3637 Print executed "if" and "elsif" statements when debugging is used. This
3638 helps a great deal to understand what a filter does.
3640 Document more things not specified clearly in RFC3028. I had all this
3641 sorted out, when out of a sudden new issues came to my mind. Oops."
3643 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
3646 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
3647 canonical form (as documented). However, after a host name lookup from
3648 the IP address, check_host() was doing a simple string comparison with
3649 addresses acquired from the DNS when checking that the found name did
3650 have the original IP as one of its addresses. Since any found IPv6
3651 addresses are likely to be in abbreviated form, the comparison could
3652 fail. Luckily, there already exists a function for doing the comparison
3653 by converting both addresses to binary, so now that is used instead of
3654 the text comparison.
3656 PH/20 There was another similar case to PH/19, when a complete host name was
3657 given in a host list; looking up its IP address could give an abbreviated
3658 form, whereas the current host's name might or might not be abbreviated.
3659 The same fix has been applied.
3665 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
3666 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
3669 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
3670 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
3672 PH/03 Typo: missing ".o" in src/pcre/Makefile.
3674 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
3675 header line, restrict the check to what is listed in RFCs 2369 and 2929.
3676 Also, for "Auto-Submitted", treat anything other than "no" as
3677 non-personal, in accordance with RFC 3834. (Previously it treated
3678 anything starting "auto-" as non-personal.)
3680 TF/01 The control=submission/name=... option had a problem with syntax
3681 errors if the name included a slash character. The /name= option
3682 now slurps the rest of the string, so it can include any characters
3683 but it must come last in the list of options (after /sender_retain
3686 PH/05 Some modifications to the interface to the fake nameserver for the new
3694 TK/01 Added the "success_on_redirect" address verification option. See
3695 NewStuff for rationale and an example.
3697 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
3699 PH/02 Patch to exigrep to allow it to work on syslog lines.
3701 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
3702 fread() to skip over the body file's header line, because in Cygwin the
3703 header line is locked and is inaccessible.
3705 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
3706 co-exist for some time) to make it clear that it is the Exim ID that is
3707 referenced, not the Message-ID: header line.
3709 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
3710 string_format() function, because snprintf() does not exist on all
3713 PH/06 The use of forbid_filter_existstest now also locks out the use of the
3714 ${stat: expansion item.
3716 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
3717 protocol synchronization error", to keep the pedants happy.
3719 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
3720 well as for IRIX systems, when gcc is being used. See the host.c source
3723 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
3725 PH/10 Named domain lists were not working if used in a queue_smtp_domains
3728 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
3729 transport and to the smtp transport in LMTP mode.
3731 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
3733 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
3734 run a filter in a subprocess. This could lead to confusion in subsequent
3735 lookups in the parent process. There should also be a search_tidyup() at
3736 the end of the subprocess.
3738 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
3739 only if the host matched helo_try_verify_hosts, which caused the
3740 verification to occur when the EHLO/HELO command was issued. The ACL just
3741 tested the remembered result. Now, if a previous verification attempt has
3742 not happened, "verify = helo" does it there and then.
3744 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
3746 TK/03 Fix log output including CR from clamd.
3748 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
3749 exist provoked a memory error which could cause a segfault.
3751 PH/15 Installed PCRE 6.2
3753 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
3755 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
3756 of the problem. Specifically, suggested +O2 rather than +O1 for the
3759 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
3761 PH/20 If a delivery was routed to a non-standard port by means of an SRV
3762 record, the port was not correctly logged when the outgoing_port log
3763 selector was set (it logged the transort's default port).
3765 PH/21 Added support for host-specific ports to manualroute, queryprogram,
3766 fallback_hosts, and "hosts" in the smtp transport.
3768 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
3769 host errors such as "Connection refused".
3771 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
3772 authentication with radiusclient 0.4.9:
3774 - Error returned from rc_read_config was caught wrongly
3775 - Username/password not passed on to radius server due to wrong length.
3777 The presumption is that some radiusclient API changes for 4.51/PH/17
3778 were not taken care of correctly. The code is still untested by me (my
3779 Linux distribution still has 0.3.2 of radiusclient), but it was
3780 contributed by a Radius user.
3782 PH/24 When doing a callout, the value of $domain wasn't set correctly when
3783 expanding the "port" option of the smtp transport.
3785 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
3786 while reading a MIME header. Thanks to Tom Hughes for a patch.
3788 PH/24 Include config.h inside local_scan.h so that configuration settings are
3791 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
3792 that in an ACL for RCPT (for example), you can examine exactly what was
3795 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
3796 commands, but it was not correctly comparing the address with the actual
3797 client host address. Thus, it would show the EHLO address in Received:
3798 header lines when this was not necessary.
3800 PH/27 Added the % operator to ${eval:}.
3802 PH/28 Exim tries to create and chdir to its spool directory when it starts;
3803 it should be ignoring failures (because with -C, for example, it has lost
3804 privilege). It wasn't ignoring creation failures other than "already
3807 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
3810 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
3811 because an input line was too long, either on its own, or by virtue of
3812 too many continuations, the temporary file was not being removed, and the
3813 return code was incorrect.
3815 PH/31 Missing "BOOL" in function definition in filtertest.c.
3817 PH/32 Applied Sieve patches from the maintainer.
3819 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
3821 PH/33 Added "verify = not_blind".
3823 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
3824 Local/Makefile (with some defaults set). These are used in built scripts
3825 such as exicyclog, but they have never been used in the exim_install
3826 script (though there are many overriding facilities there). I have
3827 arranged that the exim_install script now takes note of these two
3830 PH/35 Installed configuration files for Dragonfly.
3832 PH/36 When a locally submitted message by a trusted user did not contain a
3833 From: header, and the sender address was obtained from -f or from an SMTP
3834 MAIL command, and the trusted user did not use -F to supply a sender
3835 name, $originator_name was incorrectly used when constructing a From:
3836 header. Furthermore, $originator_name was used for submission mode
3837 messages from external hosts without From: headers in a similar way,
3838 which is clearly wrong.
3840 PH/37 Added control=suppress_local_fixups.
3842 PH/38 When log_selector = +received_sender was set, and the addition of the
3843 sender made the log line's construction buffer exactly full, or one byte
3844 less than full, an overflow happened when the terminating "\n" was
3847 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
3848 when the result of a list match is failure because a DNS lookup failed.
3850 PH/40 RM_COMMAND is now used in the building process.
3852 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
3853 the "build-* directories that it finds.
3855 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
3856 address in a domain literal was a prefix of an interface address.
3858 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
3859 when verifying a sender address, unless rewrite_headers is false.
3861 PH/44 Wrote a long comment about why errors_to addresses are verified as
3862 recipients, not senders.
3864 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
3865 the ratelimit ACL was added.
3867 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
3869 PH/46 Added extra information about PostgreSQL errors to the error string.
3871 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
3872 avoiding the need to install special zones in a real server. This is
3873 backwards compatible; if it can't find the fake resolver, it drops back.
3874 Thus, both old and new test suites can be run.
3876 TF/02 Added util/ratelimit.pl
3878 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
3879 clock is set back in time.
3881 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
3882 Candler <B.Candler@pobox.com>.
3884 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
3885 OK for addresses that are the result of redirections.
3887 PH/48 A number of further additions for the benefit of the new test suite,
3888 including a fake gethostbyname() that interfaces to the fake DNS resolver
3891 TF/06 The fix for widen_domains has also been applied to qualify_single and
3892 search_parents which are the other dnslookup options that can cause
3895 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
3898 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
3900 TF/07 Exim produced the error message "an SRV record indicated no SMTP
3901 service" if it encountered an MX record with an empty target hostname.
3902 The message is now "an MX or SRV record indicated no SMTP service".
3904 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
3905 if the DNS of the sending site is misconfigured. This is quite a
3906 common situation. This change restores the behaviour of treating a
3907 helo verification defer as a failure.
3909 PH/51 If self=fail was set on a router, the bounce message did not include the
3910 actual error message.
3916 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
3918 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
3919 command itself ended in error, the underlying message about the transport
3920 filter timeout was being overwritten with the pipe command error. Now the
3921 underlying error message should be appended to the second error message.
3923 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
3925 PH/02 Reduce the amount of output that "make" produces by default. Full output
3926 can still be requested.
3928 PH/03 The warning log line about a condition test deferring for a "warn" verb
3929 was being output only once per connection, rather than after each
3930 occurrence (because it was using the same function as for successful
3931 "warn" verbs). This seems wrong, so I have changed it.
3933 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
3934 it should not have, which might have caused a crash in the right
3935 circumstances, but probably never did.
3937 PH/04 Installed a modified version of Tony Finch's patch to make submission
3938 mode fix the return path as well as the Sender: header line, and to
3939 add a /name= option so that you can make the user's friendly name appear
3942 TF/03 Added the control = fakedefer ACL modifier.
3944 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
3945 Mark Lowes for thorough testing.
3947 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
3949 TK/03 Merged latest SRS patch from Miles Wilton.
3951 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
3952 with the definition in sysexits.h (which is #included earlier).
3953 Fortunately, Exim does not actually use EX_OK. The code used to try to
3954 preserve the sysexits.h value, by assuming that macro definitions were
3955 scanned for macro replacements. I have been disabused of this notion,
3956 so now the code just undefines EX_OK before #including unistd.h.
3958 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
3959 in the smtp transport. When a block could not be written in a single
3960 write() function, the timeout was being re-applied to each part-write.
3961 This seems wrong - if the receiver was accepting one byte at a time it
3962 would take for ever. The timeout is now adjusted when this happens. It
3963 doesn't have to be particularly precise.
3965 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
3966 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
3968 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
3969 without a domain if the check to <postmaster@domain> fails.
3971 SC/01 Eximstats: added -xls and the ability to specify output files
3972 (patch written by Frank Heydlauf).
3974 SC/02 Eximstats: use FileHandles for outputting results.
3976 SC/03 Eximstats: allow any combination of xls, txt, and html output.
3978 SC/04 Eximstats: fixed display of large numbers with -nvr option
3980 SC/05 Eximstats: fixed merging of reports with empty tables.
3982 SC/06 Eximstats: added the -include_original_destination flag
3984 SC/07 Eximstats: removed tabs and trailing whitespace.
3986 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
3988 TK/06 MBOX spool code: Add real "From " MBOX separator line
3989 so the .eml file is really in mbox format (even though
3990 most programs do not really care). Patch from Alex Miller.
3992 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
3993 The latter is generated from $received_to and is only set if the
3994 message has one envelope recipient. SA can use these headers,
3995 obviously out-of-the-box. Patch from Alex Miller.
3997 PH/08 The ${def test on a variable was returning false if the variable's
3998 value was "0", contrary to what the specification has always said!
3999 The result should be true unless the variable is empty.
4001 PH/09 The syntax error of a character other than { following "${if
4002 def:variable_name" (after optional whitespace) was not being diagnosed.
4003 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
4004 accidental colon was present, for example, could give incorrect results.
4006 PH/10 Tidied the code in a number of places where the st_size field of a stat()
4007 result is used (not including appendfile, where other changes are about
4010 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
4011 This involved changing a lot of size variables from int to off_t. It
4012 should work with maildirs and everything.
4014 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
4015 spamd dying while we are connected to it.
4017 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
4020 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
4021 function for BDB 4.3.
4023 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
4025 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
4026 the sources for PCRE have been re-arranged and more files are now
4029 PH/15 The code I had for printing potentially long long variables in PH/11
4030 above was not the best (it lost precision). The length of off_t variables
4031 is now inspected at build time, and an appropriate printing format (%ld
4032 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
4033 to be "long long int" or "long int". This is needed for the internal
4034 formatting function string_vformat().
4036 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
4037 the configuration file to be ":syslog", then the script "guesses" where
4038 the logs files are, rather than using the compiled in default. In our
4039 case the guess is not the same as the compiled default, so the script
4040 suddenly stopped working when I started to use syslog. The patch checks
4041 to see if log_file_path is "". If so, it attempts to read it from exim
4042 with no configuration file to get the compiled in version, before it
4043 falls back to the previous guessing code."
4045 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
4046 implementing BATV in an Exim configuration. See NewStuff for the gory
4049 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
4050 Makefile that are specific to HP-UX.
4052 PH/18 If the "use_postmaster" option was set for a recipient callout together
4053 with the "random" option, the postmaster address was used as the MAIL
4054 FROM address for the random test, but not for the subsequent recipient
4055 test. It is now used for both.
4057 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
4058 patch removes a few documentation additions to RFC 3028, because the
4059 latest draft now contains them. It adds the new en;ascii-case comparator
4060 and a new error check for 8bit text in MIME parts. Comparator and
4061 require names are now matched exactly. I enabled the subaddress
4062 extension, but it is not well tested yet (read: it works for me)."
4064 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
4065 rework some of the code of TK/09 above to avoid the hardwired use of
4066 "%lld" and "long long". Replaced the call to snprintf() with a call to
4069 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
4070 records point to non-existent hosts", "retry timeout exceeded", and
4071 "retry time not reached for any host after a long failure period".
4073 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
4074 experimental DomainKeys support:
4076 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
4077 (2) On an error such as an illegally used "control", the wrong name for
4078 the control was given.
4080 These problems did NOT occur unless DomainKeys support was compiled.
4082 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
4084 PH/24 Added ${if match_ip condition.
4086 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
4087 databases so that it will be absolutely obvious if a crash occurs in the
4088 DB library. This is a regular occurrence (often caused by mis-matched
4091 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
4092 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
4093 picked up on a user's system that detects such things. There doesn't seem
4094 to be a gcc warning option for this - only an attribute that has to be
4095 put on the function's prototype. It seems that in Fedora Core 4 they have
4096 set this on a number of new functions. No doubt there will be more in due
4099 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
4100 specify a transport. However, if an address that was verified by such a
4101 router was the subject of a callout, Exim crashed because it tried to
4102 read the rcpt_include_affixes from the non-existent transport. Now it
4103 just assumes that the setting of that option is false. This bug was
4104 introduced by 4.51/PH/31.
4106 PH/28 Changed -d+all to exclude +memory, because that information is very
4107 rarely of interest, but it makes the output a lot bigger. People tend to
4108 do -d+all out of habit.
4110 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
4111 code in os-type was giving problems when libc.so lives in lib64, like on
4114 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
4115 aren't the modern standard, and it seems that some systems' include files
4116 don't always have them. Exim was already checking for some of the newer
4117 ones like T_AAAA, and defining it itself. I've added checks for all the
4118 record types that Exim uses.
4120 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
4121 not automatically generating a new one, as it is supposed to. This
4122 prevented TLS from working. If the file did exist, but contained invalid
4123 data, a new version was generated, as expected. It was only the case of a
4124 non-existent file that was broken.
4126 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
4127 with a change in libdomainkeys > 0.64.
4129 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
4130 from DNS. If the selector record carries the flag, it now has
4131 precedence over the domain-wide flag.
4133 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
4135 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
4136 the use of a port name that isn't defined in /etc/services) occurred, the
4137 message was deferred as in a normal delivery, and thus remained on the
4138 spool, instead of being failed because of the mua_wrapper setting. This
4139 is now fixed, and I tidied up some of the mua_wrapper messages at the
4142 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
4143 the messages in a hash of arrays rather than using individual hashes.
4144 This is a bit cleaner and results in dramatic memory savings, albeit
4145 at a slight CPU cost.
4147 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
4148 as requested by Marc Sherman.
4150 SC/10 Eximstats: added histograms for user specified patterns as requested
4153 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
4155 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
4156 fopen() in the content-scanning modules that did not already have it.
4162 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
4163 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
4165 TK/02 Fix ACL "control" statement not being available in MIME ACL.
4167 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
4169 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
4170 to test Sieve filters that use "vacation".
4172 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
4173 that changes the way the GnuTLS parameters are stored in the cache file.
4174 The new format can be generated externally. For backward compatibility,
4175 if the data in the cache doesn't make sense, Exim assumes it has read an
4176 old-format file, and it generates new data and writes a new file. This
4177 means that you can't go back to an older release without removing the
4180 PH/03 A redirect router that has both "unseen" and "one_time" set does not
4181 work if there are any delivery delays because "one_time" forces the
4182 parent to be marked "delivered", so its unseen clone is never tried
4183 again. For this reason, Exim now forbids the simultaneous setting of
4186 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
4187 redirected to themselves ("homonym" addresses). Read the long ChangeLog
4188 entry if you want to know the details. The fix, however, neglected to
4189 consider the case when local delivery batching is involved. The test for
4190 "previously delivered" was not happening when checking to see if an
4191 address could be batched with a previous (undelivered) one; under
4192 certain circumstances this could lead to multiple deliveries to the same
4195 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
4196 in its include files, and this causes problems building Exim.
4198 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
4199 header_syntax) but Exim was just ignoring anything given after a slash.
4200 In particular, this caused confusion with an attempt to use "verify =
4201 reverse_host_lookup/defer_ok". An error is now given when options are
4202 supplied for verify items that do not have them. (Maybe reverse_host_
4203 lookup should have a defer_ok option, but that's a different point.)
4205 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
4206 defined by RFC 821) to 2048, because there were problems with some AUTH
4207 commands, and RFC 1869 says the size should be increased for extended
4208 SMTP commands that take arguments.
4210 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
4213 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
4214 "unknown" error; now it says that the functionality isn't in the binary.
4216 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
4217 an address' error message when a string expansion fails (syntax or
4218 whatever). Otherwise the password may appear in the log. Following change
4219 PH/42 below, there is no longer a chance of it appearing in a bounce
4222 PH/11 Installed exipick version 20050225.0 from John Jetmore.
4224 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
4225 of its addresses was ever tried. (Bugzilla bug #2.)
4227 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
4228 the result incorrectly in the debug output. (It correctly added a newline
4229 to what was transported.)
4231 TF/01 Added $received_time.
4233 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
4234 commented out examples of how to interface to a virus scanner and to
4235 SpamAssassin. Also added commented examples of av_scanner and
4236 spamd_address settings.
4238 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
4239 and controls are allowed in which ACLs. There were a couple of minor
4240 errors. Some of the entries in the conditions table (which is a table of
4241 where they are NOT allowed) were getting very unwieldy; rewrote them as a
4242 negation of where the condition IS allowed.
4244 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
4246 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
4247 header file does not have a version number, so I've had to invent a new
4248 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
4249 API. The code is untested by me (my Linux distribution still has 0.3.2 of
4250 radiusclient), but it was contributed by a Radius user.
4252 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
4253 files or directories, for OpenSSL.
4255 PH/19 When an Exim process that is running as root has to create an Exim log
4256 file, it does so in a subprocess that runs as exim:exim so as to get the
4257 ownership right at creation (otherwise, other Exim processes might see
4258 the file with the wrong ownership). There was no test for failure of this
4259 fork() call, which would lead to the process getting stuck as it waited
4260 for a non-existent subprocess. Forks do occasionally fail when resources
4261 run out. I reviewed all the other calls to fork(); they all seem to check
4264 PH/20 When checking for unexpected SMTP input at connect time (before writing
4265 the banner), Exim was not dealing correctly with a non-positive return
4266 from the read() function. If the client had disconnected by this time,
4267 the result was a log entry for a synchronization error with an empty
4268 string after "input=" when read() returned zero. If read() returned -1
4269 (an event I could not check), uninitialized data bytes were printed.
4270 There were reports of junk text (parts of files, etc) appearing after
4273 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
4275 PH/22 Added support for macro redefinition, and (re)definition in between
4276 driver and ACL definitions.
4278 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
4279 forgetting to use the resulting value; it was using the unexpanded value.
4281 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
4282 hadn't been configured. The fix is from Juergen Kreileder, who
4283 understands it better than I do:
4285 "Here's what I see happening with three configured cyrus_sasl
4286 authenticators configured (plain, login, cram-md5):
4288 On startup auth_cyrus_sasl_init() gets called for each of these.
4289 This means three calls to sasl_listmech() without a specified mech_list.
4290 => SASL tests which mechs of all available mechs actually work
4291 => three warnings about OTP not working
4292 => the returned list contains: plain, login, cram-md5, digest-md5, ...
4294 With the patch, sasl_listmech() also gets called three times. But now
4295 SASL's mech_list option is set to the server_mech specified in the the
4296 authenticator. Or in other words, the answer from sasl_listmech()
4297 gets limited to just the mech you're testing for (which is different
4299 => the return list contains just 'plain' or 'login', 'cram-md5' or
4300 nothing depending on the value of ob->server_mech.
4302 I've just tested the patch: Authentication still works fine,
4303 unavailable mechs specified in the exim configuration are still
4304 caught, and the auth.log warnings about OTP are gone."
4306 PH/25 When debugging is enabled, the contents of the command line are added
4307 to the debugging output, even when log_selector=+arguments is not
4310 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
4311 answer is "GNU", and only if the return is "GNU/something" is the answer
4314 PH/27 $acl_verify_message is now set immediately after the failure of a
4315 verification in an ACL, and so is available in subsequent modifiers. In
4316 particular, the message can be preserved by coding like this:
4318 warn !verify = sender
4319 set acl_m0 = $acl_verify_message
4321 Previously, $acl_verify_message was set only while expanding "message"
4322 and "log_message" when a very denied access.
4324 PH/28 Modified OS/os.c-Linux with
4326 -#ifndef OS_LOAD_AVERAGE
4327 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
4329 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
4330 nomenclature these days.)
4332 PH/29 Installed patch from the Sieve maintainer that adds the options
4333 sieve_useraddress and sieve_subaddress to the redirect router.
4335 PH/30 In these circumstances:
4336 . Two addresses routed to the same list of hosts;
4337 . First host does not offer TLS;
4338 . First host accepts first address;
4339 . First host gives temporary error to second address;
4340 . Second host offers TLS and a TLS session is established;
4341 . Second host accepts second address.
4342 Exim incorrectly logged both deliveries with the TLS parameters (cipher
4343 and peerdn, if requested) that were in fact used only for the second
4346 PH/31 When doing a callout as part of verifying an address, Exim was not paying
4347 attention to any local part prefix or suffix that was matched by the
4348 router that accepted the address. It now behaves in the same way as it
4349 does for delivery: the affixes are removed from the local part unless
4350 rcpt_include_affixes is set on the transport.
4352 PH/32 Add the sender address, as F=<...>, to the log line when logging a
4353 timeout during the DATA phase of an incoming message.
4355 PH/33 Sieve envelope tests were broken for match types other than :is. I have
4356 applied a patch sanctioned by the Sieve maintainer.
4358 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
4359 the uid or gid is negative. A case of a negative gid caused this to be
4360 noticed. The fix allows for either to be negative.
4362 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
4363 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
4366 PH/36 The change PH/12 above was broken. Fixed it.
4368 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
4369 the grounds that routing the same address twice would always produce the
4370 same answer. This might have been true once, but it is certainly no
4371 longer true now. Routing a child address may depend on the previous
4372 routing that produced that child. Some complicated redirection strategies
4373 went wrong when messages had multiple recipients, and made Exim's
4374 behaviour dependent on the order in which the addresses were given.
4376 I have moved the duplicate checking until after the routing is complete.
4377 Exim scans the addresses that are assigned to local and remote
4378 transports, and removes any duplicates. This means that more work will be
4379 done, as duplicates will always all be routed, but duplicates are
4380 presumably rare, so I don't expect this is of any significance.
4382 For deliveries to pipes, files, and autoreplies, the duplicate checking
4383 still happens during the routing process, since they are not going to be
4386 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
4387 It corrects a timeout issue with spamd. This is Ian's comment: "The
4388 background is that sometimes spamd either never reads data from a
4389 connection it has accepted, or it never writes response data. The exiscan
4390 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
4391 blindly assumes that writes won't block so it may never time out."
4393 PH/39 Allow G after quota size as well as K and M.
4395 PH/40 The value set for $authenticated_id in an authenticator may not contain
4396 binary zeroes or newlines because the value is written to log lines and
4397 to spool files. There was no check on this. Now the value is run through
4398 the string_printing() function so that such characters are converted to
4399 printable escape sequences.
4401 PH/41 $message_linecount is a new variable that contains the total number of
4402 lines in the message. Compare $body_linecount, which is the count for the
4405 PH/42 Exim no longer gives details of delivery errors for specific addresses in
4406 bounce and delay warning messages, except in certain special cases, which
4409 (a) An SMTP error message from a remote host;
4410 (b) A message specified in a :fail: redirection;
4411 (c) A message specified in a "fail" command in a system filter;
4412 (d) A message specified in a FAIL return from the queryprogram router;
4413 (e) A message specified by the cannot_route_message router option.
4415 In these cases only, Exim does include the error details in bounce and
4416 warning messages. There are also a few cases where bland messages such
4417 as "unrouteable address" or "local delivery error" are given.
4419 PH/43 $value is now also set for the "else" part of a ${run expansion.
4421 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
4422 being worked on, but at least Exim now implements the latest version to
4425 PH/45 In a pipe transport, although a timeout while waiting for the pipe
4426 process to complete was treated as a delivery failure, a timeout while
4427 writing the message to the pipe was logged, but erroneously treated as a
4428 successful delivery. Such timeouts include transport filter timeouts. For
4429 consistency with the overall process timeout, these timeouts are now
4430 treated as errors, giving rise to delivery failures by default. However,
4431 there is now a new Boolean option for the pipe transport called
4432 timeout_defer, which, if set TRUE, converts the failures into defers for
4433 both kinds of timeout. A transport filter timeout is now identified in
4436 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
4437 systems where "make" and "gmake" are different, calling "gmake" at top
4438 level broke things. I've arranged for the value of $(MAKE) to be passed
4439 from the Makefile to this script so that it can call the same version of
4443 A note about Exim versions 4.44 and 4.50
4444 ----------------------------------------
4446 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
4447 changes of various kinds. As a consequence, a big documentation update was
4448 needed. This delayed the release for rather longer than seemed good, especially
4449 in the light of a couple of (minor) security issues. Therefore, the changes
4450 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
4451 release. So 4.44 and 4.50 are in effect two different branches that both start
4454 I have left the 4.50 change log unchanged; it contains all the changes since
4455 4.43. The change log for 4.44 is below; many of its items are identical to
4456 those for 4.50. This seems to be the most sensible way to preserve the
4457 historical information.
4463 1. Minor wording change to the doc/README.SIEVE file.
4465 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
4466 computation of the current number of files was incorrect.
4468 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4469 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4472 4. Give more explanation in the error message when the command for a transport
4473 filter fails to execute.
4475 5. There are several places where Exim runs a non-Exim command in a
4476 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4477 was being done only for the command run by the queryprogram router. It is
4478 now done for all such subprocesses. The other cases are: ${run, transport
4479 filters, and the commands run by the lmtp and pipe transports.
4481 6. Added CONFIGURE_GROUP build-time option.
4483 7. Some older OS have a limit of 256 on the maximum number of file
4484 descriptors. Exim was using setrlimit() to set 1000 as a large value
4485 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4486 systems. I've change it so that if it can't get 1000, it tries for 256.
4488 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4489 was an oversight, and furthermore, ever since the addition of extra
4490 controls (e.g. 4.43/32), the checks on when to allow different forms of
4491 "control" were broken. There should now be diagnostics for all cases when a
4492 control that does not make sense is encountered.
4494 9. Added the /retain_sender option to "control=submission".
4496 10. $recipients is now available in the predata ACL (oversight).
4498 11. Tidy the search cache before the fork to do a delivery from a message
4499 received from the command line. Otherwise the child will trigger a lookup
4500 failure and thereby defer the delivery if it tries to use (for example) a
4501 cached ldap connection that the parent has called unbind on.
4503 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4504 of $address_data from the recipient verification was clobbered by the
4505 sender verification.
4507 13. The value of address_data from a sender verification is now available in
4508 $sender_address_data in subsequent conditions in the ACL statement.
4510 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
4512 15. Added a new option "connect=<time>" to callout options, to set a different
4515 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4516 was its contents. (It was OK if the option was not defined at all.)
4518 17. A "Completed" log line is now written for messages that are removed from
4519 the spool by the -Mrm option.
4521 18. New variables $sender_verify_failure and $recipient_verify_failure contain
4522 information about exactly what failed.
4524 19. Added -dd to debug only the daemon process.
4526 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
4527 handles timeouts, both on the server side and network timeouts. Renamed the
4528 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
4530 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
4531 It is now set to "smtps".
4533 22. $host_address is now set to the target address during the checking of
4534 ignore_target_hosts.
4536 23. When checking ignore_target_hosts for an ipliteral router, no host name was
4537 being passed; this would have caused $sender_host_name to have been used if
4538 matching the list had actually called for a host name (not very likely,
4539 since this list is usually IP addresses). A host name is now passed as
4542 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4543 code that specifies a non-restarting handler (typically sigaction() in
4544 modern systems) in an attempt to fix a rare and obscure crash bug.
4546 25. Narrowed the window for a race in the daemon that could cause it to ignore
4547 SIGCHLD signals. This is not a major problem, because they are used only to
4548 wake it up if nothing else does.
4550 26. A malformed maildirsize file could cause Exim to calculate negative values
4551 for the mailbox size or file count. Odd effects could occur as a result.
4552 The maildirsize information is now recalculated if the size or filecount
4555 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4556 support for a long time. Removed HAVE_SYS_VFS_H.
4558 28. Installed the latest version of exipick from John Jetmore.
4560 29. In an address list, if the pattern was not a regular expression, an empty
4561 subject address (from a bounce message) matched only if the pattern was an
4562 empty string. Non-empty patterns were not even tested. This was the wrong
4563 because it is perfectly reasonable to use an empty address as part of a
4564 database query. An empty address is now tested by patterns that are
4565 lookups. However, all the other forms of pattern expect the subject to
4566 contain a local part and a domain, and therefore, for them, an empty
4567 address still always fails if the pattern is not itself empty.
4569 30. Exim went into a mad DNS loop when attempting to do a callout where the
4570 host was specified on an smtp transport, and looking it up yielded more
4571 than one IP address.
4573 31. Re-factored the code for checking spool and log partition space into a
4574 function that finds that data and another that does the check. The former
4575 is then used to implement four new variables: $spool_space, $log_space,
4576 $spool_inodes, and $log_inodes.
4578 32. The RFC2047 encoding function was originally intended for short strings
4579 such as real names; it was not keeping to the 75-character limit for
4580 encoded words that the RFC imposes. It now respects the limit, and
4581 generates multiple encoded words if necessary. To be on the safe side, I
4582 have increased the buffer size for the ${rfc2047: expansion operator from
4585 33. It is now permitted to omit both strings after an "if" condition; if the
4586 condition is true, the result is "true". As before, when the second string
4587 is omitted, a false condition yields an empty string. This makes it less
4588 cumbersome to write custom ACL and router conditions.
4590 34. Failure to deliver a bounce message always caused it to be frozen, even if
4591 there was an errors_to setting on the router. The errors_to setting is now
4594 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4595 canonical form (fully expanded) before being placed in
4596 $sender_host_address.
4598 36. The table in the code that translates DNS record types into text (T_A to
4599 "A" for instance) was missing entries for NS and CNAME. It is just possible
4600 that this could have caused confusion if both these types were looked up
4601 for the same domain, because the text type is used as part of Exim's
4602 per-process caching. But the chance of anyone hitting this buglet seems
4605 37. The dnsdb lookup has been extended in a number of ways.
4607 (1) There is a new type, "zns", which walks up the domain tree until it
4608 finds some nameserver records. It should be used with care.
4610 (2) There is a new type, "mxh", which is like "mx" except that it returns
4611 just the host names, not the priorities.
4613 (3) It is now possible to give a list of domains (or IP addresses) to be
4614 looked up. The behaviour when one of the lookups defers can be
4615 controlled by a keyword.
4617 (4) It is now possible to specify the separator character for use when
4618 multiple records are returned.
4620 38. The dnslists ACL condition has been extended: it is now possible to supply
4621 a list of IP addresses and/or domains to be looked up in a particular DNS
4624 39. Added log_selector=+queue_time_overall.
4626 40. When running the queue in the test harness, wait just a tad after forking a
4627 delivery process, to get repeatability of debugging output.
4629 41. Include certificate and key file names in error message when GnuTLS fails
4630 to set them up, because the GnuTLS error message doesn't include the name
4631 of the failing file when there is a problem reading it.
4633 42. Allow both -bf and -bF in the same test run.
4635 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
4637 44. The "Exiscan patch" is now merged into the mainline Exim source.
4639 45. Sometimes the final signoff response after QUIT could fail to get
4640 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4641 < 0 before doing a fflush(). This bug looks as though it goes back to the
4642 introduction of TLS in release 3.20, but "sometimes" must have been rare
4643 because the tests only now provoked it.
4645 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
4646 (this can affect the format of dates).
4648 47. exim_tidydb, when checking for the continued existence of a message for
4649 which it has found a message-specific retry record, was not finding
4650 messages that were in split spool directories. Consequently, it was
4651 deleting retry records that should have stayed in existence.
4653 48. Steve fixed some bugs in eximstats.
4655 49. The SPA authentication driver was not abandoning authentication and moving
4656 on to the next authenticator when an expansion was forced to fail,
4657 contradicting the general specification for all authenticators. Instead it
4658 was generating a temporary error. It now behaves as specified.
4660 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
4661 (the order specifies the preference for clients). The order is now AES256,
4662 AES128, 3DES, ARCFOUR128.
4664 51. Small patch to Sieve code - explicitly set From: when generating an
4667 52. Exim crashed if a remote delivery caused a very long error message to be
4668 recorded - for instance if somebody sent an entire SpamAssassin report back
4669 as a large number of 550 error lines. This bug was coincidentally fixed by
4670 increasing the size of one of Exim's internal buffers (big_buffer) that
4671 happened as part of the Exiscan merge. However, to be on the safe side, I
4672 have made the code more robust (and fixed the comments that describe what
4675 53. Now that there can be additional text after "Completed" in log lines (if
4676 the queue_time_overall log selector is set), a one-byte patch to exigrep
4677 was needed to allow it to recognize "Completed" as not the last thing in
4680 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
4681 patch that reportedly fixes this has been added. I am not expert enough to
4682 create a test for it. This is what the patch creator wrote:
4684 "I found a little strange behaviour of ldap code when working with
4685 Windows 2003 AD Domain, where users was placed in more than one
4686 Organization Units. When I tried to give exim partial DN, the exit code
4687 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
4688 But simultaneously result of request was absolutely normal ldap result,
4689 so I produce this patch..."
4691 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
4692 REFERENCE, so I have modified the code to exclude the patch when that macro
4695 55. Some experimental protocols are using DNS PTR records for new purposes. The
4696 keys for these records are domain names, not reversed IP addresses. The
4697 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4698 leaves it alone. Component reversal etc. now happens only for IP addresses.
4701 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
4703 57. Double the size of the debug message buffer (to 2048) so that more of very
4704 long debug lines gets shown.
4706 58. The exicyclog utility now does better if the number of log files to keep
4707 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
4709 59. Two changes related to the smtp_active_hostname option:
4711 (1) $smtp_active_hostname is now available as a variable.
4712 (2) The default for smtp_banner uses $smtp_active_hostname instead
4713 of $primary_hostname.
4715 60. The host_aton() function is supposed to be passed a string that is known
4716 to be a valid IP address. However, in the case of IPv6 addresses, it was
4717 not checking this. This is a hostage to fortune. Exim now panics and dies
4718 if the condition is not met. A case was found where this could be provoked
4719 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4720 components; fortuitously, this particular loophole had already been fixed
4721 by change 4.50/55 above.
4723 If there are any other similar loopholes, the new check in host_aton()
4724 itself should stop them being exploited. The report I received stated that
4725 data on the command line could provoke the exploit when Exim was running as
4726 exim, but did not say which command line option was involved. All I could
4727 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4728 running as the user.
4731 61. There was a buffer overflow vulnerability in the SPA authentication code
4732 (which came originally from the Samba project). I have added a test to the
4733 spa_base64_to_bits() function which I hope fixes it.
4736 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
4737 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
4739 63. The daemon start-up calls getloadavg() while still root for those OS that
4740 need the first call to be done as root, but it missed one case: when
4741 deliver_queue_load_max is set with deliver_drop_privilege. This is
4742 necessary for the benefit of the queue runner, because there is no re-exec
4743 when deliver_drop_privilege is set.
4745 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
4746 This has been fixed.
4748 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4749 was in use, was not putting the data itself into the right store pool;
4750 consequently, it could be overwritten for a subsequent message in the same
4751 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4754 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
4756 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
4757 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
4758 the function were treating the return as a boolean value, which happened to
4759 work because 0=false and not-0=true, but is not correct code.
4761 68. The host_aton() function was not handling scoped IPv6 addresses (those
4762 with, for example, "%eth0" on the end) correctly.
4764 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
4765 negated items (that is, ~something) in unsigned ints. Some compilers
4766 apparently mutter when there is no cast.
4768 70. If an address verification called from an ACL failed, and did not produce a
4769 user-specific message (i.e. there was only a "system" message), nothing was
4770 put in $acl_verify_message. In this situation, it now puts the system
4773 71. Change 4.23/11 added synchronization checking at the start of an SMTP
4774 session; change 4.31/43 added the unwanted input to the log line - except
4775 that it did not do this in the start of session case. It now does.
4777 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
4778 This is wrong and can cause the other end to generate a synchronization
4779 error if it is another Exim or anything else that does the synchronization
4780 check. A QUIT command is no longer sent after a timeout.
4782 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
4783 during host lookups.
4785 74. The defer_ok option of callout verification was not working if it was used
4786 when verifying addresses in header lines, that is, for this case:
4788 verify = header_sender/callout=defer_ok
4790 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
4791 those file descriptors could be used for SMTP connections. If anything
4792 wrote to stderr (the example that came up was "warn" in embedded Perl), it
4793 could be sent to the SMTP client, causing chaos. The daemon now opens
4794 stdin, stdout, and stderr to /dev/null when it puts itself into the
4797 76. Arrange for output from Perl's "warn" command to be written to Exim's main
4798 log by default. The user can override this with suitable Perl magic.
4800 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
4801 the log message when discard triggers, was not working for the DATA ACL or
4802 for the non-SMTP ACL.
4804 78. Error message wording change in sieve.c.
4806 79. If smtp_accept_max_per_host was set, the number of connections could be
4807 restricted to fewer than expected, because the daemon was trying to set up
4808 a new connection before checking whether the processes handling previous
4809 connections had finished. The check for completed processes is now done
4810 earlier. On busy systems, this bug wouldn't be noticed because something
4811 else would have woken the daemon, and it would have reaped the completed
4814 80. If a message was submitted locally by a user whose login name contained one
4815 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
4816 It caused a spool format error. I have fixed the spool reading code. A
4817 related problem was that the "from" clause in the Received: line became
4818 illegal because of the space(s). It is now covered by ${quote_local_part.
4820 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
4823 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
4825 83. Patch from Sieve maintainer to fix unterminated string problem in
4826 "vacation" handling.
4828 84. Some minor changes to the Linux configuration files to help with other
4829 OS variants using glibc.
4831 85. One more patch for Sieve to update vacation handling to latest spec.
4834 ----------------------------------------------------
4835 See the note above about the 4.44 and 4.50 releases.
4836 ----------------------------------------------------
4842 1. Change 4.43/35 introduced a bug that caused file counts to be
4843 incorrectly computed when quota_filecount was set in an appendfile
4846 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4847 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4850 3. Give more explanation in the error message when the command for a transport
4851 filter fails to execute.
4853 4. There are several places where Exim runs a non-Exim command in a
4854 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4855 was being done only for the command run by the queryprogram router. It is
4856 now done for all such subprocesses. The other cases are: ${run, transport
4857 filters, and the commands run by the lmtp and pipe transports.
4859 5. Some older OS have a limit of 256 on the maximum number of file
4860 descriptors. Exim was using setrlimit() to set 1000 as a large value
4861 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4862 systems. I've change it so that if it can't get 1000, it tries for 256.
4864 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4865 was an oversight, and furthermore, ever since the addition of extra
4866 controls (e.g. 4.43/32), the checks on when to allow different forms of
4867 "control" were broken. There should now be diagnostics for all cases when a
4868 control that does not make sense is encountered.
4870 7. $recipients is now available in the predata ACL (oversight).
4872 8. Tidy the search cache before the fork to do a delivery from a message
4873 received from the command line. Otherwise the child will trigger a lookup
4874 failure and thereby defer the delivery if it tries to use (for example) a
4875 cached ldap connection that the parent has called unbind on.
4877 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4878 of $address_data from the recipient verification was clobbered by the
4879 sender verification.
4881 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4882 was its contents. (It was OK if the option was not defined at all.)
4884 11. A "Completed" log line is now written for messages that are removed from
4885 the spool by the -Mrm option.
4887 12. $host_address is now set to the target address during the checking of
4888 ignore_target_hosts.
4890 13. When checking ignore_target_hosts for an ipliteral router, no host name was
4891 being passed; this would have caused $sender_host_name to have been used if
4892 matching the list had actually called for a host name (not very likely,
4893 since this list is usually IP addresses). A host name is now passed as
4896 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4897 code that specifies a non-restarting handler (typically sigaction() in
4898 modern systems) in an attempt to fix a rare and obscure crash bug.
4900 15. Narrowed the window for a race in the daemon that could cause it to ignore
4901 SIGCHLD signals. This is not a major problem, because they are used only to
4902 wake it up if nothing else does.
4904 16. A malformed maildirsize file could cause Exim to calculate negative values
4905 for the mailbox size or file count. Odd effects could occur as a result.
4906 The maildirsize information is now recalculated if the size or filecount
4909 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4910 support for a long time. Removed HAVE_SYS_VFS_H.
4912 18. Updated exipick to current release from John Jetmore.
4914 19. Allow an empty sender to be matched against a lookup in an address list.
4915 Previously the only cases considered were a regular expression, or an
4918 20. Exim went into a mad DNS lookup loop when doing a callout where the
4919 host was specified on the transport, if the DNS lookup yielded more than
4922 21. The RFC2047 encoding function was originally intended for short strings
4923 such as real names; it was not keeping to the 75-character limit for
4924 encoded words that the RFC imposes. It now respects the limit, and
4925 generates multiple encoded words if necessary. To be on the safe side, I
4926 have increased the buffer size for the ${rfc2047: expansion operator from
4929 22. Failure to deliver a bounce message always caused it to be frozen, even if
4930 there was an errors_to setting on the router. The errors_to setting is now
4933 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4934 canonical form (fully expanded) before being placed in
4935 $sender_host_address.
4937 24. Updated eximstats to version 1.33
4939 25. Include certificate and key file names in error message when GnuTLS fails
4940 to set them up, because the GnuTLS error message doesn't include the name
4941 of the failing file when there is a problem reading it.
4943 26. Expand error message when OpenSSL has problems setting up cert/key files.
4946 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
4947 (this can affect the format of dates).
4949 28. exim_tidydb, when checking for the continued existence of a message for
4950 which it has found a message-specific retry record, was not finding
4951 messages that were in split spool directories. Consequently, it was
4952 deleting retry records that should have stayed in existence.
4954 29. eximstats updated to version 1.35
4955 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
4956 1.35 - bugfix such that pie charts by volume are generated correctly
4958 30. The SPA authentication driver was not abandoning authentication and moving
4959 on to the next authenticator when an expansion was forced to fail,
4960 contradicting the general specification for all authenticators. Instead it
4961 was generating a temporary error. It now behaves as specified.
4963 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
4964 (the order specifies the preference for clients). The order is now AES256,
4965 AES128, 3DES, ARCFOUR128.
4967 31. Small patch to Sieve code - explicitly set From: when generating an
4970 32. Exim crashed if a remote delivery caused a very long error message to be
4971 recorded - for instance if somebody sent an entire SpamAssassin report back
4972 as a large number of 550 error lines. This bug was coincidentally fixed by
4973 increasing the size of one of Exim's internal buffers (big_buffer) that
4974 happened as part of the Exiscan merge. However, to be on the safe side, I
4975 have made the code more robust (and fixed the comments that describe what
4978 33. Some experimental protocols are using DNS PTR records for new purposes. The
4979 keys for these records are domain names, not reversed IP addresses. The
4980 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4981 leaves it alone. Component reversal etc. now happens only for IP addresses.
4984 34. The host_aton() function is supposed to be passed a string that is known
4985 to be a valid IP address. However, in the case of IPv6 addresses, it was
4986 not checking this. This is a hostage to fortune. Exim now panics and dies
4987 if the condition is not met. A case was found where this could be provoked
4988 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4989 components; fortuitously, this particular loophole had already been fixed
4990 by change 4.50/55 or 4.44/33 above.
4992 If there are any other similar loopholes, the new check in host_aton()
4993 itself should stop them being exploited. The report I received stated that
4994 data on the command line could provoke the exploit when Exim was running as
4995 exim, but did not say which command line option was involved. All I could
4996 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4997 running as the user.
5000 35. There was a buffer overflow vulnerability in the SPA authentication code
5001 (which came originally from the Samba project). I have added a test to the
5002 spa_base64_to_bits() function which I hope fixes it.
5005 36. The daemon start-up calls getloadavg() while still root for those OS that
5006 need the first call to be done as root, but it missed one case: when
5007 deliver_queue_load_max is set with deliver_drop_privilege. This is
5008 necessary for the benefit of the queue runner, because there is no re-exec
5009 when deliver_drop_privilege is set.
5011 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
5012 was in use, was not putting the data itself into the right store pool;
5013 consequently, it could be overwritten for a subsequent message in the same
5014 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
5017 38. Sometimes the final signoff response after QUIT could fail to get
5018 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
5019 < 0 before doing a fflush(). This bug looks as though it goes back to the
5020 introduction of TLS in release 3.20, but "sometimes" must have been rare
5021 because the tests only now provoked it.
5027 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
5028 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
5029 Now it returns an integer. A number of calls were still expecting a T/F
5030 return. Fortuitously, in all cases, the tests worked in OK situations,
5031 which is the norm. However, things would have gone wrong on any write
5032 failures on the smtp file descriptor. This function is used when sending
5033 messages over SMTP and also when doing verify callouts.
5035 2. When Exim is called to do synchronous delivery of a locally submitted
5036 message (the -odf or -odi options), it no longer closes stderr before doing
5039 3. Implemented the mua_wrapper option.
5041 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
5043 5. Implemented the functions header_remove(), header_testname(),
5044 header_add_at_position(), and receive_remove_recipient(), and exported them
5047 6. If an ACL "warn" statement specified the addition of headers, Exim already
5048 inserted X-ACL-Warn: at the start if there was no header name. However, it
5049 was not making this test for the second and subsequent header lines if
5050 there were newlines in the string. This meant that an invalid header could
5051 be inserted if Exim was badly configured.
5053 7. Allow an ACL "warn" statement to add header lines at the start or after all
5054 the Received: headers, as well as at the end.
5056 8. Added the rcpt_4xx retry error code.
5058 9. Added postmaster_mailfrom=xxx to callout verification option.
5060 10. Added mailfrom=xxxx to the callout verification option, for verify=
5063 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
5064 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
5066 12. Inserted some casts to stop certain compilers warning when using pointer
5067 differences as field lengths or precisions in printf-type calls (mostly
5068 affecting debugging statements).
5070 13. Added optional readline() support for -be (dynamically loaded).
5072 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
5073 same clock tick as a message's arrival, so that its received time was the
5074 same as the "first fail" time on the retry record, and that message
5075 remained on the queue past the ultimate address timeout, every queue runner
5076 would try a delivery (because it was past the ultimate address timeout) but
5077 after another failure, the ultimate address timeout, which should have then
5078 bounced the address, did not kick in. This was a "< instead of <=" error;
5079 in most cases the first failure would have been in the next clock tick
5080 after the received time, and all would be well.
5082 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
5083 being recognized when the domain list was tested by the match_domain
5084 condition in an expansion string.
5086 16. Added the ${str2b64: operator.
5088 17. Exim was always calling setrlimit() to set a large limit for the number of
5089 processes, without checking whether the existing limit was already
5090 adequate. (It did check for the limit on file descriptors.) Furthermore,
5091 errors from getrlimit() and setrlimit() were being ignored. Now they are
5092 logged to the main and panic logs, but Exim does carry on, to try to do its
5093 job under whatever limits there are.
5095 18. Imported PCRE 5.0.
5097 19. Trivial typo in log message " temporarily refused connection" (the leading
5100 20. If the log selector return_path_on_delivery was set and an address was
5101 redirected to /dev/null, the delivery process crashed because it assumed
5102 that a return path would always be set for a "successful" delivery. In this
5103 case, the whole delivery is bypassed as an optimization, and therefore no
5106 21. Internal re-arrangement: the function for sending a challenge and reading
5107 a response while authentication was assuming a zero-terminated challenge
5108 string. It's now changed to take a pointer and a length, to allow for
5109 binary data in such strings.
5111 22. Added the cyrus_sasl authenticator (code supplied by MBM).
5113 23. Exim was not respecting finduser_retries when seeking the login of the
5114 uid under which it was called; it was always trying 10 times. (The default
5115 setting of finduser_retries is zero.) Also, it was sleeping after the final
5116 failure, which is pointless.
5118 24. Implemented tls_on_connect_ports.
5120 25. Implemented acl_smtp_predata.
5122 26. If the domain in control=submission is set empty, Exim assumes that the
5123 authenticated id is a complete email address when it generates From: or
5124 Sender: header lines.
5126 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
5127 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
5128 chown and chgrp in /bin and hostname in /usr/bin.
5130 28. Exim was keeping the "process log" file open after each use, just as it
5131 does for the main log. This opens the possibility of it remaining open for
5132 long periods when the USR1 signal hits a daemon. Occasional processlog
5133 errors were reported, that could have been caused by this. Anyway, it seems
5134 much more sensible not to leave this file open at all, so that is what now
5137 29. The long-running daemon process does not normally write to the log once it
5138 has entered its main loop, and it closes the log before doing so. This is
5139 so that log files can straightforwardly be renamed and moved. However,
5140 there are a couple of unusual error situations where the daemon does write
5141 log entries, and I had neglected to close the log afterwards.
5143 30. The text of an SMTP error response that was received during a remote
5144 delivery was being truncated at 512 bytes. This is too short for some of
5145 the long messages that one sometimes sees. I've increased the limit to
5148 31. It is now possible to make retry rules that apply only when a message has a
5149 specific sender, in particular, an empty sender.
5151 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
5152 it possible to be selective about when SMTP synchronization is enforced.
5154 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
5156 32. Implemented hosts_connection_nolog.
5158 33. Added an ACL for QUIT.
5160 34. Setting "delay_warning=" to disable warnings was not working; it gave a
5163 35. Added mailbox_size and mailbox_filecount to appendfile.
5165 36. Added control = no_multiline_responses to ACLs.
5167 37. There was a bug in the logic of the code that waits for the clock to tick
5168 in the case where the clock went backwards by a substantial amount such
5169 that the microsecond fraction of "now" was more than the microsecond
5170 fraction of "then" (but the whole seconds number was less).
5172 38. Added support for the libradius Radius client library this is found on
5173 FreeBSD (previously only the radiusclient library was supported).
5179 1. When certain lookups returned multiple values in the form name=value, the
5180 quoting of the values was not always being done properly. Specifically:
5181 (a) If the value started with a double quote, but contained no whitespace,
5183 (b) If the value contained whitespace other than a space character (i.e.
5184 tabs or newlines or carriage returns) it was not quoted.
5185 This fix has been applied to the mysql and pgsql lookups by writing a
5186 separate quoting function and calling it from the lookup code. The fix
5187 should probably also be applied to nisplus, ibase and oracle lookups, but
5188 since I cannot test any of those, I have not disturbed their existing code.
5190 2. A hit in the callout cache for a specific address caused a log line with no
5191 reason for rejecting RCPT. Now it says "Previous (cached) callout
5192 verification failure".
5194 3. There was an off-by-one bug in the queryprogram router. An over-long
5195 return line was truncated at 256 instead of 255 characters, thereby
5196 overflowing its buffer with the terminating zero. As well as fixing this, I
5197 have increased the buffer size to 1024 (and made a note to document this).
5199 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
5200 when Exim is waiting for an SMTP response from a remote server, Exim
5201 restarts its select() call on the socket, thereby resetting its timeout.
5202 This is not a problem when such interrupts are rare. Somebody set up a cron
5203 job to run exiwhat every 2 minutes, which is less than the normal select()
5204 timeout (5 or 10 minutes). This meant that the select() timeout never
5205 kicked in because it was always reset. I have fixed this by comparing the
5206 time when an interrupt arrives with the time at the start of the first call
5207 to select(). If more time than the timeout has elapsed, the interrupt is
5208 treated as a timeout.
5210 5. Some internal re-factoring in preparation for the addition of Sieve
5211 extensions (by MH). In particular, the "personal" test is moved to a
5212 separate function, and given an option for scanning Cc: and Bcc: (which is
5213 not set for Exim filters).
5215 6. When Exim created an email address using the login of the caller as the
5216 local part (e.g. when creating a From: or Sender: header line), it was not
5217 quoting the local part when it contained special characters such as @.
5219 7. Installed new OpenBSD configuration files.
5221 8. Reworded some messages for syntax errors in "and" and "or" conditions to
5222 try to make them clearer.
5224 9. Callout options, other than the timeout value, were being ignored when
5225 verifying sender addresses in header lines. For example, when using
5227 verify = header_sender/callout=no_cache
5229 the cache was (incorrectly) being used.
5231 10. Added a missing instance of ${EXE} to the exim_install script; this affects
5232 only the Cygwin environment.
5234 11. When return_path_on_delivery was set as a log selector, if different remote
5235 addresses in the same message used different return paths and parallel
5236 remote delivery occurred, the wrong values would sometimes be logged.
5237 (Whenever a remote delivery process finished, the return path value from
5238 the most recently started remote delivery process was logged.)
5240 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
5241 lines when AUTH and/or TLS are in use. This is the "received protocol"
5242 field. Exim used to use "asmtp" for authenticated SMTP, without any
5243 indication (in the protocol name) for TLS use. Now it follows the RFC and
5244 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
5245 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
5246 names appear in log lines as well as in Received: header lines.
5248 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
5249 extensions, and comparison tests, and to fix some bugs.
5251 14. Changes to the "personal" filter test:
5253 (1) The test was buggy in that it was just doing the equivalent of
5254 "contains" tests on header lines. For example, if a user's address was
5255 anne@some.where, the "personal" test would incorrectly be true for
5257 To: susanne@some.where
5259 This test is now done by extracting each address from the header in turn,
5260 and checking the entire address. Other tests that are part of "personal"
5261 are now done using regular expressions (for example, to check local parts
5262 of addresses in From: header lines).
5264 (2) The list of non-personal local parts in From: addresses has been
5265 extended to include "listserv", "majordomo", "*-request", and "owner-*",
5266 taken from the Sieve specification recommendations.
5268 (3) If the message contains any header line starting with "List-" it is
5269 treated as non-personal.
5271 (4) The test for "circular" in the Subject: header line has been removed
5272 because it now seems ill-conceived.
5274 15. Minor typos in src/EDITME comments corrected.
5276 16. Installed latest exipick from John Jetmore.
5278 17. If headers_add on a router specified a text string that was too long for
5279 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
5280 of string_sprintf() is now avoided.
5282 18. $message_body_size was not set (it was always zero) when running the DATA
5283 ACL and the local_scan() function.
5285 19. For the "mail" command in an Exim filter, no default was being set for
5286 the once_repeat time, causing a random time value to be used if "once" was
5287 specified. (If the value happened to be <= 0, no repeat happened.) The
5288 default is now 0s, meaning "never repeat". The "vacation" command was OK
5289 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
5290 (I found it when inspecting the code).
5292 20. There is now an overall timeout for performing a callout verification. It
5293 defaults to 4 times the callout timeout, which applies to individual SMTP
5294 commands during the callout. The overall timeout applies when there is more
5295 than one host that can be tried. The timeout is checked before trying the
5296 next host. This prevents very long delays if there are a large number of
5297 hosts and all are timing out (e.g. when the network connections are timing
5298 out). The value of the overall timeout can be changed by specifying an
5299 additional sub-option for "callout", called "maxwait". For example:
5301 verify = sender/callout=5s,maxwait=20s
5303 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
5304 to the end before writing, but this should make it even safer).
5306 22. Exim was forgetting that it had advertised PIPELINING for the second and
5307 subsequent messages on an SMTP connection. It was also not resetting its
5308 memory on STARTTLS and an internal HELO.
5310 23. When Exim logs an SMTP synchronization error within a session, it now
5311 records whether PIPELINING has been advertised or not.
5313 24. Added 3 instances of "(long int)" casts to time_t variables that were being
5314 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
5315 rather than long int.
5317 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
5319 26. Added the never_mail option to autoreply.
5325 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
5326 crash if the getsockname() call failed; this can happen if a connection is
5327 closed very soon after it is established. The problem was simply in the
5328 order in which certain operations were done, causing Exim to try to write
5329 to the SMTP stream before it had set up the file descriptor. The bug has
5330 been fixed by making things happen in the correct order.
5336 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
5337 before the connection was closed, thus losing the rejection response.
5339 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
5340 some early Solaris releases, but causes trouble in current releases where
5341 socklen_t is defined.
5343 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
5346 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
5349 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
5350 that releasing the top of it at the end releases what was used for sub-
5351 expansions (unless the block got too big). However, discard this block if
5352 the first thing is a variable or header, so that we can use its block when
5353 it is dynamic (useful for very large $message_headers, for example).
5355 6. Lookups now cache *every* query, not just the most recent. A new, separate
5356 store pool is used for this. It can be recovered when all lookup caches are
5357 flushed. Lookups now release memory at the end of their result strings.
5358 This has involved some general refactoring of the lookup sources.
5360 7. Some code has been added to the store_xxx() functions to reduce the amount
5361 of flapping under certain conditions.
5363 8. log_incoming_interface used to affect only the <= reception log lines. Now
5364 it causes the local interface and port to be added to several more SMTP log
5365 lines, for example "SMTP connection from", and rejection lines.
5367 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
5369 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
5371 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
5372 could be overwritten at the end of the current message (or the start of a
5373 new message if it was set in a HELO ACL). The value is now preserved for
5374 the duration of the SMTP connection.
5376 12. If a transport had a headers_rewrite setting, and a matching header line
5377 contained an unqualified address, that address was qualified, even if it
5378 did not match any rewriting rules. The underlying bug was that the values
5379 of the flags that permit the existence of unqualified sender and recipient
5380 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
5381 non-local messages, and by -bnq for local messages) were not being
5382 preserved with the message after it was received.
5384 13. When Exim was logging an SMTP synchronization error, it could sometimes log
5385 "next input=" as part of the text comprising the host identity instead of
5386 the correct text. The code was using the same buffer for two different
5387 strings. However, depending on which order the printing function evaluated
5388 its arguments, the bug did not always show up. Under Linux, for example, my
5389 test suite worked just fine.
5391 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
5392 doesn't work with some older versions of Perl. It has been changed to "my",
5393 which in any case is probably the better facility to use.
5395 15. A really picky compiler found some instances of statements for creating
5396 error messages that either had too many or two few arguments for the format
5399 16. The size of the buffer for calls to the DNS resolver has been increased
5400 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
5401 for addresses that have a lot of PTR records. This alleviates a problem; it
5402 does not fully solve it.
5404 17. A dnsdb lookup for PTR records that receives more data than will fit in the
5405 buffer now truncates the list and logs the incident, which is the same
5406 action as happens when Exim is looking up a host name and its aliases.
5407 Previously in this situation something unpredictable would happen;
5408 sometimes it was "internal error: store_reset failed".
5410 18. If a server dropped the connection unexpectedly when an Exim client was
5411 using GnuTLS and trying to read a response, the client delivery process
5412 crashed while trying to generate an error log message.
5414 19. If a "warn" verb in an ACL added multiple headers to a message in a single
5415 string, for example:
5417 warn message = H1: something\nH2: something
5419 the text was added as a single header line from Exim's point of view
5420 though it ended up OK in the delivered message. However, searching for the
5421 second and subsequent header lines using $h_h2: did not work. This has been
5422 fixed. Similarly, if a system filter added multiple headers in this way,
5423 the routers could not see them.
5425 20. Expanded the error message when iplsearch is called with an invalid key to
5426 suggest using net-iplsearch in a host list.
5428 21. When running tests using -bh, any delays imposed by "delay" modifiers in
5429 ACLs are no longer actually imposed (and a message to that effect is
5432 22. If a "gecos" field in a passwd entry contained escaped characters, in
5433 particular, if it contained a \" sequence, Exim got it wrong when building
5434 a From: or a Sender: header from that name. A second bug also caused
5435 incorrect handling when an unquoted " was present following a character
5436 that needed quoting.
5438 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
5439 was not being matched caselessly.
5441 24. Arranged for all hyphens in the exim.8 source to be escaped with
5444 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
5445 properly. Recipient callout cache records were still being keyed to include
5446 the sender, even when use_sender was set false. This led to far more
5447 callouts that were necessary. The sender is no longer included in the key
5448 when use_sender is false.
5450 26. Added "control = submission" modifier to ACLs.
5452 27. Added the ${base62d: operator to decode base 62 numbers.
5454 28. dnsdb lookups can now access SRV records.
5456 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
5457 the configuration file.
5459 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
5460 (-v) mode. This makes the output for a verbose queue run more intelligible.
5462 31. Added a use_postmaster feature to recipient callouts.
5464 32. Added the $body_zerocount variable, containing the number of binary zero
5465 bytes in the message body.
5467 33. The time of last modification of the "new" subdirectory is now used as the
5468 "mailbox time last read" when there is a quota error for a maildir
5471 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
5473 35. Added +ignore_unknown as a special item in host lists.
5475 36. Code for decoding IPv6 addresses in host lists is now included, even if
5476 IPv6 support is not being compiled. This fixes a bug in which an IPv6
5477 address was recognized as an IP address, but was then not correctly decoded
5478 into binary, causing unexpected and incorrect effects when compared with
5485 1. Very minor rewording of debugging text in manualroute to say "list of
5486 hosts" instead of "hostlist".
5488 2. If verify=header_syntax was set, and a header line with an unqualified
5489 address (no domain) and a large number of spaces between the end of the
5490 name and the colon was received, the reception process suffered a buffer
5491 overflow, and (when I tested it) crashed. This was caused by some obsolete
5492 code that should have been removed. The fix is to remove it!
5494 3. When running in the test harness, delay a bit after writing a bounce
5495 message to get a bit more predictability in the log output.
5497 4. Added a call to search_tidyup() just before forking a reception process. In
5498 theory, someone could use a lookup in the expansion of smtp_accept_max_
5499 per_host which, without the tidyup, could leave open a database connection.
5501 5. Added the variables $recipient_data and $sender_data which get set from a
5502 lookup success in an ACL "recipients" or "senders" condition, or a router
5503 "senders" option, similar to $domain_data and $local_part_data.
5505 6. Moved the writing of debug_print from before to after the "senders" test
5508 7. Change 4.31/66 (moving the time when the Received: is generated) caused
5509 problems for message scanning, either using a data ACL, or using
5510 local_scan() because the Received: header was not generated till after they
5511 were called (in order to set the time as the time of reception completion).
5512 I have revised the way this works. The header is now generated after the
5513 body is received, but before the ACL or local_scan() are called. After they
5514 are run, the timestamp in the header is updated.
5520 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
5521 before starting a queue runner without re-exec. This happened only when
5522 deliver_drop_privilege was set or when the Exim user was set to root. The
5523 effect of the bug was that timeouts during subsequent deliveries caused
5524 crashes instead of being properly handled. The handler is now left at its
5525 default (and expected) setting.
5527 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
5528 message, again when deliver_drop_privilege is set or Exim is run as root.
5529 The bug described in (1) was not present in this case, but the tidying up
5530 of the other signals was missing. I have made the two cases consistent.
5532 3. The ignore_target_hosts setting on a manualroute router was being ignored
5533 for hosts that were looked up using the /MX notation.
5535 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
5538 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
5539 operated on the sender address. After changing the $sender_address to <>
5540 for the sender address verify, Exim was re-instated it as the original
5541 (before rewriting) address, but remembering that it had rewritten it, so it
5542 wasn't rewriting it again. This bug also had the effect of breaking the
5543 sender address verification caching when the sender address was rewritten.
5545 6. The ignore_target_hosts option was being ignored by the ipliteral router.
5546 This has been changed so that if the ip literal address matches
5547 ignore_target_hosts, the router declines.
5549 7. Added expansion conditions match_domain, match_address, and match_local_
5550 part (NOT match_host).
5552 8. The placeholder for the Received: header didn't have a length field set.
5554 9. Added code to Exim itself and to exim_lock to test for a specific race
5555 condition that could lead to file corruption when using MBX delivery. The
5556 issue is with the lockfile that is created in /tmp. If this file is removed
5557 after a process has opened it but before that process has acquired a lock,
5558 there is the potential for a second process to recreate the file and also
5559 acquire a lock. This could lead to two Exim processes writing to the file
5560 at the same time. The added code performs the same test as UW imapd; it
5561 checks after acquiring the lock that its file descriptor still refers to
5562 the same named file.
5564 10. The buffer for building added header lines was of fixed size, 8192 bytes.
5565 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
5568 11. Added the smtp_active_hostname option. If used, this will typically be made
5569 to depend on the incoming interface address. Because $interface_address is
5570 not set up until the daemon has forked a reception process, error responses
5571 that can happen earlier (such as "too many connections") no longer contain
5574 12. If an expansion in a condition on a "warn" statement fails because a lookup
5575 defers, the "warn" statement is abandoned, and the next ACL statement is
5576 processed. Previously this caused the whole ACL to be aborted.
5578 13. Added the iplsearch lookup type.
5580 14. Added ident_timeout as a log selector.
5582 15. Added tls_certificate_verified as a log selector.
5584 16. Added a global option tls_require_ciphers (compare the smtp transport
5585 option of the same name). This controls incoming TLS connections.
5587 17. I finally figured out how to make tls_require_ciphers do a similar thing
5588 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
5589 before starting the TLS session.
5591 18. Tabs are now shown as \t in -bP output.
5593 19. If the log selector return_path_on_delivery was set, Exim crashed when
5594 bouncing a message because it had too many Received: header lines.
5596 20. If two routers both had headers_remove settings, and the first one included
5597 a superfluous trailing colon, the final name in the first list and the
5598 first name in the second list were incorrectly joined into one item (with a
5599 colon in the middle).
5605 1. Added -C and -D options to the exinext utility, mainly to make it easier
5606 to include in the automated testing, but these could be helpful when
5607 multiple configurations are in use.
5609 2. The exinext utility was not formatting the output nicely when there was
5610 an alternate port involved in the retry record key, nor when there was a
5611 message id as well (for retries that were specific to a specific message
5612 and a specific host). It was also confused by IPv6 addresses, because of
5613 the additional colons they contain. I have fixed the IPv4 problem, and
5614 patched it up to do a reasonable job for IPv6.
5616 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
5617 delivery, the log line now contains "pipelined" if PIPELINING was used.
5619 4. An SMTP transport process used to panic and die if the bind() call to set
5620 an explicit outgoing interface failed. This has been changed; it is now
5621 treated in the same way as a connect() failure.
5623 5. A reference to $sender_host_name in the part of a conditional expansion
5624 that was being skipped was still causing a DNS lookup. This no longer
5627 6. The def: expansion condition was not recognizing references to header lines
5628 that used bh_ and bheader_.
5630 7. Added the _cache feature to named lists.
5632 8. The code for checking quota_filecount in the appendfile transport was
5633 allowing one more file than it should have been.
5635 9. For compatibility with Sendmail, the command line option
5643 and sets the incoming protocol and host name (for trusted callers). The
5644 host name and its colon can be omitted when only the protocol is to be set.
5645 Note the Exim already has two private options, -pd and -ps, that refer to
5646 embedded Perl. It is therefore impossible to set a protocol value of "d" or
5647 "s", but I don't think that's a major issue.
5649 10. A number of refactoring changes to the code, none of which should affect
5652 (a) The number of logging options was getting close to filling up the
5653 32-bit word that was used as a bit map. I have split them into two classes:
5654 those that are passed in the argument to log_write(), and those that are
5655 only ever tested independently outside of that function. These are now in
5656 separate 32-bit words, so there is plenty of room for expansion again.
5657 There is no change in the user interface or the logging behaviour.
5659 (b) When building, for example, log lines, the code previously used a
5660 macro that called string_cat() twice, in order to add two strings. This is
5661 not really sufficiently general. Furthermore, there was one instance where
5662 it was actually wrong because one of the argument was used twice, and in
5663 one call a function was used. (As it happened, calling the function twice
5664 did not affect the overall behaviour.) The macro has been replaced by a
5665 function that can join an arbitrary number of extra strings onto a growing
5668 (c) The code for expansion conditions now uses a table and a binary chop
5669 instead of a serial search (which was left over from when there were very
5670 few conditions). Also, it now recognizes conditions like "pam" even when
5671 the relevant support is not compiled in: a suitably worded error message is
5672 given if an attempt is made to use such a condition.
5674 11. Added ${time_interval:xxxxx}.
5676 12. A bug was causing one of the ddress fields not to be passed back correctly
5677 from remote delivery subprocesses. The field in question was not being
5678 subsequently used, so this caused to problems in practice.
5680 13. Added new log selectors queue_time and deliver_time.
5682 14. Might have fixed a bug in maildirsizefile handling that threw up
5683 "unexpected character" debug warnings, and recalculated the data
5684 unnecessarily. In any case, I expanded the warning message to give more
5687 15. Added the message "Restricted characters in address" to the statements in
5688 the default ACL that block characters like @ and % in local parts.
5690 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
5691 Three changes have been made:
5693 (a) There was a serious bug; a negative response to MAIL caused the whole
5694 recipient domain to be cached as invalid, thereby blocking all messages
5695 to all local parts at the same domain, from all senders. This bug has
5696 been fixed. The domain is no longer cached after a negative response to
5697 MAIL if the sender used is not empty.
5699 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
5702 (c) A new callout option, "use_sender" has been added for people who want
5703 the modified behaviour.
5709 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
5712 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
5713 indeed breaks things for older releases.
5715 3. Added additional logging to the case where there is a problem reading data
5716 from a filter that is running in a subprocess using a pipe, in order to
5717 try to track down a specific problem.
5719 4. Testing facility fudge: when running in the test harness and attempting
5720 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
5721 getting "No route to host". Convert this to a timeout.
5723 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
5726 6. Some OS don't have socklen_t but use size_t instead. This affects the
5727 fifth argument of getsockopt() amongst other things. This is now
5728 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
5729 can be set for individual OS. I have set it for SunOS5, OSF1, and
5730 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
5731 some earlier ones do not.
5733 7. Change 4.30/15 was not doing the test caselessly.
5735 8. The standard form for an IPv6 address literal was being rejected by address
5736 parsing in, for example, MAIL and RCPT commands. An example of this kind of
5737 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5738 this, as well as the form without the "IPv6" on the front (but only when
5739 address literals are enabled, of course).
5741 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
5743 10. Exim crashed if a message with an empty sender address specified by -f
5744 encountered a router with an errors_to setting. This could be provoked only
5745 by a command such as
5749 where an empty string was supplied; "<>" did not hit this bug.
5751 11. Installed PCRE release 4.5.
5753 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
5754 remained set. It is now erased.
5756 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
5757 times from message ids (which are base 36 rather than the normal 62).
5759 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
5760 were being counted as actual protocol errors, and logged if the log
5761 selector +smtp_protocol_error was set. One cannot be perfect in this test,
5762 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
5763 and DATA following a set of rejected RCPTs do not count as protocol errors.
5764 In other words, Exim assumes they were pipelined, though this may not
5765 actually be the case. Of course, in all cases the client gets an
5766 appropriate error code.
5768 15. If a lookup fails in an ACL condition, a message about the failure may
5769 be available; it is used if testing the ACL cannot continue, because most
5770 such messages specify what the cause of the deferral is. However, some
5771 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
5772 that caused an old message to be retained and used if a later statement
5773 caused a defer, replacing the real cause of the deferral.
5775 16. If an IP address had so many PTR records that the DNS lookup buffer
5776 was not large enough to hold them, Exim could crash while trying to process
5777 the truncated data. It now detects and logs this case.
5779 17. Further to 4.21/58, another change has been made: if (and only if) the
5780 first line of a message (the first header line) ends with CRLF, a bare LF
5781 in a subsequent header line has a space inserted after it, so as not to
5782 terminate the header.
5784 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
5785 unnecessarily, used atoi() instead of strtol(), and didn't check the
5786 termination when getting file sizes from file names by regex.
5788 19. Completely re-implemented the support for maildirsize files, in the light
5789 of a number of problems with the previous contributed implementation
5790 (4.30/29). In particular:
5792 . If the quota is zero, the maildirsize file is maintained, but no quota is
5795 . If the maildir directory does not exist, it is created before any attempt
5796 to write a maildirsize file.
5798 . The quota value in the file is just a cache; if the quota is changed in
5799 the transport, the new value overrides.
5801 . A regular expression is available for excluding directories from the
5804 20. The autoreply transport checks the characters in options that define the
5805 message's headers; it allows continued headers, but it was checking with
5806 isspace() after an embedded newline instead of explicitly looking for a
5809 21. If all the "regular" hosts to which an address was routed had passed their
5810 expiry times, and had not reached their retry times, the address was
5811 bounced, even if fallback hosts were defined. Now Exim should go on to try
5814 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
5815 equivalent code in the SMTP transport. Some hosts send humungous responses
5816 to HELO/EHLO, more than 1024 it seems.
5818 23. Refactoring: code in filter.c used (void *) for "any old type" but this
5819 gives compiler warnings in some environments. I've now done it "properly",
5822 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
5823 (because of problems with the built-in one) was declared to return uschar *
5824 instead of char *, causing compiler failure.
5826 25. Fixed a file descriptor leak when processing alias/forward files.
5828 26. Fixed a minor format string issue in dbfn.c.
5830 27. Typo in exim.c: ("dmbnz" for "dbmnz").
5832 28. If a filter file refered to $h_xxx or $message_headers, and the headers
5833 contained RFC 2047 "words", Exim's memory could, under certain conditions,
5836 29. When a sender address is verified, it is cached, to save repeating the test
5837 when there is more than one recipient in a message. However, when the
5838 verification involves a callout, it is possible for different callout
5839 options to be set for different recipients. It is too complicated to keep
5840 track of this in the cache, so now Exim always runs a verification when a
5841 callout is required, relying on the callout cache for the optimization.
5842 The overhead is duplication of the address routing, but this should not be
5845 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
5846 to be verified with callout=postmaster, and the main callout worked but the
5847 postmaster check failed, the verification correctly failed. However, if a
5848 subsequent RCPT command asked for sender verification *without* the
5849 postmaster check, incorrect caching caused this verification also to fail,
5852 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
5853 it was not caching the DNS options (qualify_single, search_parents) that
5854 were used when the lookup failed. A subsequent lookup with different
5855 options therefore always gave the same answer, though there were cases
5856 where it should not have. (Example: a "domains = !$mx_any" option on a
5857 dnslookup router: the "domains" option is always processed without any
5858 widening, but the router might have qualify_single set.) Now Exim uses the
5859 cached value only when the same options are set.
5861 32. Added John Jetmore's "exipick" utility to the distribution.
5863 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
5864 than a timeout (e.g. a certificate is required, and is not provided), an
5865 Exim server now closes the connection immediately. Previously it waited for
5866 the client to close - but if the client is SSL, it seems that they each
5867 wait for each other, leading to a delay before one of them times out.
5869 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
5870 maintained 0.8.x compatibility because I don't think many are using it, and
5871 it is clearly obsolete.
5873 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
5876 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
5877 client certificate was expired. A simple patch fixes this, though I don't
5878 understand the full logic of why the verify callback is called multiple
5881 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
5882 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
5883 which causes problems with some clients (such as the Certicom SSL Plus
5884 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
5885 disables the coutermeasure allowing Eudora to connect."
5887 38. Exim was not checking that a write() to a log file succeeded. This could
5888 lead to Bad Things if a log got too big, in particular if it hit a file
5889 size limit. Exim now panics and dies if it cannot write to a log file, just
5890 as it does if it cannot open a log file.
5892 39. Modified OS/Makefile-Linux so that it now contains
5894 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
5896 The two -D definitions ensure that Exim is compiled with large file
5897 support, which makes it possible to handle log files that are bigger than
5900 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
5901 instance) a domain was checked against a named list that involved a lookup,
5902 causing $domain_data to be set, then another domain was checked against the
5903 same list, then the first domain was re-checked, the value of $domain_data
5904 after the final check could be wrong. In particular, if the second check
5905 failed, it could be set empty. This bug probably also applied to
5908 41. The strip_trailing_dot option was not being applied to the address given
5909 with the -f command-line option.
5911 42. The code for reading a message's header from the spool was incrementing
5912 $received_count, but never initializing it. This meant that the value was
5913 incorrect (doubled) while delivering a message in the same process in which
5914 it was received. In the most common configuration of Exim, this never
5915 happens - a fresh exec is done - but it can happen when
5916 deliver_drop_privilege is set.
5918 43. When Exim logs an SMTP synchronization error - client data sent too soon -
5919 it now includes up to 150 characters of the unexpected data in the log
5922 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
5923 and building data strings. The size of both of these buffers was 10 000
5924 bytes - far larger than anybody would *ever* want, thought I. Needless to
5925 say, somebody hit the limit. I have increased the maximum line length to
5926 20 000 and the maximum data length of concatenated lines to 100 000. I have
5927 also fixed two bugs, because there was no checking on these buffers. Tsk,
5928 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
5929 buffer is too small.
5931 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
5932 lsearch lookups. Now it does.
5934 46. When parsing a route_list item in a manualroute router, a fixed-length
5935 buffer was used for the list of hosts. I made this 1024 bytes long,
5936 thinking that nobody would ever have a list of hosts that long. Wrong.
5937 Somebody had a whole pile of complicated expansion conditions, and the
5938 string was silently truncated, leading to an expansion error. It turns out
5939 that it is easier to change to an unlimited length (owing to other changes
5940 that have happened since this code was originally written) than to build
5941 structure for giving a limitation error. The length of the item that
5942 expands into the list of hosts is now unlimited.
5944 47. The lsearch lookup could not handle data where the length of text line was
5945 more than 4095 characters. Such lines were truncated, leading to shortened
5946 data being returned. It should now handle lines of any length.
5948 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
5949 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
5952 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
5953 replacing strings such as BIN_DIRECTORY in a source file: the replacement
5954 no longer happens in comment lines. A list of replacements is now placed
5955 at the head of all of the source files, except those whose only change is
5956 to replace PERL_COMMAND in the very first #! line.
5958 50. Replaced the slow insertion sort in queue.c, for sorting the list of
5959 messages on the queue, with a bottom-up merge sort, using code contributed
5960 by Michael Haardt. This should make operations like -bp somewhat faster on
5961 large queues. It won't affect queue runners, except when queue_run_in_order
5964 51. Installed eximstats 1.31 in the distribution.
5966 52. Added support for SRV lookups to the dnslookup router.
5968 53. If an ACL referred to $message_body or $message_body_end, the value was not
5969 reset for any messages that followed in the same SMTP session.
5971 54. The store-handling optimization for building very long strings was not
5972 differentiating between the different store pools. I don't think this
5973 actually made any difference in practice, but I've tidied it.
5975 55. While running the routers to verify a sender address, $sender_address
5976 was still set to the sender address. This is wrong, because when routing to
5977 send a bounce to the sender, it would be empty. Therefore, I have changed
5978 it so that, while verifying a sender address, $sender_address is set to <>.
5979 (There is no change to what happens when verifying a recipient address.)
5981 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
5982 target A or AAAA records (if not already returned) without resetting the
5983 qualify_single or search_parents options of the DNS resolver. These are
5984 inappropriate in this case because the targets of MX and SRV records must
5985 be FQDNs. A broken DNS record could cause trouble if it happened to have a
5986 target that, when qualified, matched something in the local domain. These
5987 two options are now turned off when doing these lookups.
5989 57. It seems that at least some releases of Reiserfs (which does not have the
5990 concept of a fixed number of inodes) returns zero and not -1 for the
5991 number of available inodes. This interacted badly with check_spool_inodes,
5992 which assumed that -1 was the "no such thing" setting. What I have done is
5993 to check that the total number of inodes is greater than zero before doing
5994 the test of how many are available.
5996 58. When a "warn" ACL statement has a log_message modifier, the message is
5997 remembered, and not repeated. This is to avoid a lot of repetition when a
5998 message has many recipients that cause the same warning to be written.
5999 However, Exim was preserving the list of already written lines for an
6000 entire SMTP session, which doesn't seem right. The memory is now reset if a
6001 new message is started.
6003 59. The "rewrite" debugging flag was not showing the result of rewriting in the
6004 debugging output unless log_rewrite was also set.
6006 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
6007 of (int)(handle) when we know that handle contains (void *)(-1).
6009 61. The Exim daemon panic-logs an error return when it closes the incoming
6010 connection. However "connection reset by peer" seems to be common, and
6011 isn't really an error worthy of noting specially, so that particular error
6014 62. When Exim is trying to find all the local interfaces, it used to panic and
6015 die if the ioctl to get the interface flags failed. However, it seems that
6016 on at least one OS (Solaris 9) it is possible to have an interface that is
6017 included in the list of interfaces, but for which you get a failure error
6018 for this call. This happens when the interface is not "plumbed" into a
6019 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
6020 failure of the "get flags" call assumes that the interface is down.
6022 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
6023 makes life easier for people who are doing arithmetic on fields extracted
6024 from dates, where you often get leading zeros that should not be
6025 interpreted as octal.
6027 64. Added qualify_domain to the redirect router, to override the global
6030 65. If a pathologically long header line contained very many addresses (the
6031 report of this problem mentioned 10 000) and each of them was rewritten,
6032 Exim could use up a very large amount of memory. (It kept on making new
6033 copies of the header line as it rewrote, and never released the old ones.)
6034 At the expense of a bit more processing, the header rewriting function has
6035 been changed so that it no longer eats memory in this way.
6037 66. The generation of the Received: header has been moved from the time that a
6038 message starts to be received, to the time that it finishes. The timestamp
6039 in the Received: header should now be very close to that of the <= log
6040 line. There are two side-effects of this change:
6042 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
6043 logged header lines no longer include the local Received: line, because
6044 it has not yet been created. The same applies to a copy of the message
6045 that is returned to a non-SMTP sender when a message is rejected.
6047 (b) When a filter file is tested using -bf, no additional Received: header
6048 is added to the test message. After some thought, I decided that this
6051 This change does not affect the value of $received_for. It is still set
6052 after address rewriting, but before local_scan() is called.
6054 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
6056 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
6057 gave an unhelpful panic error message, and a defer error. I have managed to
6058 change this behaviour so that it now rejects any supplied certificate,
6059 which seems right, as the list of acceptable certificates is empty.
6061 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
6062 gave an unhelpful defer error. I have not managed to make this reject any
6063 supplied certificates, but the error message it gives is "no certificate
6064 supplied", which is not helpful.
6066 70. exigrep's output now also includes lines that are not associated with any
6067 message, but which match the given pattern. Implemented by a patch from
6068 Martin Sluka, which also tidied up the Perl a bit.
6070 71. Recipient callout verification, like sender verification, was using <> in
6071 the MAIL FROM command. This isn't really the right thing, since the actual
6072 sender may affect whether the remote host accepts the recipient or not. I
6073 have changed it to use the actual sender in the callout; this means that
6074 the cache record is now keyed on a recipient/sender pair, not just the
6075 recipient address. There doesn't seem to be a real danger of callout loops,
6076 since a callout by the remote host to check the sender would use <>.
6077 [SEE ABOVE: changed after hitting problems.]
6079 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
6080 temporary errors. However, in the case of such a code being given after
6081 the end of a data transmission (i.e. after ".") Exim was failing to write
6082 a retry record for the message. (Yes, there was some broken host that was
6083 actually sending 8xx at this point.)
6085 73. An unknown lookup type in a host list could cause Exim to panic-die when
6086 the list was checked. (An example that provoked this was putting <; in the
6087 middle of a list instead of at the start.) If this happened during a DATA
6088 ACL check, a -D file could be left lying around. This kind of configuration
6089 error no longer causes Exim to die; instead it causes a defer error. The
6090 incident is still logged to the main and panic logs.
6092 74. Buglet left over from Exim 3 conversion. The message "too many messages
6093 in one connection" was written to the rejectlog but not the mainlog, except
6094 when address rewriting (yes!) was being logged.
6096 75. Added write_rejectlog option.
6098 76. When a system filter was run not as root (that is, when system_filter_user
6099 was set), the values of the $n variables were not being returned to the
6100 main process; thus, they were not subsequently available in the $sn
6103 77. Added +return_path_on_delivery log selector.
6105 78. A connection timeout was being treated differently from recipients deferred
6106 when testing hosts_max_try with a message that was older than the host's
6107 retry timeout. (The host should not be counted, thus allowing all hosts to
6108 be tried at least once before bouncing.) This may have been the cause of an
6109 occasionally reported bug whereby a message would remain on the queue
6110 longer than the retry timeout, but would be bounced if a delivery was
6111 forced. I say "may" because I never totally pinned down the problem;
6112 setting up timeout/retry tests is difficult. See also the next item.
6114 79. The ultimate address timeout was not being applied to errors that involved
6115 a combination of host plus message (for example, a timeout on a MAIL
6116 command). When an address resolved to a number of possible hosts, and they
6117 were not all tried for each delivery (e.g. because of hosts_max_try), a
6118 message could remain on the queue longer than the retry timeout.
6120 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
6123 81. Fixed an obscure SMTP outgoing bug which required at least the following
6124 conditions: (a) there was another message waiting for the same server;
6125 (b) the server returned 5xx to all RCPT commands in the first message so
6126 that the message was not completed; (c) the server dropped the connection
6127 or gave a negative response to the RSET that Exim sends to abort the
6128 transaction. The observed case was a dropped connection after DATA that had
6129 been sent in pipelining mode. That is, the server had advertised PIPELINING
6130 but was not implementing it correctly. The effect of the bug was incorrect
6131 behaviour, such as trying another host, and this could lead to a crash.
6137 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
6138 and daemon.c were passed as pointers to ints; they should have been
6139 pointers to socklen_t variables (which are typically unsigned ints).
6141 2. Some signed/unsigned type warnings in the os.c file for Linux have been
6144 3. Fixed a really odd bug that affected only the testing scheme; patching a
6145 certain fixed string in the binary changed the value of another string that
6146 happened to be identical to the end of the original first string.
6148 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
6149 name", it returns that address as the IP address. On some operating
6150 systems (e.g. Solaris), it also passes back the IP address string as the
6151 "host name". However, on others (e.g. Linux), it passes back an empty
6152 string. Exim wasn't checking for this, and was changing the host name to an
6153 empty string, assuming it had been canonicalized.
6155 5. Although rare, it is permitted to have more than one PTR record for a given
6156 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
6157 all the names associated with an address, because they do in Solaris.
6158 However, it seems that they do not in Linux for data that comes from the
6159 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
6160 I found this out when I moved to a new Linux workstation and tried to run
6161 the Exim test suite.
6163 To get round this problem I have changed the code so that it now does its
6164 own call to the DNS to look up PTR records when searching for a host name.
6165 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
6166 addresses that are only in /etc/hosts are still found.
6168 This behaviour is, however, controlled by an option called host_lookup_
6169 order, which defaults to "bydns:byaddr". If people want to use the other
6170 order, or indeed, just use one or the other means of lookup, they can
6171 specify it in this variable.
6173 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
6174 some operating systems, this comes back from gethostbyaddr() as an empty
6175 string, and this is what Exim used to test for. However, it seems that in
6176 other systems, "." is yielded. Exim now tests for this case too.
6178 7. The values of check_spool_space and check_log_space are now held internally
6179 as a number of kilobytes instead of an absolute number of bytes. If a
6180 numbers is specified without 'K' or 'M', it is rounded up to the nearest
6181 kilobyte. This means that much larger values can be stored.
6183 8. Exim monitor: an attempt to get the action menu when not actually pointing
6184 at a message produces an empty menu entitled "No message selected". This
6185 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
6186 no entries in it ("Shell widget menu has zero width and/or height"). So I
6187 have added a single, blank menu entry in this case.
6189 9. Added ${quote_local_part.
6191 10. MIME decoding is now applied to the contents of Subject: header lines when
6194 11. Now that a reference to $sender_host_address automatically causes a reverse
6195 lookup to occur if necessary (4.13/18), there is no need to arrange for a
6196 host lookup before query-style lookups in lists that might use this
6197 variable. This has therefore been abolished, and the "net-" prefix is no
6198 longer necessary for query-style lookups.
6200 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
6201 have been a typo for LFLAGS, so it has been changed.
6203 13. The install script calls Exim with "-C /dev/null" in order to find the
6204 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
6205 to be output. However, since Exim outputs its version number before the
6206 error, it didn't break the script. It just looked ugly. I fixed this by
6207 always allowing "-C /dev/null" if the caller is root.
6209 14. Ignore overlarge ACL variable number when reading spool file - insurance
6210 against a later release with more variables having written the file.
6212 15. The standard form for an IPv6 address literal was being rejected by EHLO.
6213 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
6214 this, as well as the form without the "IPv6" on the front.
6216 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
6217 OS/Makefile-Darwin file.
6219 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
6220 only to LDAP libraries that do not have LDAP_OPT_DEREF.
6222 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
6223 variable. However, some OS use ints for inodes. I've added cast to long int
6224 to get rid of the compiler warning.
6226 19. I had forgotten to lock out "/../" in configuration file names when
6227 ALT_CONFIG_PREFIX was set.
6229 20. Routers used for verification do not need to specify transports. However,
6230 if such a router generated a host list, and callout was configured, Exim
6231 crashed, because it could not find a port number from the (non-existent)
6232 transport. It now assumes port 25 in this circumstance.
6234 21. Added the -t option to exigrep.
6236 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
6237 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
6238 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
6239 them is set, LOOKUP_LSEARCH is forced.
6241 23. "exim -bV" now outputs a list of lookups that are included in the binary.
6243 24. Added sender and host information to the "rejected by local_scan()" log
6244 line; previously there was no indication of these.
6246 25. Added .include_if_exists.
6248 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
6249 Linux. It turns out that not all file systems support this. Apparently some
6250 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
6251 do it.) To cope with this, the error EINVAL, which means that sync-ing is
6252 not supported on the file descriptor, is now ignored when Exim is trying to
6253 sync a directory. This applies only to Linux.
6255 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
6257 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
6258 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
6261 29. Added support for maildirsize files from supplied patch (modified a bit).
6263 30. The use of :fail: followed by an empty string could lead Exim to respond to
6264 sender verification failures with (e.g.):
6266 550 Verification failed for <xxx>
6267 550 Sender verify failed
6269 where the first response line was missing the '-' that indicates it is not
6270 the final line of the response.
6272 31. The loop for finding the name of the user that called Exim had a hardwired
6273 limit of 10; it now uses the value of finduser_retries, which is used for
6274 all other user lookups.
6276 32. Added $received_count variable, available in data and not_smtp ACLs, and at
6279 33. Exim was neglecting to zero errno before one call of strtol() when
6280 expanding a string and expecting an integer value. On some systems this
6281 resulted in spurious "integer overflow" errors. Also, it was casting the
6282 result into an int without checking.
6284 34. Testing for a connection timeout using "timeout_connect" in the retry rules
6285 did not work. The code looks as if it has *never* worked, though it appears
6286 to have been documented since at least release 1.62. I have made it work.
6288 35. The "timeout_DNS" error in retry rules, also documented since at least
6289 1.62, also never worked. As it isn't clear exactly what this means, and
6290 clearly it isn't a major issue, I have abolished the feature by treating it
6291 as "timeout", and writing a warning to the main and panic logs.
6293 36. The display of retry rules for -brt wasn't always showing the error code
6296 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
6297 timeout_connect_A, timeout_connect_MX.
6299 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
6300 to the empty sender.
6302 39. The daemon was not analysing the content of -oX till after it had closed
6303 stderr and disconnected from the controlling terminal. This meant that any
6304 syntax errors were only noted on the panic log, and the return code from
6305 the command was 0. By re-arranging the code a little, I've made the
6306 decoding happen first, so such errors now appear on stderr, and the return
6307 code is 1. However, the actual setting up of the sockets still happens in
6308 the disconnected process, so errors there are still only recorded on the
6311 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
6312 connections (as happens on some IP stacks) was logged at start up time as
6313 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
6314 it from "IPv6 and IPv4", which means that two separate sockets are being
6317 41. The debug output for gethostbyname2() or getipnodebyname() failures now
6318 says whether AF_INET or AF_INET6 was passed as an argument.
6320 42. Exiwhat output was messed up when time zones were included in log
6323 43. Exiwhat now gives more information about the daemon's listening ports,
6324 and whether -tls-on-connect was used.
6326 44. The "port" option of the smtp transport is now expanded.
6328 45. A "message" modifier in a "warn" statement in a non-message ACL was being
6329 silently ignored. Now an error message is written to the main and panic
6332 46. There's a new ACL modifier called "logwrite" which writes to a log file
6333 as soon as it is encountered.
6335 47. Added $local_user_uid and $local_user_gid at routing time.
6337 48. Exim crashed when trying to verify a sender address that was being
6340 49. Exim was recognizing only a space character after ".include". It now also
6341 recognizes a tab character.
6343 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
6344 extracting the relevant information from the specification. The man page no
6345 longer contains scrambled data for the -d option, and I've added a section
6346 at the front about calling Exim under different names.
6348 51. Added "extra_headers" argument to the "mail" command in filter files.
6350 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
6353 53. Installed eximstats 1.29.
6355 54. Added transport_filter_timeout as a generic transport option.
6357 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
6358 Cc: header lines. This was required by RFC 822, but it not required by RFC
6361 56. Exim used to add From:, Date:, and Message-Id: header lines to any
6362 incoming messages that did not have them. Now it does so only if the
6363 message originates locally, that is, if there is no associated remote host
6364 address. When Resent- header lines are present, this applies to the Resent-
6365 lines rather than the non-Resent- lines.
6367 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
6368 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
6370 58. Messages for configuration errors now include the name of the main
6371 configuration file - useful now that there may be more than one file in a
6372 list (.included file names were always shown).
6374 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
6375 for those rare installations that do not start the daemon as root or run it
6376 setuid root. I've cut out the call to initgroups() if the daemon is not
6379 60. The Exim user and group can now be bound into the binary as text strings
6380 that are looked up at the start of Exim's processing.
6382 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
6384 62. Added $mailstore_basename variable.
6386 63. Installed patch to sieve.c from Michael Haardt.
6388 64. When Exim failed to open the panic log after failing to open the main log,
6389 the original message it was trying to log was written to stderr and debug
6390 output, but if they were not available (the usual case in production), it
6391 was lost. Now it is written to syslog before the two lines that record the
6392 failures to open the logs.
6394 65. Users' Exim filters run in subprocesses under the user's uid. It is
6395 possible for a "deliver" command or an alias in a "personal" command to
6396 provoke an address rewrite. If logging of address rewriting is configured,
6397 this fails because the process is not running as root or exim. There may be
6398 a better way of dealing with this, but for the moment (because 4.30 needs
6399 to be released), I have disabled address rewrite logging when running a
6400 filter in a non-root, non-exim process.
6406 1. The buildconfig auxiliary program wasn't quoting the value set for
6407 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
6408 not defined. This bug was masked in 4.22 by the effect that was fixed in
6411 2. Some messages that were rejected after a message id was allocated were
6412 shown as "incomplete" by exigrep. It no longer does this for messages that
6413 are rejected by local_scan() or the DATA or non-SMTP ACLs.
6415 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
6416 have allow_domain_literals set, the ID did not get logged in the <= line.
6417 Domain literals are now always recognized in Message-ID: header lines.
6419 4. The first argument for a ${extract expansion item is the key name or field
6420 number. Leading and trailing spaces in this item were not being ignored,
6421 causing some misleading effects.
6423 5. When deliver_drop_privilege was set, single queue runner processes started
6424 manually (i.e. by the command "exim -q") or by the daemon (which uses the
6425 same command in the process it spins off) were not dropping privilege.
6427 6. When the daemon running as "exim" started a queue runner, it always
6428 re-executed Exim in the spun-off process. This is a waste of effort when
6429 deliver_drop_privilege is set. The new process now just calls the
6430 queue-runner function directly.
6436 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
6439 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
6440 ignored. Though the use of -oP was forcing the writing of a pid file, it
6441 was always written to the default place.
6443 3. If the message "no IP address found for host xxxx" is generated during
6444 incoming verification, it is now followed by identification of the incoming
6445 connection (so you can more easily find what provoked it).
6447 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
6449 5. Added some features to "harden" Exim a bit more against certain attacks:
6451 (a) There is now a build-time option called FIXED_NEVER_USERS that can
6452 be put in Local/Makefile. This is like the never_users runtime option,
6453 but it cannot be overridden. The default setting is "root".
6455 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
6456 prefix string with which any file named in a -C command line option
6459 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
6460 is retained for -C and -D only if the caller of Exim is root. Without
6461 it, the exim user may also use -C and -D and retain privilege.
6463 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
6464 command line option is disabled.
6466 6. Macro names set by the -D option must start with an upper case letter, just
6467 like macro names defined in the configuration file.
6469 7. Added "dereference=" facility to LDAP.
6471 8. Two instances of the typo "uknown" in the source files are fixed.
6473 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
6474 the Configure-Makefile script screwed up while processing it.
6476 10. Incorporated PCRE 4.4.
6478 11. The SMTP synchronization check was not operating right at the start of an
6479 SMTP session. For example, it could not catch a HELO sent before the client
6480 waited for the greeting. There is now a check for outstanding input at the
6481 point when the greeting is written. Because of the duplex, asynchronous
6482 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
6483 way, but not yet received, when the check is performed.
6485 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
6486 on TCP/IP sockets, because this apparently causes some broken clients to
6489 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
6490 unchanged) from the Cygwin maintainer.
6492 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
6493 was supported instead of testing for mbx. Effectively a typo.
6495 15. The spa authenticator server code was not checking that the input it
6496 received was valid base64.
6498 16. The debug output line for the "set" modifier in ACLs was not showing the
6499 name of the variable that was being set.
6501 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
6503 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
6504 lookup on its own. Something else was needed to trigger the lookup. For
6505 example, a match in host_lookup or the need for a host name in a host list.
6506 Now, if $sender_host_name is referenced and the host name has not yet been
6507 looked up, a lookup is performed. If the lookup fails, the variable remains
6508 empty, and $host_lookup_failed is set to "1".
6510 19. Added "eqi" as a case-independent comparison operator.
6512 20. The saslauthd authentication condition could segfault if neither service
6513 nor realm was specified.
6515 21. If an overflowing value such as "2048M" was set for message_size_limit, the
6516 error message that was logged was misleading, and incoming SMTP
6517 connections were dropped. The message is now more accurate, and temporary
6518 errors are given to SMTP connections.
6520 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
6521 (except RSET) with a 421 error, until QUIT is received. However, it was
6522 failing to send a response to QUIT.
6524 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
6525 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
6526 ACL is now run after the helo_try_verify_hosts code.
6528 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
6529 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
6530 case-independent, so other case variants are also recognized). Apparently
6531 some systems use these upper case variants.
6533 25. If more than two messages were waiting for the same host, and a transport
6534 filter was specified for the transport, Exim sent two messages over the
6535 same TCP/IP connection, and then failed with "socket operation on non-
6536 socket" when it tried to send the third.
6538 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
6540 27. The extern definition of crypt16() in expand.c was not being excluded when
6541 the OS had its own crypt16() function.
6543 28. Added bounce_return_body as a new option, and bounce_return_size_limit
6544 as a preferred synonym for return_size_limit, both as an option and as an
6547 29. Added LIBS=-liconv to OS/Makefile-OSF1.
6549 30. Changed the default configuration ACL to relax the local part checking rule
6550 for addresses that are not in any local domains. For these addresses,
6551 slashes and pipe symbols are allowed within local parts, but the sequence
6552 /../ is explicitly forbidden.
6554 31. SPA server authentication was not clearing the challenge buffer before
6557 32. log_message in a "warn" ACL statement was writing to the reject log as
6558 well as to the main log, which contradicts the documentation and doesn't
6559 seem right (because no rejection is happening). So I have stopped it.
6561 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
6562 However, I am unable to do any testing of this.
6564 34. Fixed an infelicity in the appendfile transport. When checking directories
6565 for a mailbox, to see if any needed to be created, it was accidentally
6566 using path names with one or more superfluous leading slashes; tracing
6567 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
6569 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
6570 contents are added to the log line that is written for every discarded
6571 recipient. (Previously a log_message setting was ignored.)
6573 36. The ${quote: operator now quotes the string if it is empty.
6575 37. The install script runs exim in order to find its version number. If for
6576 some reason other than non-existence or emptiness, which it checks, it
6577 could not run './exim', it was installing it with an empty version number,
6578 i.e. as "exim-". This error state is now caught, and the installation is
6581 38. An argument was missing from the function that creates an error message
6582 when Exim fails to connect to the socket for saslauthd authentication.
6583 This could cause Exim to crash, or give a corrupted message.
6585 39. Added isip, isip4, and isip6 to ${if conditions.
6587 40. The ACL variables $acl_xx are now saved with the message, and can be
6588 accessed later in routers, transports, and filters.
6590 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
6591 strings in the file are not string-expanded.
6593 42. If a MAIL command specified a SIZE value that was too large to fit into an
6594 int variable, the check against message_size_limit failed. Such values are
6595 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
6596 day this will have to be increased, but I don't think I want to be around
6597 when emails are that large.
6604 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
6605 iconv() is not standard in FreeBSD.
6607 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
6608 IPv6 enabled. The observed symptom was a segmentation fault on return from
6609 the function os_common_find_running_interfaces() in src/os.c.
6611 3. In the check_special_case() function in daemon.c I had used "errno" as an
6612 argument name, which causes warnings on some systems. This was basically a
6613 typo, since it was named "eno" in the comments!
6615 4. The code that waits for the clock to tick (at a resolution of some fraction
6616 of a second) so as to ensure message-id uniqueness was always waiting for
6617 at least one whole tick, when it could have waited for less. [This is
6618 almost certainly not relevant at current processor speeds, where it is
6619 unlikely to ever wait at all. But we try to future-proof.]
6621 5. The function that sleeps for a time interval that includes fractions of a
6622 second contained a race. It did not block SIGALRM between setting the
6623 timer, and suspending (a couple of lines later). If the interval was short
6624 and the sigsuspend() was delayed until after it had expired, the suspension
6625 never ended. On busy systems this could lead to processes getting stuck for
6628 6. Some uncommon configurations may cause a lookup to happen in a queue runner
6629 process, before it forks any delivery processes. The open lookup caching
6630 mechanism meant that the open file or database connection was passed into
6631 the delivery process. The problem was that delivery processes always tidy
6632 up cached lookup data. This could cause a problem for the next delivery
6633 process started by the queue runner, because the external queue runner
6634 process does not know about the closure. So the next delivery process
6635 still has data in the lookup cache. In the case of a file lookup, there was
6636 no problem because closing a file descriptor in a subprocess doesn't affect
6637 the parent. However, if the lookup was caching a connection to a database,
6638 the connection was closed, and the second delivery process was likely to
6639 see errors such as "PGSQL: query failed: server closed the connection
6640 unexpectedly". The problem has been fixed by closing all cached lookups
6641 in a queue runner before running a delivery process.
6643 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
6644 seem to have the "const" qualifier which it has on other OS. I've
6647 8. Change 4.21/2 was too strict. It is only if there are two authenticators
6648 *of the same type* (client or server) with the same public name that an
6649 error should be diagnosed.
6651 9. When Exim looked up a host name for an IP address, but failed to find the
6652 original IP address when looking up the host name (a safety check), it
6653 output the message "<ip address> does not match any IP for NULL", which was
6654 confusing, to say the least. The bug was that the host name should have
6655 appeared instead of "NULL".
6657 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
6658 user that is built into the binary, and the -C or -D options is used, root
6659 privilege is dropped before the configuration file is read. In addition,
6660 logging is switched to stderr instead of the normal log files. If the
6661 configuration then re-defines the Exim user, the unprivileged environment
6662 is probably not what is expected, so Exim logs a panic warning message (but
6665 However, if deliver_drop_privilege is set, the unprivileged state may well
6666 be exactly what is intended, so the warning has been cut out in that case,
6667 and Exim is allowed to try to write to its normal log files.
6673 1. smtp_return_error_details was not giving details for temporary sender
6674 or receiver verification errors.
6676 2. Diagnose a configuration error if two authenticators have the same public
6679 3. Exim used not to create the message log file for a message until the first
6680 delivery attempt. This could be confusing when incoming messages were held
6681 for policy or load reasons. The message log file is now created at the time
6682 the message is received, and an initial "Received" line is written to it.
6684 4. The automatically generated man page for command line options had a minor
6685 bug that caused no ill effects; however, a more serious problem was that
6686 the procedure for building the man page automatically didn't always
6687 operate. Consequently, release 4.20 contains an out-of-date version. This
6688 shouldn't happen again.
6690 5. When building Exim with embedded Perl support, the script that builds the
6691 Makefile was calling 'perl' to find its compile-time parameters, ignoring
6692 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
6694 6. The freeze_tell option was not being used for messages that were frozen on
6695 arrival, either by an ACL or by local_scan().
6697 7. Added the smtp_incomplete_transaction log selector.
6699 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
6700 was accepting AUTH without a new EHLO.
6702 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
6703 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
6706 10. Logging of TCP/IP connections (when configured) now happens in the main
6707 daemon process instead of the child process, so that the TCP/IP connection
6708 count is more accurate (but it can never be perfect).
6710 11. The use of "drop" in a nested ACL was not being handled correctly in the
6711 outer ACL. Now, if condition failure induced by the nested "drop" causes
6712 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
6713 or "require"), the connection is dropped.
6715 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
6716 that yield "discard" can now be used with an "accept" or a "discard" verb,
6717 but an error is generated for any others (because I can't see a useful way
6718 to define what should happen).
6720 13. When an ACL is read dynamically from a file (or anywhere else), the lines
6721 are now processed in the same way as lines in the Exim configuration file.
6722 In particular, continuation lines are supported.
6724 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
6726 15. Added -ti meaning -t -i.
6728 16. Check for letters, digits, hyphens, and dots in the names of dnslist
6729 domains, and warn by logging if others are found.
6731 17. At least on BSD, alignment is not guaranteed for the array of ifreq's
6732 returned from GIFCONF when Exim is trying to find the list of interfaces on
6733 a host. The code in os.c has been modified to copy each ifreq to an aligned
6734 structure in all cases.
6736 Also, in some cases, the returned ifreq's were being copied to a 'struct
6737 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
6738 means the last couple of bytes of an IPv6 address could be chopped if the
6739 ifreq contained only a normal sockaddr (14 bytes storage).
6741 18. Named domain lists were not supported in the hosts_treat_as_local option.
6742 An entry such as +xxxx was not recognized, and was treated as a literal
6745 19. Ensure that header lines added by a DATA ACL are included in the reject log
6746 if the ACL subsequently rejects the message.
6748 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
6749 MD5 (which is deprecated).
6751 21. When testing a filter file using -bf, Exim was writing a message when it
6752 took the sender from a "From " line in the message, but it was not doing so
6753 when it took $return_path from a Return-Path: header line. It now does.
6755 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
6756 with a valid header line field name (a series of printing characters
6757 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
6759 23. Changed "disc" in the source to "disk" to conform to the documentation and
6760 the book and for uniformity.
6762 24. Ignore Sendmail's -Ooption=value command line item.
6764 25. When execve() failed while trying to run a command in a pipe transport,
6765 Exim was returning EX_UNAVAILABLE (69) from the subprocess. However, this
6766 could be confused with a return value of 69 from the command itself. This
6767 has been changed to 127, the value the shell returns if it is asked to run
6768 a non-existent command. The wording for the related log line suggests a
6769 non-existent command as the problem.
6771 26. If received_header_text expands to an empty string, do not add a Received:
6772 header line to the message. (Well, it adds a token one on the spool, but
6773 marks it "old" so that it doesn't get used or transmitted.)
6775 27. Installed eximstats 1.28 (addition of -nt option).
6777 28. There was no check for failure on the call to getsockname() in the daemon
6778 code. This can fail if there is a shortage of resources on the system, with
6779 ENOMEM, for example. A temporary error is now given on failure.
6781 29. Contrary to the C standard, it seems that in some environments, the
6782 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
6783 program. Exim now does this explicitly; it affects the formatting of
6784 timestamps using strftime().
6786 30. If exiqsumm was given junk data, it threw up some uninitialized variable
6787 complaints. I've now initialized all the variables, to avoid this.
6789 32. Header lines added by a system filter were not being "seen" during
6790 transport-time rewrites.
6792 33. The info_callback() function passed to OpenSSL is set up with type void
6793 (*)(SSL *, int, int), as described somewhere. However, when calling the
6794 function (actually a macro) that sets it up, the type void(*)() is
6795 expected. I've put in a cast to prevent warnings from picky compilers.
6797 34. If a DNS black list lookup found a CNAME record, but there were no A
6798 records associated with the domain it pointed at, Exim crashed.
6800 35. If a DNS black list lookup returned more than one A record, Exim ignored
6801 all but the first. It now scans all returned addresses if a particular IP
6802 value is being sought. In this situation, the contents of the
6803 $dnslist_value variable are a list of all the addresses, separated by a
6806 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
6807 to accept a host name and all its aliases if the forward lookup for any of
6808 them yielded the IP address of the incoming connection. Now it accepts only
6809 those names whose forward lookup yields the correct IP address. Any other
6810 names are discarded. This closes a loophole whereby a rogue DNS
6811 administrator could create reverse DNS records to break through a
6812 wildcarded host restriction in an ACL.
6814 37. If a user filter or a system filter that ran in a subprocess used any of
6815 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
6816 the wrong values were passed to the pipe command ($thisaddress had the
6817 value of $0, $0 had the value of $1, etc). This bug was introduced by
6818 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
6820 38. Improved the line breaking for long SMTP error messages from ACLs.
6821 Previously, if there was no break point between 40 and 75 characters, Exim
6822 left the rest of the message alone. Two changes have been made: (a) I've
6823 reduced the minimum length to 35 characters; (b) if it can't find a break
6824 point between 35 and 75 characters, it looks ahead and uses the first one
6825 that it finds. This may give the occasional overlong line, but at least the
6826 remaining text gets split now.
6828 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
6829 file descriptors might be low, and that setting 1000 would always raise it.
6830 It turns out that in some environments, the limit is already over 1000 and
6831 that lowering it causes trouble. So now Exim takes care not to decrease it.
6833 40. When delivering a message, the value of $return_path is set to $sender_
6834 address at the start of routing (routers may change the value). By an
6835 oversight, this default was not being set up when an address was tested by
6836 -bt or -bv, which affected the outcome if any router or filter referred to
6839 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
6840 log only when "message" or "log_message" are set. However, if one of the
6841 conditions was an address verification, or a call to a nested ACL, the
6842 messages generated by the underlying test were being passed through. This
6843 no longer happens. The underlying message is available in $acl_verify_
6844 message for both "message" and "log_message" expansions, so it can be
6845 passed through if needed.
6847 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
6848 new expansion $bh_ to give the encoded byte string without charset
6849 translation. Translation happens only if iconv() is available; HAVE_ICONV
6850 indicates this at build time. HEADERS_CHARSET gives the charset to
6851 translate to; headers_charset can change it in the configuration, and
6852 "headers charset" can change it in an individual filter file.
6854 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
6855 that creates RFC 2047 encoded "words" labels them as that charset instead
6856 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
6857 expansion operator; (ii) when Exim creates a From: line for a local
6858 message; (iii) when a header line is rewritten to include a "phrase" part.
6860 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
6861 buggy, causing it to skip the first lines of messages whose message ID
6862 ended in 'D'. This would not have bitten before Exim release 4.14, because
6863 message IDs were unlikely to end in 'D' before then. The effect was to have
6864 incorrect size information for certain domains.
6866 45. #include "config.h" was missing at the start of the crypt16.c module. This
6867 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
6870 46. If there was a timeout during a "random" callout check, Exim treated it as
6871 a failure of the random address, and carried on sending RSET and the real
6872 address. If the delay was just some slowness somewhere, the response to the
6873 original RCPT would be taken as a response to RSET and so on, causing
6874 mayhem of various kinds.
6876 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
6877 when I implemented it. It didn't allow for the fact that some option values
6878 may legitimately be negative (e.g. size_addition), and it didn't even do
6879 the right test for positive values.
6881 48. Domain names in DNS records are case-independent. Exim always looks them up
6882 in lower case. Some resolvers return domain names in exactly the case they
6883 appear in the zone file, that is, they may contain uppercase letters. Not
6884 all resolvers do this - some return always lower case. Exim was treating a
6885 change of case by a resolver as a change of domain, similar to a widening
6886 of a domain abbreviation. This triggered its re-routing code and so it was
6887 trying to route what was effectively the same domain again. This normally
6888 caused routing to fail (because the router wouldn't handle the domain
6889 twice). Now Exim checks for this case specially, and just changes the
6890 casing of the domain that it ultimately uses when it transmits the message
6893 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
6896 50. If a filter generated a file delivery with a non-absolute name (possible if
6897 no home directory exists for the router), the forbid_file option was not
6900 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
6901 the existing equality matching.
6903 52. Exim was using ints instead of ino_t variables in some places where it was
6904 dealing with inode numbers.
6906 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
6907 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
6908 called TMPDIR, and if it finds it is different, it changes its value.
6910 54. The smtp_printf() function is now made available to local_scan() so
6911 additional output lines can be written before returning. There is also an
6912 smtp_fflush() function to enable the detection of a dropped connection.
6913 The variables smtp_input and smtp_batched_input are exported to
6916 55. Changed the default runtime configuration: the message "Unknown user"
6917 has been removed from the ACL, and instead placed on the localuser router,
6918 using the cannot_route_message feature. This means that any verification
6919 failures that generate their own messages won't get overridden. Similarly,
6920 the "Unrouteable address" message that was in the ACL for unverifiable
6921 relay addresses has also been removed.
6923 56. Added hosts_avoid_esmtp to the smtp transport.
6925 57. The exicyclog script was not checking for the esoteric option
6926 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
6927 will work only if exicyclog is run under the appropriate euid.
6929 58. Following a discussion on the list, the rules by which Exim recognises line
6930 endings on incoming messages have been changed. The -dropcr and drop_cr
6931 options are now no-ops, retained only for backwards compatibility. The
6932 following line terminators are recognized: LF CRLF CR. However, special
6933 processing applies to CR:
6935 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
6936 nor a local message in the state where . is a terminator.
6938 (ii) If a bare CR is encountered in a header line, an extra space is added
6939 after the line terminator so as not to end the header. The reasoning
6940 behind this is that bare CRs in header lines are most likely either
6941 to be mistakes, or people trying to play silly games.
6943 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
6944 was being incorrectly given as 18 bytes larger than it should have been.
6945 This is a VOB (very old bug).
6947 60. This may never have affected anything current, but just in case it has:
6948 When the local host is found other than at the start of a list of hosts,
6949 the local host, those with the same MX, and any that follow, are discarded.
6950 When the list in question was part of a longer list of hosts, the following
6951 hosts (not currently being processed) were also being discarded. This no
6952 longer happens. I'm not sure if this situation could ever has previously
6955 61. Added the "/MX" feature to lists of hosts in the manualroute and query
6958 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
6959 header. This is something that is recommended in a new Internet Draft, and
6960 is something that is documented as being done by Sendmail. There are two
6961 possible values. For messages generated by the autoreply transport, Exim
6964 Auto-Submitted: auto-replied
6966 whereas for all other generated messages (e.g. bounces) it adds
6968 Auto-Submitted: auto-generated
6970 63. The "personal" condition in filters now includes a test for the
6971 Auto-Submitted: header. If it contains the string "auto-" the message it
6972 not considered personal.
6974 64. Added rcpt_include_affixes as a generic transport option.
6976 65. Added queue_only_override (default true).
6978 66. Added the syslog_duplication option.
6980 67. If what should have been the first header line of a message consisted of
6981 a space followed by a colon, Exim was mis-interpreting it as a header line.
6982 It isn't of course - it is syntactically invalid and should therefore be
6983 treated as the start of the message body. The misbehaviour could have
6984 caused a number of strange effects, including loss of data in subsequent
6985 header lines, and spool format errors.
6987 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
6988 client host had authenticated. This control can now be exercised by an ACL
6989 for more flexibility.
6991 69. By default, callouts do not happen when testing with -bh. There is now a
6992 variant, -bhc, which does actually run the callout code, including
6993 consulting and updating the callout cache.
6995 70. Added support for saslauthd authentication, courtesy of Alexander
6998 71. If statvfs() failed on the spool or log directories while checking their
6999 size for availability, Exim confusingly gave the error "space shortage".
7000 Furthermore, in debugging mode it crashed with a floating point exception.
7001 These checks are done if check_{spool,log}_{space,inodes} are set, and when
7002 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
7003 serious problem, Exim now writes to the main and panic logs when this
7004 happens, with details of the failure. It then refuses to accept the
7005 incoming message, giving the message "spool directory problem" or "log
7006 directory problem" with a 421 code for SMTP messages.
7008 72. When Exim is about to re-exec itself, it ensures that the file descriptors
7009 0, 1, and 2 exist, because some OS complain for execs without them (see
7010 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
7011 descriptors. However, the code omitted to check that the open succeeded,
7012 causing mysterious errors if for some reason the permissions on /dev/null
7013 got screwed. Now Exim writes a message to the main and panic logs, and
7014 bombs out if it can't open /dev/null.
7016 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
7017 interact so that it is all more flexible. It is supposed to remain
7018 backwards compatible. Also added extra_local_interfaces.
7020 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
7021 to bomb out with an assertion failure - to the client this appears as a
7022 connection drop. This problem occurs in the part of the code that was taken
7023 from the Samba project. Fortunately, the assertion is in a very simple
7024 function, so I have fixed this by reproducing the function inline in the
7025 one place where it is called, and arranging for authentication to fail
7026 instead of killing the process with assert().
7028 75. The SPA client code was not working when the server requested OEM rather
7029 than Unicode encoding.
7031 76. Added code to make require_files with a specific uid setting more usable in
7032 the case where statting the file as root fails - usually a non-root-mounted
7033 NFS file system. When this happens and the failure is EACCES, Exim now
7034 forks a subprocess and does the per-uid checking as the relevant uid.
7036 77. Added process_log_path.
7038 78. If log_file_path was not explicitly set, a setting of check_log_space or
7039 check_log_inodes was ignored.
7041 79. If a space check for the spool or log partitions fails, the incident is now
7042 logged. Of course, in the latter case the data may get lost...
7044 80. Added the %p formatting code to string_format() so that it can be used to
7045 print addresses in debug_print(). Adjusted all the address printing in the
7046 debugging in store.c to use %p rather than %d.
7048 81. There was a concern that a line of code in smtp_in.c could overflow a
7049 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
7050 initially expressed, the concern was not well-founded, because trailing
7051 spaces are removed early. However, if the trailing spaces were followed by
7052 a NULL, they did not get removed, so the overflow was possible. Two fixes
7055 (a) I re-wrote the offending code in a cleaner fashion.
7056 (b) If an incoming SMTP command contains a NULL character, it is rejected
7059 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
7060 runs initgroups(), so that if the Exim user is in any additional groups,
7061 they will be used during message reception.
7067 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / exim.git / blob