1 Change log file for Exim from version 4.21
2 ------------------------------------------
3 This document describes *changes* to previous versions, that might
4 affect Exim's operation, with an unchanged configuration file. For new
5 options, and new features, see the NewStuff file next to this ChangeLog.
8 Since Exim version 4.90
9 -----------------------
11 JH/01 Replace the store_release() internal interface with store_newblock(),
12 which internalises the check required to safely use the old one, plus
13 the allocate and data copy operations duplicated in both (!) of the
16 JH/02 Disallow '/' characters in queue names specified for the "queue=" ACL
17 modifier. This matches the restriction on the commandline.
19 JH/03 Fix pgsql lookup for multiple result-tuples with a single column.
20 Previously only the last row was returned.
22 JH/04 Bug 2217: Tighten up the parsing of DKIM signature headers. Previously
23 we assumed that tags in the header were well-formed, and parsed the
24 element content after inspecting only the first char of the tag.
25 Assumptions at that stage could crash the receive process on malformed
28 JH/05 Bug 2215: Fix crash associated with dnsdb lookup done from DKIM ACL.
29 While running the DKIM ACL we operate on the Permanent memory pool so that
30 variables created with "set" persist to the DATA ACL. Also (at any time)
31 DNS lookups that fail create cache records using the Permanent pool. But
32 expansions release any allocations made on the current pool - so a dnsdb
33 lookup expansion done in the DKIM ACL releases the memory used for the
34 DNS negative-cache, and bad things result. Solution is to switch to the
35 Main pool for expansions.
36 While we're in that code, add checks on the DNS cache during store_reset,
37 active in the testsuite.
38 Problem spotted, and debugging aided, by Wolfgang Breyha.
40 JH/06 Fix issue with continued-connections when the DNS shifts unreliably.
41 When none of the hosts presented to a transport match an already-open
42 connection, close it and proceed with the list. Previously we would
43 queue the message. Spotted by Lena with Yahoo, probably involving
46 JH/07 Bug 2214: Fix SMTP responses resulting from non-accept result of MIME ACL.
47 Previously a spurious "250 OK id=" response was appended to the proper
50 JH/08 The "support for" informational output now, which built with Content
51 Scanning support, has a line for the malware scanner interfaces compiled
52 in. Interface can be individually included or not at build time.
54 JH/09 The "aveserver", "kavdaemon" and "mksd" interfaces are now not included
55 by the template makefile "src/EDITME". The "STREAM" support for an older
56 ClamAV interface method is removed.
58 JH/10 Bug 2223: Fix mysql lookup returns for the no-data case (when the number of
59 rows affected is given instead).
61 JH/11 The runtime Berkeley DB library version is now additionally output by
62 "exim -d -bV". Previously only the compile-time version was shown.
64 JH/12 Bug 2230: Fix cutthrough routing for nonfirst messages in an initiating
65 SMTP connection. Previously, when one had more receipients than the
66 first, an abortive onward connection was made. Move to full support for
67 multiple onward connections in sequence, handling cutthrough connection
68 for all multi-message initiating connections.
70 JH/13 Bug 2229: Fix cutthrough routing for nonstandard port numbers defined by
71 routers. Previously, a multi-recipient message would fail to match the
72 onward-connection opened for the first recipient, and cause its closure.
74 JH/14 Bug 2174: A timeout on connect for a callout was also erroneously seen as
75 a timeout on read on a GnuTLS initiating connection, resulting in the
76 initiating connection being dropped. This mattered most when the callout
77 was marked defer_ok. Fix to keep the two timeout-detection methods
80 JH/15 Relax results from ACL control request to enable cutthrough, in
81 unsupported situations, from error to silently (except under debug)
82 ignoring. This covers use with PRDR, frozen messages, queue-only and
85 HS/01 Fix Buffer overflow in base64d() (CVE-2018-6789)
87 JH/16 Fix bug in DKIM verify: a buffer overflow could corrupt the malloc
88 metadata, resulting in a crash in free().
90 PP/01 Fix broken Heimdal GSSAPI authenticator integration.
91 Broken in f2ed27cf5, missing an equals sign for specified-initialisers.
92 Broken also in d185889f4, with init system revamp.
94 JH/17 Bug 2113: Fix conversation closedown with the Avast malware scanner.
95 Previously we abruptly closed the connection after reading a malware-
96 found indication; now we go on to read the "scan ok" response line,
99 JH/18 Bug 2239: Enforce non-usability of control=utf8_downconvert in the mail
100 ACL. Previously, a crash would result.
102 JH/19 Speed up macro lookups during configuration file read, by skipping non-
103 macro text after a replacement (previously it was only once per line) and
104 by skipping builtin macros when searching for an uppercase lead character.
106 JH/20 DANE support moved from Experimental to mainline. The Makefile control
107 for the build is renamed.
109 JH/21 Fix memory leak during multi-message connections using STARTTLS. A buffer
110 was allocated for every new TLS startup, meaning one per message. Fix
111 by only allocating once (OpenSSL) or freeing on TLS-close (GnuTLS).
113 JH/22 Bug 2236: When a DKIM verification result is overridden by ACL, DMARC
114 reported the original. Fix to report (as far as possible) the ACL
115 result replacing the original.
117 JH/23 Fix memory leak during multi-message connections using STARTTLS under
118 OpenSSL. Certificate information is loaded for every new TLS startup,
119 and the resources needed to be freed.
121 JH/24 Bug 2242: Fix exim_dbmbuild to permit directoryless filenames.
123 JH/25 Fix utf8_downconvert propagation through a redirect router. Previously it
130 JH/01 Rework error string handling in TLS interface so that the caller in
131 more cases is responsible for logging. This permits library-sourced
132 string to be attached to addresses during delivery, and collapses
133 pairs of long lines into single ones.
135 PP/01 Allow PKG_CONFIG_PATH to be set in Local/Makefile and use it correctly
136 during configuration. Wildcards are allowed and expanded.
138 JH/02 Rework error string handling in DKIM to pass more info back to callers.
139 This permits better logging.
141 JH/03 Rework the transport continued-connection mechanism: when TLS is active,
142 do not close it down and have the child transport start it up again on
143 the passed-on TCP connection. Instead, proxy the child (and any
144 subsequent ones) for TLS via a unix-domain socket channel. Logging is
145 affected: the continued delivery log lines do not have any DNSSEC, TLS
146 Certificate or OCSP information. TLS cipher information is still logged.
148 JH/04 Shorten the log line for daemon startup by collapsing adjacent sets of
149 identical IP addresses on different listening ports. Will also affect
152 PP/02 Bug 2070: uClibc defines __GLIBC__ without providing glibc headers;
153 add noisy ifdef guards to special-case this sillyness.
154 Patch from Bernd Kuhls.
156 JH/05 Tighten up the checking in isip4 (et al): dotted-quad components larger
157 than 255 are no longer allowed.
159 JH/06 Default openssl_options to include +no_ticket, to reduce load on peers.
160 Disable the session-cache too, which might reduce our load. Since we
161 currrectly use a new context for every connection, both as server and
162 client, there is no benefit for these.
163 GnuTLS appears to not support tickets server-side by default (we don't
164 call gnutls_session_ticket_enable_server()) but client side is enabled
165 by default on recent versions (3.1.3 +) unless the PFS priority string
168 PP/03 Add $SOURCE_DATE_EPOCH support for reproducible builds, per spec at
169 <https://reproducible-builds.org/specs/source-date-epoch/>.
171 JH/07 Fix smtp transport use of limited max_rcpt under mua_wrapper. Previously
172 the check for any unsuccessful recipients did not notice the limit, and
173 erroneously found still-pending ones.
175 JH/08 Pipeline CHUNKING command and data together, on kernels that support
176 MSG_MORE. Only in-clear (not on TLS connections).
178 JH/09 Avoid using a temporary file during transport using dkim. Unless a
179 transport-filter is involved we can buffer the headers in memory for
180 creating the signature, and read the spool data file once for the
181 signature and again for transmission.
183 JH/10 Enable use of sendfile in Linux builds as default. It was disabled in
184 4.77 as the kernel support then wasn't solid, having issues in 64bit
185 mode. Now, it's been long enough. Add support for FreeBSD also.
187 JH/11 Bug 2104: Fix continued use of a transport connection with TLS. In the
188 case where the routing stage had gathered several addresses to send to
189 a host before calling the transport for the first, we previously failed
190 to close down TLS in the old transport process before passing the TCP
191 connection to the new process. The new one sent a STARTTLS command
192 which naturally failed, giving a failed delivery and bloating the retry
193 database. Investigation and fix prototype from Wolfgang Breyha.
195 JH/12 Fix check on SMTP command input synchronisation. Previously there were
196 false-negatives in the check that the sender had not preempted a response
197 or prompt from Exim (running as a server), due to that code's lack of
198 awareness of the SMTP input buffering.
200 PP/04 Add commandline_checks_require_admin option.
201 Exim drops privileges sanely, various checks such as -be aren't a
202 security problem, as long as you trust local users with access to their
203 own account. When invoked by services which pass untrusted data to
204 Exim, this might be an issue. Set this option in main configuration
205 AND make fixes to the calling application, such as using `--` to stop
208 JH/13 Do pipelining under TLS. Previously, although safe, no advantage was
209 taken. Now take care to pack both (client) MAIL,RCPT,DATA, and (server)
210 responses to those, into a single TLS record each way (this usually means
211 a single packet). As a side issue, smtp_enforce_sync now works on TLS
214 PP/05 OpenSSL/1.1: use DH_bits() for more accurate DH param sizes. This
215 affects you only if you're dancing at the edge of the param size limits.
216 If you are, and this message makes sense to you, then: raise the
217 configured limit or use OpenSSL 1.1. Nothing we can do for older
220 JH/14 For the "sock" variant of the malware scanner interface, accept an empty
221 cmdline element to get the documented default one. Previously it was
224 JH/15 Fix a crash in the smtp transport caused when two hosts in succession
225 are unsuable for non-message-specific reasons - eg. connection timeout,
226 banner-time rejection.
228 JH/16 Fix logging of delivery remote port, when specified by router, under
231 PP/06 Repair manualroute's ability to take options in any order, even if one
232 is the name of a transport.
235 HS/01 Cleanup, prevent repeated use of -p/-oMr (CVE-2017-1000369)
237 JH/17 Change the list-building routines interface to use the expanding-string
238 triplet model, for better allocation and copying behaviour.
240 JH/18 Prebuild the data-structure for "builtin" macros, for faster startup.
241 Previously it was constructed the first time a possibly-matching string
242 was met in the configuration file input during startup; now it is done
245 JH/19 Bug 2141: Use the full-complex API for Berkeley DB rather than the legacy-
246 compatible one, to avoid the (poorly documented) possibility of a config
247 file in the working directory redirecting the DB files, possibly correpting
248 some existing file. CVE-2017-10140 assigned for BDB.
250 JH/20 Bug 2147: Do not defer for a verify-with-callout-and-random which is not
251 cache-hot. Previously, although the result was properly cached, the
252 initial verify call returned a defer.
254 JH/21 Bug 2151: Avoid using SIZE on the MAIL for a callout verify, on any but
255 the main verify for receipient in uncached-mode.
257 JH/22 Retire historical build files to an "unsupported" subdir. These are
258 defined as "ones for which we have no current evidence of testing".
260 JH/23 DKIM: enforce the DNS pubkey record "h" permitted-hashes optional field,
261 if present. Previously it was ignored.
263 JH/24 Start using specified-initialisers in C structure init coding. This is
264 a C99 feature (it's 2017, so now considered safe).
266 JH/25 Use one-bit bitfields for flags in the "addr" data structure. Previously
267 if was a fixed-sized field and bitmask ops via macros; it is now more
270 PP/07 GitHub PR 56: Apply MariaDB build fix.
271 Patch provided by Jaroslav Škarvada.
273 PP/08 Bug 2161: Fix regression in sieve quoted-printable handling introduced
274 during Coverity cleanups [4.87 JH/47]
275 Diagnosis and fix provided by Michael Fischer v. Mollard.
277 JH/26 Fix DKIM bug: when the pseudoheader generated for signing was exactly
278 the right size to place the terminating semicolon on its own folded
279 line, the header hash was calculated to an incorrect value thanks to
280 the (relaxed) space the fold became.
282 HS/02 Fix Bug 2130: large writes from the transport subprocess where chunked
283 and confused the parent.
285 JH/27 Fix SOCKS bug: an unitialized pointer was deref'd by the transport process
286 which could crash as a result. This could lead to undeliverable messages.
288 JH/28 Logging: "next input sent too soon" now shows where input was truncated
291 JH/29 Fix queue_run_in_order to ignore the PID portion of the message ID. This
292 matters on fast-turnover and PID-randomising systems, which were getting
293 out-of-order delivery.
295 JH/30 Fix a logging bug on aarch64: an unsafe routine was previously used for
296 a possibly-overlapping copy. The symptom was that "Remote host closed
297 connection in response to HELO" was logged instead of the actual 4xx
300 JH/31 Fix CHUNKING code to properly flush the unwanted chunk after an error.
301 Previously only that bufferd was discarded, resulting in SYMTP command
304 JH/32 DKIM: when a message has multiple signatures matching an identity given
305 in dkim_verify_signers, run the dkim acl once for each. Previously only
306 one run was done. Bug 2189.
308 JH/33 Downgrade an unfound-list name (usually a typo in the config file) from
309 "panic the current process" to "deliberately defer". The panic log is
310 still written with the problem list name; the mail and reject logs now
311 get a temp-reject line for the message that was being handled, saying
312 something like "domains check lookup or other defer". The SMTP 451
313 message is still "Temporary local problem".
315 JH/34 Bug 2199: Fix a use-after-free while reading smtp input for header lines.
316 A crafted sequence of BDAT commands could result in in-use memory beeing
317 freed. CVE-2017-16943.
319 HS/03 Bug 2201: Fix checking for leading-dot on a line during headers reading
320 from SMTP input. Previously it was always done; now only done for DATA
321 and not BDAT commands. CVE-2017-16944.
323 JH/35 Bug 2201: Flush received data in BDAT mode after detecting an error fatal
324 to the message (such as an overlong header line). Previously this was
325 not done and we did not exit BDAT mode. Followon from the previous item
326 though a different problem.
332 JH/01 Bug 1922: Support IDNA2008. This has slightly different conversion rules
333 than -2003 did; needs libidn2 in addition to libidn.
335 JH/02 The path option on a pipe transport is now expanded before use.
337 PP/01 GitHub PR 50: Do not call ldap_start_tls_s on ldapi:// connections.
338 Patch provided by "Björn", documentation fix added too.
340 JH/03 Bug 2003: fix Proxy Protocol v2 handling: the address size field was
341 missing a wire-to-host endian conversion.
343 JH/04 Bug 2004: fix CHUNKING in non-PIPELINEING mode. Chunk data following
344 close after a BDAT command line could be taken as a following command,
345 giving a synch failure. Fix by only checking for synch immediately
346 before acknowledging the chunk.
348 PP/02 GitHub PR 52: many spelling fixes, which include fixing parsing of
349 no_require_dnssec option and creation of _HAVE_TRANSPORT_APPEND_MAILDIR
350 macro. Patches provided by Josh Soref.
352 JH/05 Have the EHLO response advertise VRFY, if there is a vrfy ACL defined.
353 Previously we did not; the RFC seems ambiguous and VRFY is not listed
354 by IANA as a service extension. However, John Klensin suggests that we
357 JH/06 Bug 2017: Fix DKIM verification in -bh test mode. The data feed into
358 the dkim code may be unix-mode line endings rather than smtp wire-format
359 CRLF, so prepend a CR to any bare LF.
361 JH/07 Rationalise the coding for callout smtp conversations and transport ones.
362 As a side-benfit, callouts can now use PIPELINING hence fewer round-trips.
364 JH/08 Bug 2016: Fix DKIM verification vs. CHUNKING. Any BDAT commands after
365 the first were themselves being wrongly included in the feed into dkim
366 processing; with most chunk sizes in use this resulted in an incorrect
367 body hash calculated value.
369 JH/09 Bug 2014: permit inclusion of a DKIM-Signature header in a received
370 DKIM signature block, for verification. Although advised against by
371 standards it is specifically not ruled illegal.
373 JH/10 Bug 2025: Fix reception of (quoted) local-parts with embedded spaces.
375 JH/11 Bug 2029: Fix crash in DKIM verification when a message signature block is
376 missing a body hash (the bh= tag).
378 JH/12 Bug 2018: Re-order Proxy Protocol startup versus TLS-on-connect startup.
379 It seems that HAProxy sends the Proxy Protocol information in clear and
380 only then does a TLS startup, so do the same.
382 JH/13 Bug 2027: Avoid attempting to use TCP Fast Open for non-transport client
383 TCP connections (such as for Spamd) unless the daemon successfully set
384 Fast Open mode on its listening sockets. This fixes breakage seen on
385 too-old kernels or those not configured for Fast Open, at the cost of
386 requiring both directions being enabled for TFO, and TFO never being used
387 by non-daemon-related Exim processes.
389 JH/14 Bug 2000: Reject messages recieved with CHUNKING but with malformed line
390 endings, at least on the first header line. Try to canonify any that get
391 past that check, despite the cost.
393 JH/15 Angle-bracket nesting (an error inserted by broken sendmails) levels are
394 now limited to an arbitrary five deep, while parsing addresses with the
395 strip_excess_angle_brackets option enabled.
397 PP/03 Bug 2018: For Proxy Protocol and TLS-on-connect, do not over-read and
398 instead leave the unprompted TLS handshake in socket buffer for the
399 TLS library to consume.
401 PP/04 Bug 2018: Also handle Proxy Protocol v2 safely.
403 PP/05 FreeBSD compat: handle that Ports no longer create /usr/bin/perl
405 JH/16 Drop variables when they go out of scope. Memory management drops a whole
406 region in one operation, for speed, and this leaves assigned pointers
407 dangling. Add checks run only under the testsuite which checks all
408 variables at a store-reset and panics on a dangling pointer; add code
409 explicitly nulling out all the variables discovered. Fixes one known
410 bug: a transport crash, where a dangling pointer for $sending_ip_address
411 originally assigned in a verify callout, is re-used.
413 PP/06 Drop '.' from @INC in various Perl scripts.
415 PP/07 Switch FreeBSD iconv to always use the base-system libc functions.
417 PP/08 Reduce a number of compilation warnings under clang; building with
418 CC=clang CFLAGS+=-Wno-dangling-else -Wno-logical-op-parentheses
419 should be warning-free.
421 JH/17 Fix inbound CHUNKING when DKIM disabled at runtime.
423 HS/01 Fix portability problems introduced by PP/08 for platforms where
424 realloc(NULL) is not equivalent to malloc() [SunOS et al].
426 HS/02 Bug 1974: Fix missing line terminator on the last received BDAT
427 chunk. This allows us to accept broken chunked messages. We need a more
428 general solution here.
430 PP/09 Wrote util/chunking_fixqueue_finalnewlines.pl to help recover
431 already-broken messages in the queue.
433 JH/18 Bug 2061: Fix ${extract } corrupting an enclosing ${reduce } $value.
435 JH/19 Fix reference counting bug in routing-generated-address tracking.
441 JH/01 Use SIZE on MAIL FROM in a cutthrough connection, if the destination
442 supports it and a size is available (ie. the sending peer gave us one).
444 JH/02 The obsolete acl condition "demime" is removed (finally, after ten
445 years of being deprecated). The replacements are the ACLs
446 acl_smtp_mime and acl_not_smtp_mime.
448 JH/03 Upgrade security requirements imposed for hosts_try_dane: previously
449 a downgraded non-dane trust-anchor for the TLS connection (CA-style)
450 or even an in-clear connection were permitted. Now, if the host lookup
451 was dnssec and dane was requested then the host is only used if the
452 TLSA lookup succeeds and is dnssec. Further hosts (eg. lower priority
453 MXs) will be tried (for hosts_try_dane though not for hosts_require_dane)
454 if one fails this test.
455 This means that a poorly-configured remote DNS will make it incommunicado;
456 but it protects against a DNS-interception attack on it.
458 JH/04 Bug 1810: make continued-use of an open smtp transport connection
459 non-noisy when a race steals the message being considered.
461 JH/05 If main configuration option tls_certificate is unset, generate a
462 self-signed certificate for inbound TLS connections.
464 JH/06 Bug 165: hide more cases of password exposure - this time in expansions
465 in rewrites and routers.
467 JH/07 Retire gnutls_require_mac et.al. These were nonfunctional since 4.80
468 and logged a warning sing 4.83; now they are a configuration file error.
470 JH/08 Bug 1836: Fix crash in VRFY handling when handed an unqualified name
471 (lacking @domain). Apply the same qualification processing as RCPT.
473 JH/09 Bug 1804: Avoid writing msglog files when in -bh or -bhc mode.
475 JH/10 Support ${sha256:} applied to a string (as well as the previous
478 JH/11 Cutthrough: avoid using the callout hints db on a verify callout when
479 a cutthrough deliver is pending, as we always want to make a connection.
480 This also avoids re-routing the message when later placing the cutthrough
481 connection after a verify cache hit.
482 Do not update it with the verify result either.
484 JH/12 Cutthrough: disable when verify option success_on_redirect is used, and
485 when routing results in more than one destination address.
487 JH/13 Cutthrough: expand transport dkim_domain option when testing for dkim
488 signing (which inhibits the cutthrough capability). Previously only
489 the presence of an option was tested; now an expansion evaluating as
490 empty is permissible (obviously it should depend only on data available
491 when the cutthrough connection is made).
493 JH/14 Fix logging of errors under PIPELINING. Previously the log line giving
494 the relevant preceding SMTP command did not note the pipelining mode.
496 JH/15 Fix counting of empty lines in $body_linecount and $message_linecount.
497 Previously they were not counted.
499 JH/16 DANE: treat a TLSA lookup response having all non-TLSA RRs, the same
500 as one having no matching records. Previously we deferred the message
501 that needed the lookup.
503 JH/17 Fakereject: previously logged as a normal message arrival "<="; now
504 distinguished as "(=".
506 JH/18 Bug 1867: make the fail_defer_domains option on a dnslookup router work
507 for missing MX records. Previously it only worked for missing A records.
509 JH/19 Bug 1850: support Radius libraries that return REJECT_RC.
511 JH/20 Bug 1872: Ensure that acl_smtp_notquit is run when the connection drops
512 after the data-go-ahead and data-ack. Patch from Jason Betts.
514 JH/21 Bug 1846: Send DMARC forensic reports for reject and quarantine results,
515 even for a "none" policy. Patch from Tony Meyer.
517 JH/22 Fix continued use of a connection for further deliveries. If a port was
518 specified by a router, it must also match for the delivery to be
521 JH/23 Bug 1874: fix continued use of a connection for further deliveries.
522 When one of the recipients of a message was unsuitable for the connection
523 (has no matching addresses), we lost track of needing to mark it
524 deferred. As a result mail would be lost.
526 JH/24 Bug 1832: Log EHLO response on getting conn-close response for HELO.
528 JH/25 Decoding ACL controls is now done using a binary search; the source code
529 takes up less space and should be simpler to maintain. Merge the ACL
530 condition decode tables also, with similar effect.
532 JH/26 Fix problem with one_time used on a redirect router which returned the
533 parent address unchanged. A retry would see the parent address marked as
534 delivered, so not attempt the (identical) child. As a result mail would
537 JH/27 Fix a possible security hole, wherein a process operating with the Exim
538 UID can gain a root shell. Credit to http://www.halfdog.net/ for
539 discovery and writeup. Ubuntu bug 1580454; no bug raised against Exim
542 JH/28 Enable {spool,log} filesystem space and inode checks as default.
543 Main config options check_{log,spool}_{inodes,space} are now
544 100 inodes, 10MB unless set otherwise in the configuration.
546 JH/29 Fix the connection_reject log selector to apply to the connect ACL.
547 Previously it only applied to the main-section connection policy
550 JH/30 Bug 1897: fix callouts connection fallback from TLS to cleartext.
552 PP/01 Changed default Diffie-Hellman parameters to be Exim-specific, created
553 by me. Added RFC7919 DH primes as an alternative.
555 PP/02 Unbreak build via pkg-config with new hash support when crypto headers
556 are not in the system include path.
558 JH/31 Fix longstanding bug with aborted TLS server connection handling. Under
559 GnuTLS, when a session startup failed (eg because the client disconnected)
560 Exim did stdio operations after fclose. This was exposed by a recent
561 change which nulled out the file handle after the fclose.
563 JH/32 Bug 1909: Fix OCSP proof verification for cases where the proof is
564 signed directly by the cert-signing cert, rather than an intermediate
565 OCSP-signing cert. This is the model used by LetsEncrypt.
567 JH/33 Bug 1914: Ensure socket is nonblocking before draining after SMTP QUIT.
569 HS/01 Fix leak in verify callout under GnuTLS, about 3MB per recipient on
570 an incoming connection.
572 HS/02 Bug 1802: Do not half-close the connection after sending a request
575 HS/03 Use "auto" as the default EC curve parameter. For OpenSSL < 1.0.2
576 fallback to "prime256v1".
578 JH/34 SECURITY: Use proper copy of DATA command in error message.
579 Could leak key material. Remotely exploitable. CVE-2016-9963.
585 JH/01 Bug 1664: Disable OCSP for GnuTLS library versions at/before 3.3.16
586 and 3.4.4 - once the server is enabled to respond to an OCSP request
587 it does even when not requested, resulting in a stapling non-aware
588 client dropping the TLS connection.
590 TF/01 Code cleanup: Overhaul the debug_selector and log_selector machinery to
591 support variable-length bit vectors. No functional change.
593 TF/02 Improve the consistency of logging incoming and outgoing interfaces.
594 The I= interface field on outgoing lines is now after the H= remote
595 host field, same as incoming lines. There is a separate
596 outgoing_interface log selector which allows you to disable the
599 JH/02 Bug 728: Close logfiles after a daemon-process "exceptional" log write.
600 If not running log_selector +smtp_connection the mainlog would be held
601 open indefinitely after a "too many connections" event, including to a
602 deleted file after a log rotate. Leave the per net connection logging
603 leaving it open for efficiency as that will be quickly detected by the
604 check on the next write.
606 HS/01 Bug 1671: Fix post transport crash.
607 Processing the wait-<transport> messages could crash the delivery
608 process if the message IDs didn't exist for some reason. When
609 using 'split_spool_directory=yes' the construction of the spool
610 file name failed already, exposing the same netto behaviour.
612 JH/03 Bug 425: Capture substrings in $regex1, $regex2 etc from regex &
613 mime_regex ACL conditions.
615 JH/04 Bug 1686: When compiled with EXPERIMENTAL_DSN_INFO: Add extra information
616 to DSN fail messages (bounces): remote IP, remote greeting, remote response
617 to HELO, local diagnostic string.
619 JH/05 Downgrade message for a TLS-certificate-based authentication fail from
620 log line to debug. Even when configured with a tls authenticator many
621 client connections are expected to not authenticate in this way, so
622 an authenticate fail is not an error.
624 HS/02 Add the Exim version string to the process info. This way exiwhat
625 gives some more detail about the running daemon.
627 JH/06 Bug 1395: time-limit caching of DNS lookups, to the TTL value. This may
628 matter for fast-change records such as DNSBLs.
630 JH/07 Bug 1678: Always record an interface option value, if set, as part of a
631 retry record, even if constant. There may be multiple transports with
632 different interface settings and the retry behaviour needs to be kept
635 JH/08 Bug 1586: exiqgrep now refuses to run if there are unexpected arguments.
637 JH/09 Bug 1700: ignore space & tab embedded in base64 during decode.
639 JH/10 Bug 840: fix log_defer_output option of pipe transport
641 JH/11 Bug 830: use same host for all RCPTS of a message, even under
642 hosts_randomize. This matters a lot when combined with mua_wrapper.
644 JH/12 Bug 1706: percent and underbar characters are no longer escaped by the
645 ${quote_pgsql:<string>} operator.
647 JH/13 Bug 1708: avoid misaligned access in cached lookup.
649 JH/14 Change header file name for freeradius-client. Relevant if compiling
650 with Radius support; from the Gentoo tree and checked under Fedora.
652 JH/15 Bug 1712: Introduce $prdr_requested flag variable
654 JH/16 Bug 1714: Permit an empty string as expansion result for transport
655 option transport_filter, meaning no filtering.
657 JH/17 Bug 1713: Fix non-PDKIM_DEBUG build. Patch from Jasen Betts.
659 JH/18 Bug 1709: When built with TLS support, the tls_advertise_hosts option now
660 defaults to "*" (all hosts). The variable is now available when not built
661 with TLS, default unset, mainly to enable keeping the testsuite sane.
662 If a server certificate is not supplied (via tls_certificate) an error is
663 logged, and clients will find TLS connections fail on startup. Presumably
664 they will retry in-clear.
665 Packagers of Exim are strongly encouraged to create a server certificate
666 at installation time.
668 HS/03 Add -bP config_file as a synonym for -bP configure_file, for consistency
669 with the $config_file variable.
671 JH/19 Two additional event types: msg:rcpt:defer and msg:rcpt:host:defer. Both
672 in transport context, after the attempt, and per-recipient. The latter type
673 is per host attempted. The event data is the error message, and the errno
674 information encodes the lookup type (A vs. MX) used for the (first) host,
675 and the trailing two digits of the smtp 4xx response.
677 GF/01 Bug 1715: Fix for race condition in exicyclog, where exim could attempt
678 to write to mainlog (or rejectlog, paniclog) in the window between file
679 creation and permissions/ownership being changed. Particularly affects
680 installations where exicyclog is run as root, rather than exim user;
681 result is that the running daemon panics and dies.
683 JH/20 Bug 1701: For MySQL lookups, support MySQL config file option group names.
685 JH/21 Bug 1720: Add support for priority groups and weighted-random proxy
686 selection for the EXPERIMENTAL_SOCKS feature, via new per-proxy options
687 "pri" and "weight". Note that the previous implicit priority given by the
688 list order is no longer honoured.
690 JH/22 Bugs 963, 1721: Fix some corner cases in message body canonicalization
693 JH/23 Move SOCKS5 support from Experimental to mainline, enabled for a build
694 by defining SUPPORT_SOCKS.
696 JH/26 Move PROXY support from Experimental to mainline, enabled for a build
697 by defining SUPPORT_PROXY. Note that the proxy_required_hosts option
698 is renamed to hosts_proxy, and the proxy_{host,target}_{address,port}.
699 variables are renamed to proxy_{local,external}_{address,port}.
701 JH/27 Move Internationalisation support from Experimental to mainline, enabled
702 for a build by defining SUPPORT_I18N
704 JH/28 Bug 1745: Fix redis lookups to handle (quoted) spaces embedded in parts
705 of the query string, and make ${quote_redis:} do that quoting.
707 JH/29 Move Events support from Experimental to mainline, enabled by default
708 and removable for a build by defining DISABLE_EVENT.
710 JH/30 Updated DANE implementation code to current from Viktor Dukhovni.
712 JH/31 Fix bug with hosts_connection_nolog and named-lists which were wrongly
713 cached by the daemon.
715 JH/32 Move Redis support from Experimental to mainline, enabled for a build
716 by defining LOOKUP_REDIS. The libhiredis library is required.
718 JH/33 Bug 1748: Permit ACL dnslists= condition in non-smtp ACLs if explicit
719 keys are given for lookup.
721 JH/34 Bug 1192: replace the embedded copy of PolarSSL RSA routines in the DKIM
722 support, by using OpenSSL or GnuTLS library ones. This means DKIM is
723 only supported when built with TLS support. The PolarSSL SHA routines
724 are still used when the TLS library is too old for convenient support.
726 JH/35 Require SINGLE_DH_USE by default in OpenSSL (main config option
727 openssl_options), for security. OpenSSL forces this from version 1.1.0
728 server-side so match that on older versions.
730 JH/36 Bug 1778: longstanding bug in memory use by the ${run } expansion: A fresh
731 allocation for $value could be released as the expansion processing
732 concluded, but leaving the global pointer active for it.
734 JH/37 Bug 1769: Permit a VRFY ACL to override the default 252 response,
735 and to use the domains and local_parts ACL conditions.
737 JH/38 Fix cutthrough bug with body lines having a single dot. The dot was
738 incorrectly not doubled on cutthrough transmission, hence seen as a
739 body-termination at the receiving system - resulting in truncated mails.
740 Commonly the sender saw a TCP-level error, and retransmitted the message
741 via the normal store-and-forward channel. This could result in duplicates
742 received - but deduplicating mailstores were liable to retain only the
743 initial truncated version.
745 JH/39 Bug 1781: Fix use of DKIM private-keys having trailing '=' in the base-64.
747 JH/40 Fix crash in queryprogram router when compiled with EXPERIMENTAL_SRS.
749 JH/41 Bug 1792: Fix selection of headers to sign for DKIM: bottom-up. While
750 we're in there, support oversigning also; bug 1309.
752 JH/42 Bug 1796: Fix error logged on a malware scanner connection failure.
754 HS/04 Add support for keep_environment and add_environment options.
756 JH/43 Tidy coding issues detected by gcc --fsanitize=undefined. Some remain;
757 either intentional arithmetic overflow during PRNG, or testing config-
760 JH/44 Bug 1800: The combination of a -bhc commandline option and cutthrough
761 delivery resulted in actual delivery. Cancel cutthrough before DATA
764 JH/45 Fix cutthrough, when connection not opened by verify and target hard-
765 rejects a recipient: pass the reject to the originator.
767 JH/46 Multiple issues raised by Coverity. Some were obvious or plausible bugs.
768 Many were false-positives and ignorable, but it's worth fixing the
771 JH/47 Fix build on HP-UX and older Solaris, which need (un)setenv now also
772 for the new environment-manipulation done at startup. Move the routines
773 from being local to tls.c to being global via the os.c file.
775 JH/48 Bug 1807: Fix ${extract } for the numeric/3-string case. While preparsing
776 an extract embedded as result-arg for a map, the first arg for extract
777 is unavailable so we cannot tell if this is a numbered or keyed
778 extraction. Accept either.
784 JH/01 Bug 1545: The smtp transport option "retry_include_ip_address" is now
787 JH/02 The smtp transport option "multi_domain" is now expanded.
789 JH/03 The smtp transport now requests PRDR by default, if the server offers
792 JH/04 Certificate name checking on server certificates, when exim is a client,
793 is now done by default. The transport option tls_verify_cert_hostnames
794 can be used to disable this per-host. The build option
795 EXPERIMENTAL_CERTNAMES is withdrawn.
797 JH/05 The value of the tls_verify_certificates smtp transport and main options
798 default to the word "system" to access the system default CA bundle.
799 For GnuTLS, only version 3.0.20 or later.
801 JH/06 Verification of the server certificate for a TLS connection is now tried
802 (but not required) by default. The verification status is now logged by
803 default, for both outbound TLS and client-certificate supplying inbound
806 JH/07 Changed the default rfc1413 lookup settings to disable calls. Few
809 JH/08 The EXPERIMENTAL_DSN compile option is no longer needed; all Delivery
810 Status Notification (bounce) messages are now MIME format per RFC 3464.
811 Support for RFC 3461 DSN options NOTIFY,ENVID,RET,ORCPT can be advertised
812 under the control of the dsn_advertise_hosts option, and routers may
813 have a dsn_lasthop option.
815 JH/09 A timeout of 2 minutes is now applied to all malware scanner types by
816 default, modifiable by a malware= option. The list separator for
817 the options can now be changed in the usual way. Bug 68.
819 JH/10 The smtp_receive_timeout main option is now expanded before use.
821 JH/11 The incoming_interface log option now also enables logging of the
822 local interface on delivery outgoing connections.
824 JH/12 The cutthrough-routing facility now supports multi-recipient mails,
825 if the interface and destination host and port all match.
827 JH/13 Bug 344: The verify = reverse_host_lookup ACL condition now accepts a
830 JH/14 Bug 1573: The spam= ACL condition now additionally supports Rspamd.
831 Patch from Andrew Lewis.
833 JH/15 Bug 670: The spamd_address main option (for the spam= ACL condition)
834 now supports optional time-restrictions, weighting, and priority
835 modifiers per server. Patch originally by <rommer@active.by>.
837 JH/16 The spamd_address main option now supports a mixed list of local
838 and remote servers. Remote servers can be IPv6 addresses, and
839 specify a port-range.
841 JH/17 Bug 68: The spamd_address main option now supports an optional
842 timeout value per server.
844 JH/18 Bug 1581: Router and transport options headers_add/remove can
845 now have the list separator specified.
847 JH/19 Bug 392: spamd_address, and clamd av_scanner, now support retry
850 JH/20 Bug 1571: Ensure that $tls_in_peerdn is set, when verification fails
853 JH/21 Support for the A6 type of dns record is withdrawn.
855 JH/22 Bug 608: The result of a QUIT or not-QUIT toplevel ACL now matters
856 rather than the verbs used.
858 JH/23 Bug 1572: Increase limit on SMTP confirmation message copy size
859 from 255 to 1024 chars.
861 JH/24 Verification callouts now attempt to use TLS by default.
863 HS/01 DNSSEC options (dnssec_require_domains, dnssec_request_domains)
864 are generic router options now. The defaults didn't change.
866 JH/25 Bug 466: Add RFC2322 support for MIME attachment filenames.
867 Original patch from Alexander Shikoff, worked over by JH.
869 HS/02 Bug 1575: exigrep falls back to autodetection of compressed
870 files if ZCAT_COMMAND is not executable.
872 JH/26 Bug 1539: Add timeout/retry options on dnsdb lookups.
874 JH/27 Bug 286: Support SOA lookup in dnsdb lookups.
876 JH/28 Bug 1588: Do not use the A lookup following an AAAA for setting the FQDN.
877 Normally benign, it bites when the pair was led to by a CNAME;
878 modern usage is to not canonicalize the domain to a CNAME target
879 (and we were inconsistent anyway for A-only vs AAAA+A).
881 JH/29 Bug 1632: Removed the word "rejected" from line logged for ACL discards.
883 JH/30 Check the forward DNS lookup for DNSSEC, in addition to the reverse,
884 when evaluating $sender_host_dnssec.
886 JH/31 Check the HELO verification lookup for DNSSEC, adding new
887 $sender_helo_dnssec variable.
889 JH/32 Bug 1397: Enable ECDHE on OpenSSL, just the NIST P-256 curve.
891 JH/33 Bug 1346: Note MAIL cmd seen in -bS batch, to avoid smtp_no_mail log.
893 JH/34 Bug 1648: Fix a memory leak seen with "mailq" and large queues.
895 JH/35 Bug 1642: Fix support of $spam_ variables at delivery time. Was
896 documented as working, but never had. Support all but $spam_report.
898 JH/36 Bug 1659: Guard checking of input smtp commands again pseudo-command
899 added for tls authenticator.
901 HS/03 Add perl_taintmode main config option
907 TL/01 When running the test suite, the README says that variables such as
908 no_msglog_check are global and can be placed anywhere in a specific
909 test's script, however it was observed that placement needed to be near
910 the beginning for it to behave that way. Changed the runtest perl
911 script to read through the entire script once to detect and set these
912 variables, reset to the beginning of the script, and then run through
913 the script parsing/test process like normal.
915 TL/02 The BSD's have an arc4random API. One of the functions to induce
916 adding randomness was arc4random_stir(), but it has been removed in
917 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
918 function when detected.
920 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
921 cause callback expansion.
923 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
924 syntax errors in an expansion can be treated as a string instead of
925 logging or causing an error, due to the internal use of bool_lax
926 instead of bool when processing it.
928 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
929 server certificates when making smtp deliveries.
931 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
933 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
935 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
937 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
938 Merged patch from Sebastian Wiedenroth.
940 JH/05 Fix results-pipe from transport process. Several recipients, combined
941 with certificate use, exposed issues where response data items split
942 over buffer boundaries were not parsed properly. This eventually
943 resulted in duplicates being sent. This issue only became common enough
944 to notice due to the introduction of connection certificate information,
945 the item size being so much larger. Found and fixed by Wolfgang Breyha.
947 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
948 size buffer was used, resulting in syntax errors when an expansion
951 JH/07 Add support for directories of certificates when compiled with a GnuTLS
952 version 3.3.6 or later.
954 JH/08 Rename the TPDA experimental facility to Event Actions. The #ifdef
955 is EXPERIMENTAL_EVENT, the main-configuration and transport options
956 both become "event_action", the variables become $event_name, $event_data
957 and $event_defer_errno. There is a new variable $verify_mode, usable in
958 routers, transports and related events. The tls:cert event is now also
959 raised for inbound connections, if the main configuration event_action
962 TL/06 In test suite, disable OCSP for old versions of openssl which contained
963 early OCSP support, but no stapling (appears to be less than 1.0.0).
965 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
966 server certificate names available under the smtp transport option
967 "tls_verify_cert_hostname" now do not permit multi-component wildcard
970 JH/10 Time-related extraction expansions from certificates now use the main
971 option "timezone" setting for output formatting, and are consistent
972 between OpenSSL and GnuTLS compilations. Bug 1541.
974 JH/11 Fix a crash in mime ACL when meeting a zero-length, quoted or RFC2047-
975 encoded parameter in the incoming message. Bug 1558.
977 JH/12 Bug 1527: Autogrow buffer used in reading spool files. Since they now
978 include certificate info, eximon was claiming there were spoolfile
981 JH/13 Bug 1521: Fix ldap lookup for single-attr request, multiple-attr return.
983 JH/14 Log delivery-related information more consistently, using the sequence
984 "H=<name> [<ip>]" wherever possible.
986 TL/07 Bug 1547: Omit RFCs from release. Draft and RFCs have licenses which
987 are problematic for Debian distribution, omit them from the release
990 JH/15 Updates and fixes to the EXPERIMENTAL_DSN feature.
992 JH/16 Fix string representation of time values on 64bit time_t architectures.
995 JH/17 Fix a null-indirection in certextract expansions when a nondefault
996 output list separator was used.
1001 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
1002 checkers that were complaining about end of non-void function with no
1005 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
1006 This was a regression introduced in 4.83 by another bugfix.
1008 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
1010 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
1011 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
1017 TF/01 Correctly close the server side of TLS when forking for delivery.
1019 When a message was received over SMTP with TLS, Exim failed to clear up
1020 the incoming connection properly after forking off the child process to
1021 deliver the message. In some situations the subsequent outgoing
1022 delivery connection happened to have the same fd number as the incoming
1023 connection previously had. Exim would try to use TLS and fail, logging
1024 a "Bad file descriptor" error.
1026 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
1027 utilities have not been installed.
1029 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
1030 temporary space as the ACL may create new global variables.
1032 TL/01 LDAP support uses per connection or global context settings, depending
1033 upon the detected version of the libraries at build time.
1035 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
1036 to extract and use the src ip:port in logging and expansions as if it
1037 were a direct connection from the outside internet. PPv2 support was
1038 updated based on HAProxy spec change in May 2014.
1040 JH/02 Add ${listextract {number}{list}{success}{fail}}.
1042 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
1043 Properly escape header and check for NULL return.
1045 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
1048 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
1050 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
1051 characters in header names, implemented as a verify condition.
1052 Contributed by Michael Fischer v. Mollard.
1054 TL/05 Rename SPF condition results err_perm and err_temp to standardized
1055 results permerror and temperror. Previous values are deprecated but
1056 still accepted. In a future release, err_perm and err_temp will be
1057 completely removed, which will be a backward incompatibility if the
1058 ACL tests for either of these two old results. Patch contributed by
1059 user bes-internal on the mailing list.
1061 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
1063 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
1064 selectors, in both main and reject logs.
1066 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
1069 JH/07 Add malware type "sock" for talking to simple daemon.
1071 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
1073 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
1074 routers/transports under cutthrough routing.
1076 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
1077 numbers. Touch up "bool" conditional to keep the same definition.
1079 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
1081 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
1083 JH/12 Expand items in router/transport headers_add or headers_remove lists
1084 individually rather than the list as a whole. Bug 1452.
1086 Required for reasonable handling of multiple headers_ options when
1087 they may be empty; requires that headers_remove items with embedded
1088 colons must have them doubled (or the list-separator changed).
1090 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
1091 view the policy declared in the DMARC record. Currently, $dmarc_status
1092 is a combined value of both the record presence and the result of the
1095 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
1097 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
1098 dnslookup router and the smtp transport (applying to the forward
1101 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
1102 of ldap servers used for a specific lookup. Patch provided by Heiko
1105 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
1106 New variable $lookup_dnssec_authenticated for observability.
1108 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
1109 Patch submitted by Lars Timman.
1111 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
1113 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
1114 Requires trusted mode and valid format message id, aborts otherwise.
1115 Patch contributed by Heiko Schlichting.
1117 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
1118 certextract with support for various fields. Bug 1358.
1120 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
1121 is requested by default, modifiable by smtp transport option
1124 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
1125 operate on certificate variables to give certificate fingerprints
1126 Also new ${sha256:cert_variable}.
1128 JH/23 The PRDR feature is moved from being Experimental into the mainline.
1130 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
1131 Christian Aistleitner.
1133 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
1135 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
1136 file. Patch from Wolfgang Breyha.
1138 JH/25 Expand the coverage of the delivery $host and $host_address to
1139 client authenticators run in verify callout. Bug 1476.
1141 JH/26 Port service names are now accepted for tls_on_connect_ports, to
1142 align with daemon_smtp_ports. Bug 72.
1144 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
1145 support and error reporting did not work properly.
1147 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
1148 and is readable. Patch from Andrew Colin Kissa.
1150 TL/14 Enhance documentation of ${run expansion and how it parses the
1151 commandline after expansion, particularly in the case when an
1152 unquoted variable expansion results in an empty value.
1154 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
1156 PP/02 Fix internal collision of T_APL on systems which support RFC3123
1157 by renaming away from it. Addresses GH issue 15, reported by
1160 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
1162 TL/15 SECURITY: prevent double expansion in math comparison functions
1163 (can expand unsanitized data). Not remotely exploitable.
1170 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
1172 PP/02 Make -n do something, by making it not do something.
1173 When combined with -bP, the name of an option is not output.
1175 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
1178 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
1179 $sender_host_name and config options to manage this, and basic check
1182 PP/05 DSCP support for outbound connections and control modifier for inbound.
1184 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
1185 (Only plugin which currently uses this is kerberos4, which nobody should
1186 be using, but we should make it available and other future plugins might
1187 conceivably use it, even though it would break NAT; stuff *should* be
1188 using channel bindings instead).
1190 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
1191 name; added for Sendmail compatibility; requires admin caller.
1192 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
1193 just ignore it); requires trusted caller.
1194 Also parse but ignore: -Ac -Am -X<logfile>
1197 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
1199 TL/02 Add +smtp_confirmation as a default logging option.
1201 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
1202 Patch by Magnus Holmgren from 2007-02-20.
1204 TL/04 Bugzilla 1281 - Spec typo.
1205 Bugzilla 1283 - Spec typo.
1206 Bugzilla 1290 - Spec grammar fixes.
1208 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
1210 TL/06 Add Experimental DMARC support using libopendmarc libraries.
1212 TL/07 Fix an out of order global option causing a segfault. Reported to dev
1213 mailing list by by Dmitry Isaikin.
1215 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
1217 JH/02 Support "G" suffix to numbers in ${if comparisons.
1219 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
1221 NM/01 Bugzilla 1197 - Spec typo
1222 Bugzilla 1196 - Spec examples corrections
1224 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
1226 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
1227 gnutls_enable_pkcs11, but renamed to more accurately indicate its
1230 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
1231 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
1233 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
1234 "acl {{name}{arg}...}", and optional args on acl condition
1237 JH/05 Permit multiple router/transport headers_add/remove lines.
1239 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
1241 JH/07 Avoid using a waiting database for a single-message-only transport.
1242 Performance patch from Paul Fisher. Bugzilla 1262.
1244 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
1247 JH/09 Add $headers_added variable, with content from use of ACL modifier
1248 add_header (but not yet added to the message). Bugzilla 199.
1250 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
1251 Pulled from Bugzilla 817 by Wolfgang Breyha.
1253 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
1255 (nb: this is the same fix as in Exim 4.80.1)
1257 JH/11 Add A= logging on delivery lines, and a client_set_id option on
1260 JH/12 Add optional authenticated_sender logging to A= and a log_selector
1263 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
1265 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
1266 advertise SMTP AUTH mechanism to us, instead of a generic
1267 protocol violation error. Also, make Exim more robust to bad
1268 data from the Dovecot auth socket.
1270 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
1272 When a queue runner is handling a message, Exim first routes the
1273 recipient addresses, during which it prunes them based on the retry
1274 hints database. After that it attempts to deliver the message to
1275 any remaining recipients. It then updates the hints database using
1278 So if a recipient address works intermittently, it can get repeatedly
1279 deferred at routing time. The retry hints record remains fresh so the
1280 address never reaches the final cutoff time.
1282 This is a fairly common occurrence when a user is bumping up against
1283 their storage quota. Exim had some logic in its local delivery code
1284 to deal with this. However it did not apply to per-recipient defers
1285 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
1287 This change adds a proper retry rule check during routing so that the
1288 final cutoff time is checked against the message's age. We only do
1289 this check if there is an address retry record and there is not a
1290 domain retry record; this implies that previous attempts to handle
1291 the address had the retry_use_local_parts option turned on. We use
1292 this as an approximation for the destination being like a local
1293 delivery, as in LMTP.
1295 I suspect this new check makes the old local delivery cutoff check
1296 redundant, but I have not verified this so I left the code in place.
1298 TF/02 Correct gecos expansion when From: is a prefix of the username.
1300 Test 0254 submits a message to Exim with the header
1304 When I ran the test suite under the user fanf2, Exim expanded
1305 the header to contain my full name, whereas it should have added
1306 a Resent-Sender: header. It erroneously treats any prefix of the
1307 username as equal to the username.
1309 This change corrects that bug.
1311 GF/01 DCC debug and logging tidyup
1312 Error conditions log to paniclog rather than rejectlog.
1313 Debug lines prefixed by "DCC: " to remove any ambiguity.
1315 TF/03 Avoid unnecessary rebuilds of lookup-related code.
1317 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
1318 Bug spotted by Jeremy Harris; was flawed since initial commit.
1319 Would have resulted in OCSP responses post-SNI triggering an Exim
1320 NULL dereference and crash.
1322 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
1324 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
1325 Bug detection, analysis and fix by Samuel Thibault.
1326 Bugzilla 1331, Debian bug #698092.
1328 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
1330 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
1331 Server implementation by Todd Lyons, client by JH.
1332 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
1333 config variable "prdr_enable" controls whether the server
1334 advertises the facility. If the client requests PRDR a new
1335 acl_data_smtp_prdr ACL is called once for each recipient, after
1336 the body content is received and before the acl_smtp_data ACL.
1337 The client is controlled by both of: a hosts_try_prdr option
1338 on the smtp transport, and the server advertisement.
1339 Default client logging of deliveries and rejections involving
1340 PRDR are flagged with the string "PRDR".
1342 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
1343 fclose(). Diagnosis by Todd Lyons.
1345 PP/17 Update configure.default to handle IPv6 localhost better.
1346 Patch by Alain Williams (plus minor tweaks).
1349 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
1350 This is now consistent with GnuTLS, and is now documented: the
1351 previous undocumented portable approach to treating the option as
1352 unset was to force an expansion failure. That still works, and
1353 an empty string is now equivalent.
1355 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
1356 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
1357 not performing validation itself.
1359 PP/20 Added force_command boolean option to pipe transport.
1360 Patch from Nick Koston, of cPanel Inc.
1362 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
1365 TF/04 Added udpsend ACL modifier and hexquote expansion operator
1367 PP/21 Fix eximon continuous updating with timestamped log-files.
1368 Broken in a format-string cleanup in 4.80, missed when I repaired the
1369 other false fix of the same issue.
1370 Report and fix from Heiko Schlichting.
1373 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
1374 Report from Prashanth Katuri.
1376 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
1377 It's SecureTransport, so affects any MacOS clients which use the
1378 system-integrated TLS libraries, including email clients.
1380 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
1381 using a MIME ACL for non-SMTP local injection.
1382 Report and assistance in diagnosis by Warren Baker.
1384 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
1386 JH/16 Fix comparisons for 64b. Bugzilla 1385.
1388 TL/09 Add expansion variable $authenticated_fail_id to keep track of
1389 last id that failed so it may be referenced in subsequent ACL's.
1391 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
1394 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
1395 ldap library initialization, allowing self-signed CA's to be
1396 used. Also properly sets require_cert option later in code by
1397 using NULL (global ldap config) instead of ldap handle (per
1398 session). Bug diagnosis and testing by alxgomz.
1400 TL/12 Enhanced documentation in the ratelimit.pl script provided in
1401 the src/util/ subdirectory.
1403 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
1404 renamed to Transport Post Delivery Action by Jeremy Harris, as
1407 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
1408 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
1409 redis_servers = needs to be configured which will be used by the redis
1410 lookup. Patch from Warren Baker, of The Packet Hub.
1412 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
1414 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
1415 hostname or reverse DNS when processing a host list. Used suggestions
1416 from multiple comments on this bug.
1418 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
1420 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
1421 Missed a few lines, added it to make the runtest require no keyboard
1424 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
1425 contains upper case chars. Make router use caseful_local_part.
1427 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
1428 support when GnuTLS has been built with p11-kit.
1434 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
1436 This, or similar/improved, will also be change PP/11 of 4.82.
1442 PP/01 Handle short writes when writing local log-files.
1443 In practice, only affects FreeBSD (8 onwards).
1444 Bugzilla 1053, with thanks to Dmitry Isaikin.
1446 NM/01 Bugzilla 949 - Documentation tweak
1448 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
1451 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
1453 PP/02 Implemented gsasl authenticator.
1455 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
1457 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
1458 `pkg-config foo` for cflags/libs.
1460 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
1461 with rest of GSASL and with heimdal_gssapi.
1463 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
1464 `pkg-config foo` for cflags/libs for the TLS implementation.
1466 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
1467 properties get this fed in as external SSF. A number of robustness
1468 and debugging improvements to the cyrus_sasl authenticator.
1470 PP/08 cyrus_sasl server now expands the server_realm option.
1472 PP/09 Bugzilla 1214 - Log authentication information in reject log.
1473 Patch by Jeremy Harris.
1475 PP/10 Added dbmjz lookup type.
1477 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
1479 PP/12 MAIL args handles TAB as well as SP, for better interop with
1480 non-compliant senders.
1481 Analysis and variant patch by Todd Lyons.
1483 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
1484 Bug report from Lars Müller <lars@samba.org> (via SUSE),
1485 Patch from Dirk Mueller <dmueller@suse.com>
1487 PP/13 tls_peerdn now print-escaped for spool files.
1488 Observed some $tls_peerdn in wild which contained \n, which resulted
1489 in spool file corruption.
1491 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
1492 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
1493 or write after TLS renegotiation, which otherwise led to messages
1496 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
1497 as a tracking header (ie: a signed header comes before the signature).
1498 Patch from Wolfgang Breyha.
1500 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
1501 comma-sep list; embedded commas doubled.
1503 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
1505 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
1507 Report and patch from Dmitry Banschikov.
1509 PP/16 Removed "dont_insert_empty_fragments" from "openssl_options".
1510 Removed SSL_clear() after SSL_new() which led to protocol negotiation
1511 failures. We appear to now support TLS1.1+ with Exim.
1513 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
1514 lets Exim select keys and certificates based upon TLS SNI from client.
1515 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
1516 before an outbound SMTP session. New log_selector, +tls_sni.
1518 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
1519 NULL dereference. Report and patch from Alun Jones.
1521 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
1522 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
1523 Not seeing resolver debug output on NetBSD, but suspect this is a
1524 resolver implementation change.
1526 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
1527 Left warnings. Added "eximon gdb" invocation mode.
1529 PP/21 Defaulting "accept_8bitmime" to true, not false.
1531 PP/22 Added -bw for inetd wait mode support.
1533 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
1534 locate the relevant includes and libraries. Made this the default.
1536 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
1537 Bugzilla 1246, report and most of solution from Tomasz Kusy.
1539 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
1540 This may cause build issues on older platforms.
1542 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
1543 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
1544 gnutls_require_mac & gnutls_require_protocols (no longer supported).
1545 Added SNI support via GnuTLS too.
1546 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
1548 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
1550 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
1551 Applied second patch from Janne, implementing suggestion to default
1552 multiple-strings-in-record handling to match SPF spec.
1554 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
1556 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
1557 read-only, out of scope).
1558 Patch from Wolfgang Breyha, report from Stuart Northfield.
1560 PP/29 Fix three issues highlighted by clang analyser static analysis.
1561 Only crash-plausible issue would require the Cambridge-specific
1562 iplookup router and a misconfiguration.
1563 Report from Marcin Mirosław.
1565 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
1567 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
1568 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
1569 As part of this, removing so much warning spew let me fix some minor
1570 real issues in debug logging.
1572 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
1573 assignment on my part. Fixed.
1575 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
1576 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
1577 Janne Snabb (who went above and beyond: thank you).
1579 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
1580 string otherwise requires a connection and a bunch more work and it's
1581 relatively easy to get wrong. Should also expose TLS library linkage
1584 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
1585 64-bit ${eval} (JH/03).
1587 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
1588 GNU libc to support some of the 64-bit stuff, should not lead to
1589 conflicts. Defined before os.h is pulled in, so if a given platform
1590 needs to override this, it can.
1592 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
1593 protection layer was required, which is not implemented.
1594 Bugzilla 1254, patch from Wolfgang Breyha.
1596 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
1597 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
1598 tls_dhparam take prime identifiers. Also unbreak combination of
1599 OpenSSL+DH_params+TLSSNI.
1601 PP/39 Disable SSLv2 by default in OpenSSL support.
1607 PP/01 Solaris build fix for Oracle's LDAP libraries.
1608 Bugzilla 1109, patch from Stephen Usher.
1610 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
1612 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
1615 TF/02 Fix a couple more cases where we did not log the error message
1616 when unlink() failed. See also change 4.74-TF/03.
1618 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
1619 lock up or crash if it happened to be inside a call to libc when it
1620 got a SIGUSR1 from exiwhat.
1622 The SIGUSR1 handler appends the current process status to the process
1623 log which is later printed by exiwhat. It used to use the general
1624 purpose logging code to do this, but several functions it calls are
1625 not safe for signals.
1627 The new output code in the SIGUSR1 handler is specific to the process
1628 log, and simple enough that it's easy to inspect for signal safety.
1629 Removing some special cases also simplifies the general logging code.
1630 Removing the spurious timestamps from the process log simplifies
1633 TF/04 Improved ratelimit ACL condition.
1635 The /noupdate option has been deprecated in favour of /readonly which
1636 has clearer semantics. The /leaky, /strict, and /readonly update modes
1637 are mutually exclusive. The update mode is no longer included in the
1638 database key; it just determines when the database is updated. (This
1639 means that when you upgrade Exim will forget old rate measurements.)
1641 Exim now checks that the per_* options are used with an update mode that
1642 makes sense for the current ACL. For example, when Exim is processing a
1643 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
1644 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
1645 must specify per_mail/readonly. If you omit the update mode it defaults to
1646 /leaky where that makes sense (as before) or /readonly where required.
1648 The /noupdate option is now undocumented but still supported for
1649 backwards compatibility. It is equivalent to /readonly except that in
1650 ACLs where /readonly is required you may specify /leaky/noupdate or
1651 /strict/noupdate which are treated the same as /readonly.
1653 A useful new feature is the /count= option. This is a generalization
1654 of the per_byte option, so that you can measure the throughput of other
1655 aggregate values. For example, the per_byte option is now equivalent
1656 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
1658 The per_rcpt option has been generalized using the /count= mechanism
1659 (though it's more complicated than the per_byte equivalence). When it is
1660 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
1661 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
1662 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
1663 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
1664 Note that using per_rcpt with a non-readonly update mode in more than
1665 one ACL will cause the recipients to be double-counted. (The per_mail
1666 and per_byte options don't have this problem.)
1668 The handling of very low rates has changed slightly. If the computed rate
1669 is less than the event's count (usually one) then this event is the first
1670 after a long gap. In this case the rate is set to the same as this event's
1671 count, so that the first message of a spam run is counted properly.
1673 The major new feature is a mechanism for counting the rate of unique
1674 events. The new per_addr option counts the number of different
1675 recipients that someone has sent messages to in the last time period. It
1676 behaves like per_rcpt if all the recipient addresses are different, but
1677 duplicate recipient addresses do not increase the measured rate. Like
1678 the /count= option this is a general mechanism, so the per_addr option
1679 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
1680 example, measure the rate that a client uses different sender addresses
1681 with the options per_mail/unique=$sender_address. There are further
1682 details in the main documentation.
1684 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
1686 TF/06 Removed a few PCRE remnants.
1688 TF/07 Automatically extract Exim's version number from tags in the git
1689 repository when doing development or release builds.
1691 PP/02 Raise smtp_cmd_buffer_size to 16kB.
1692 Bugzilla 879. Patch from Paul Fisher.
1694 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
1695 Heavily based on revision 40f9a89a from Simon Arlott's tree.
1698 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
1700 PP/05 Variable $av_failed, true if the AV scanner deferred.
1701 Bugzilla 1078. Patch from John Horne.
1703 PP/06 Stop make process more reliably on build failure.
1704 Bugzilla 1087. Patch from Heiko Schlittermann.
1706 PP/07 Make maildir_use_size_file an _expandable_ boolean.
1707 Bugzilla 1089. Patch from Heiko Schlittermann.
1709 PP/08 Handle ${run} returning more data than OS pipe buffer size.
1710 Bugzilla 1131. Patch from Holger Weiß.
1712 PP/09 Handle IPv6 addresses with SPF.
1713 Bugzilla 860. Patch from Wolfgang Breyha.
1715 PP/10 GnuTLS: support TLS 1.2 & 1.1.
1717 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
1720 PP/11 match_* no longer expand right-hand-side by default.
1721 New compile-time build option, EXPAND_LISTMATCH_RHS.
1722 New expansion conditions, "inlist", "inlisti".
1724 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
1726 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
1728 PP/14 fix log_write() format string regression from TF/03.
1729 Bugzilla 1152. Patch from Dmitry Isaikin.
1735 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
1737 PP/02 Harmonised TLS library version reporting; only show if debugging.
1738 Layout now matches that introduced for other libraries in 4.74 PP/03.
1740 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
1742 PP/04 New "dns_use_edns0" global option.
1744 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
1747 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
1748 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
1750 TK/01 Updated PolarSSL code to 0.14.2.
1751 Bugzilla 1097. Patch from Andreas Metzler.
1753 PP/07 Catch divide-by-zero in ${eval:...}.
1754 Fixes bugzilla 1102.
1756 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
1759 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
1760 format-string attack -- SECURITY: remote arbitrary code execution.
1762 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
1763 time unintentionally subject to list matching rules, letting the header
1764 cause arbitrary Exim lookups (of items which can occur in lists, *not*
1765 arbitrary string expansion). This allowed for information disclosure.
1767 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
1768 INT_MIN/-1 -- value coerced to INT_MAX.
1774 NM/01 Workaround for PCRE version dependency in version reporting
1777 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
1778 This fixes portability to compilers other than gcc, notably
1779 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
1781 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
1782 makefiles for portability to HP-UX and POSIX correctness.
1784 PP/01 Permit LOOKUP_foo enabling on the make command-line.
1785 Also via indented variable definition in the Makefile.
1786 (Debugging by Oliver Heesakkers).
1788 PP/02 Restore caching of spamd results with expanded spamd_address.
1789 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
1791 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
1792 Improves build reliability. Fix from: Frank Elsner
1794 NM/02 Fix wide character breakage in the rfc2047 coding
1795 Fixes bug 1064. Patch from Andrey N. Oktyabrski
1797 NM/03 Allow underscore in dnslist lookups
1798 Fixes bug 1026. Patch from Graeme Fowler
1800 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
1801 Code patches from Adam Ciarcinski of NetBSD.
1803 NM/04 Fixed exiqgrep to cope with mailq missing size issue
1806 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
1807 is logged, to avoid truncation. Patch from John Horne.
1809 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
1810 Patch from Jakob Hirsch.
1812 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
1813 SQL string expansion failure details.
1814 Patch from Andrey Oktyabrski.
1816 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
1817 Patch from Simon Arlott.
1819 PP/09 New lookups functionality failed to compile on old gcc which rejects
1820 extern declarations in function scope.
1821 Patch from Oliver Fleischmann
1823 PP/10 Use sig_atomic_t for flags set from signal handlers.
1824 Check getgroups() return and improve debugging.
1825 Fixed developed for diagnosis in bug 927 (which turned out to be
1828 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
1829 Patch from Mark Zealey.
1831 PP/12 Bugzilla 1056: Improved spamd server selection.
1832 Patch from Mark Zealey.
1834 PP/13 Bugzilla 1086: Deal with maildir quota file races.
1835 Based on patch from Heiko Schlittermann.
1837 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
1838 Patch from Uwe Doering, sign-off by Michael Haardt.
1840 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
1841 variable declaration deep within a block. Bug and patch from
1844 PP/15 lookups-Makefile IRIX compatibility coercion.
1846 PP/16 Make DISABLE_DKIM build knob functional.
1848 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
1849 Patch by Simon Arlott
1851 TF/03 Fix valgrind.h portability to C89 compilers that do not support
1852 variable argument macros. Our copy now differs from upstream.
1858 TF/01 Failure to get a lock on a hints database can have serious
1859 consequences so log it to the panic log.
1861 TF/02 Log LMTP confirmation messages in the same way as SMTP,
1862 controlled using the smtp_confirmation log selector.
1864 TF/03 Include the error message when we fail to unlink a spool file.
1866 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
1867 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
1868 for maintaining out-of-tree patches for some time.
1870 PP/01 Bugzilla 139: Documentation and portability issues.
1871 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
1872 Handle per-OS dynamic-module compilation flags.
1874 PP/02 Let /dev/null have normal permissions.
1875 The 4.73 fixes were a little too stringent and complained about the
1876 permissions on /dev/null. Exempt it from some checks.
1877 Reported by Andreas M. Kirchwitz.
1879 PP/03 Report version information for many libraries, including
1880 Exim version information for dynamically loaded libraries. Created
1881 version.h, now support a version extension string for distributors
1882 who patch heavily. Dynamic module ABI change.
1884 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
1885 privilege escalation vulnerability whereby the Exim run-time user
1886 can cause root to append content of the attacker's choosing to
1889 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
1892 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
1893 If dropping privileges for untrusted macros, we disabled normal logging
1894 on the basis that it would fail; for the Exim run-time user, this is not
1895 the case, and it resulted in successful deliveries going unlogged.
1896 Fixed. Reported by Andreas Metzler.
1902 PP/01 Date: & Message-Id: revert to normally being appended to a message,
1903 only prepend for the Resent-* case. Fixes regression introduced in
1904 Exim 4.70 by NM/22 for Bugzilla 607.
1906 PP/02 Include check_rfc2047_length in configure.default because we're seeing
1907 increasing numbers of administrators be bitten by this.
1909 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
1911 PP/03 Bugzilla 994: added openssl_options main configuration option.
1913 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
1915 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
1917 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
1919 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
1920 without a peer certificate, leading to a segfault because of an
1921 assumption that peers always have certificates. Be a little more
1922 paranoid. Problem reported by Martin Tscholak.
1924 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
1925 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
1926 NB: ClamAV planning to remove STREAM in "middle of 2010".
1927 CL also introduces -bmalware, various -d+acl logging additions and
1928 more caution in buffer sizes.
1930 PP/09 Implemented reverse_ip expansion operator.
1932 PP/10 Bugzilla 937: provide a "debug" ACL control.
1934 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
1936 PP/12 Bugzilla 973: Implement --version.
1938 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
1940 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
1942 PP/15 Bugzilla 816: support multiple condition rules on Routers.
1944 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
1945 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
1946 ignore trailing whitespace.
1948 JJ/02 prevent non-panic DKIM error from being sent to paniclog
1950 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
1953 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
1954 Notification from Dr Andrew Aitchison.
1956 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
1957 ExtendedDetectionInfo response format.
1958 Notification from John Horne.
1960 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
1963 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
1964 XSL and documented dependency on system catalogs, with examples of how
1967 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
1970 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
1971 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
1972 configuration file which is writeable by the Exim user or group.
1974 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
1975 of configuration files to cover files specified with the -C option if
1976 they are going to be used with root privileges, not just the default
1979 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
1980 option (effectively making it always true).
1982 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
1983 files to be used while preserving root privileges.
1985 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
1986 that rogue child processes cannot use them.
1988 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
1989 run-time user, instead of root.
1991 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
1992 Exim run-time user without dropping privileges.
1994 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
1995 result string, instead of calling string_vformat() twice with the same
1998 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
1999 for other users. Others should always drop root privileges if they use
2000 -C on the command line, even for a whitelisted configure file.
2002 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
2004 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
2010 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
2011 $data_path, and $header_path variables; fixed documentation bugs and
2014 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
2015 exipick to access non-standard spools, including the "frozen" queue
2018 NM/01 Bugzilla 965: Support mysql stored procedures.
2019 Patch from Alain Williams
2021 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
2023 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
2024 Patch from Andreas Metzler
2026 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
2027 Patch from Kirill Miazine
2029 NM/05 Bugzilla 671: Added umask to procmail example.
2031 JJ/03 installed exipick 20100323.0, fixing doc bug
2033 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
2034 directory. Notification and patch from Dan Rosenberg.
2036 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
2038 TK/02 Improve log output when DKIM signing operation fails.
2040 MH/01 Treat the transport option dkim_domain as a colon separated
2041 list, not as a single string, and sign the message with each element,
2042 omitting multiple occurences of the same signer.
2044 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
2045 Bugzilla 985, 986. Patch by Simon Arlott
2047 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
2048 Patch by Simon Arlott
2050 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
2051 MBX locking. Notification from Dan Rosenberg.
2057 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
2059 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
2061 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
2063 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
2065 NM/04 Bugzilla 915: Flag broken perl installation during build.
2071 TK/01 Added patch by Johannes Berg that expands the main option
2072 "spamd_address" if it starts with a dollar sign.
2074 TK/02 Write list of recipients to X-Envelope-Sender header when building
2075 the mbox-format spool file for content scanning (suggested by Jakob
2078 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
2079 (http://www.dcc-servers.net/) support via dccifd. Activated by
2080 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
2082 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
2083 by Mark Daniel Reidel <mr@df.eu>.
2085 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
2086 When building exim an external PCRE library is now needed -
2087 PCRE is a system library on the majority of modern systems.
2088 See entry on PCRE_LIBS in EDITME file.
2090 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
2091 conversation. Added nologin parameter to request.
2092 Patch contributed by Kirill Miazine.
2094 TF/01 Do not log submission mode rewrites if they do not change the address.
2096 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
2098 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
2099 log files in place. Contributed by Roberto Lima.
2101 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
2103 TF/03 Bugzilla 615: When checking the local_parts router precondition
2104 after a local_part_suffix or local_part_prefix option, Exim now
2105 does not use the address's named list lookup cache, since this
2106 contains cached lookups for the whole local part.
2108 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
2109 Robert Millan. Documentation is in experimental-spec.txt.
2111 TF/04 Bugzilla 668: Fix parallel build (make -j).
2113 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
2115 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
2116 Patch provided by Jan Srzednicki.
2118 TF/05 Leading white space used to be stripped from $spam_report which
2119 wrecked the formatting. Now it is preserved.
2121 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
2122 that they are available at delivery time.
2124 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
2126 TF/08 TLS error reporting now respects the incoming_interface and
2127 incoming_port log selectors.
2129 TF/09 Produce a more useful error message if an SMTP transport's hosts
2130 setting expands to an empty string.
2132 NM/06 Bugzilla 744: EXPN did not work under TLS.
2133 Patch provided by Phil Pennock.
2135 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
2136 Patch provided by Richard Godbee.
2138 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
2139 acl_smtp_notquit, added index entry.
2141 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
2142 Patch provided by Eugene Bujak.
2144 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
2145 accept(). Patch provided by Maxim Dounin.
2147 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
2148 Patch provided by Phil Pennock.
2150 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
2152 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
2153 Patch provided by Brad "anomie" Jorsch.
2155 NM/14 Bugzilla 622: Added timeout setting to transport filter.
2156 Patch provided by Dean Brooks.
2158 TK/05 Add native DKIM support (does not depend on external libraries).
2160 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
2161 Patch provided by Graeme Fowler.
2163 NM/16 Bugzilla 851: Documentation example syntax fix.
2165 NM/17 Changed NOTICE file to remove references to embedded PCRE.
2167 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
2170 NM/19 Bugzilla 745: TLS version reporting.
2171 Patch provided by Phil Pennock.
2173 NM/20 Bugzilla 167: bool: condition support.
2174 Patch provided by Phil Pennock.
2176 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
2177 clients. Patch provided by Phil Pennock.
2179 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
2180 Patch provided by Brad "anomie" Jorsch.
2182 NM/23 Bugzilla 687: Fix misparses in eximstats.
2183 Patch provided by Heiko Schlittermann.
2185 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
2186 Patch provided by Heiko Schlittermann.
2188 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
2189 plus update to original patch.
2191 NM/26 Bugzilla 799: Documentation correction for ratelimit.
2193 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
2194 Patch provided by David Brownlee.
2196 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
2198 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
2200 NM/30 Bugzilla 888: TLS documentation bugfixes.
2202 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
2204 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
2205 Unlike the original bugzilla I have changed all shell scripts in src tree.
2207 NM/33 Bugzilla 898: Transport filter timeout fix.
2208 Patch by Todd Rinaldo.
2210 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
2211 Patch by Serge Demonchaux.
2213 NM/35 Bugzilla 39: Base64 decode bug fixes.
2214 Patch by Jakob Hirsch.
2216 NM/36 Bugzilla 909: Correct connect() call in dcc code.
2218 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
2220 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
2222 NM/39 Bugzilla 911: Fixed MakeLinks build script.
2228 TK/01 Add preliminary DKIM support. Currently requires a forked version of
2229 ALT-N's libdkim that I have put here:
2230 http://duncanthrax.net/exim-experimental/
2232 Note to Michael Haardt: I had to rename some vars in sieve.c. They
2233 were called 'true' and it seems that C99 defines that as a reserved
2234 keyword to be used with 'bool' variable types. That means you could
2235 not include C99-style headers which use bools without triggering
2236 build errors in sieve.c.
2238 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
2239 as mailq or other aliases. Changed the --help handling significantly
2240 to do whats expected. exim_usage() emits usage/help information.
2242 SC/01 Added the -bylocaldomain option to eximstats.
2244 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
2246 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
2248 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
2254 PH/01 Another patch from the Sieve maintainer.
2256 PH/02 When an IPv6 address is converted to a string for single-key lookup
2257 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
2258 dots are used instead of colons so that keys in lsearch files need not
2259 contain colons. This was done some time before quoting was made available
2260 in lsearch files. However, iplsearch files do require colons in IPv6 keys
2261 (notated using the quote facility) so as to distinguish them from IPv4
2262 keys. This meant that lookups for IP addresses in host lists did not work
2263 for iplsearch lookups.
2265 This has been fixed by arranging for IPv6 addresses to be expressed with
2266 colons if the lookup type is iplsearch. This is not incompatible, because
2267 previously such lookups could never work.
2269 The situation is now rather anomalous, since one *can* have colons in
2270 ordinary lsearch keys. However, making the change in all cases is
2271 incompatible and would probably break a number of configurations.
2273 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
2276 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
2277 conversion specification without a maximum field width, thereby enabling
2278 a rogue spamd server to cause a buffer overflow. While nobody in their
2279 right mind would setup Exim to query an untrusted spamd server, an
2280 attacker that gains access to a server running spamd could potentially
2281 exploit this vulnerability to run arbitrary code as the Exim user.
2283 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
2284 $primary_hostname instead of what libspf2 thinks the hosts name is.
2286 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
2287 a directory entry by the name of the lookup key. Previously, if a
2288 symlink pointed to a non-existing file or a file in a directory that
2289 Exim lacked permissions to read, a lookup for a key matching that
2290 symlink would fail. Now it is enough that a matching directory entry
2291 exists, symlink or not. (Bugzilla 503.)
2293 PH/03 The body_linecount and body_zerocount variables are now exported in the
2296 PH/04 Added the $dnslist_matched variable.
2298 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
2299 This means they are set thereafter only if the connection becomes
2302 PH/06 Added the client_condition to authenticators so that some can be skipped
2303 by clients under certain conditions.
2305 PH/07 The error message for a badly-placed control=no_multiline_responses left
2306 "_responses" off the end of the name.
2308 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
2310 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
2311 (without spaces) instead of just copying the configuration text.
2313 PH/10 Added the /noupdate option to the ratelimit ACL condition.
2315 PH/11 Added $max_received_linelength.
2317 PH/12 Added +ignore_defer and +include_defer to host lists.
2319 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
2320 way in which PCRE > 7.0 is built.
2322 PH/14 Implemented queue_only_load_latch.
2324 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
2325 MAIL command. The effect was to mangle the value on 64-bit systems.
2327 PH/16 Another patch from the Sieve maintainer.
2329 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
2331 PH/18 If a system quota error occurred while trying to create the file for
2332 a maildir delivery, the message "Mailbox is full" was not appended to the
2333 bounce if the delivery eventually timed out. Change 4.67/27 below applied
2334 only to a quota excession during the actual writing of the file.
2336 PH/19 It seems that peer DN values may contain newlines (and other non-printing
2337 characters?) which causes problems in log lines. The DN values are now
2338 passed through string_printing() before being added to log lines.
2340 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
2341 and InterBase are left for another time.)
2343 PH/21 Added message_body_newlines option.
2345 PH/22 Guard against possible overflow in moan_check_errorcopy().
2347 PH/23 POSIX allows open() to be a macro; guard against that.
2349 PH/24 If the recipient of an error message contained an @ in the local part
2350 (suitably quoted, of course), incorrect values were put in $domain and
2351 $local_part during the evaluation of errors_copy.
2357 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
2358 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
2361 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
2362 issue a MAIL command.
2364 PH/02 In an ACL statement such as
2366 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
2368 if a client was not listed at all, or was listed with a value other than
2369 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
2370 the condition was not true (as it should be), so access was not denied.
2371 The bug was that the ! inversion was incorrectly passed on to the second
2372 item. This has been fixed.
2374 PH/03 Added additional dnslists conditions == and =& which are different from
2375 = and & when the dns lookup returns more than one IP address.
2377 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
2378 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
2380 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
2381 FSYNC, which compiles an option called disable_fsync that allows for
2382 bypassing fsync(). The documentation is heavily laced with warnings.
2384 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
2386 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
2387 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
2388 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
2389 including adding "make clean"; (3) Added -fPIC when compiling the test
2390 dynamically loaded module, to get rid of a warning.
2392 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
2393 message fails, move_frozen_messages = true and ignore_bounce_errors_after
2394 = 0s. The bug is otherwise harmless.
2396 PH/07 There was a bug in the dovecot authenticator such that the value of
2397 $auth1 could be overwritten, and so not correctly preserved, after a
2398 successful authentication. This usually meant that the value preserved by
2399 the server_setid option was incorrect.
2401 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
2403 PH/09 Installed PCRE release 7.0.
2405 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
2406 run for batched SMTP input. It is now run at the start of every message
2407 in the batch. While fixing this I discovered that the process information
2408 (output by running exiwhat) was not always getting set for -bs and -bS
2409 input. This is fixed, and it now also says "batched" for BSMTP.
2411 PH/11 Added control=no_pipelining.
2413 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
2414 patch, slightly modified), and move the expansion of helo_data till after
2415 the connection is made in the smtp transport (so it can use these
2418 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
2420 PH/14 Added log_selector = +pid.
2422 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
2424 PH/16 Add ${if forany and ${if forall.
2426 PH/17 Added dsn_from option to vary the From: line in DSNs.
2428 PH/18 Flush SMTP output before performing a callout, unless control =
2429 no_callout_flush is set.
2431 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
2432 was true (the default) a successful delivery failed to delete the retry
2433 item, thus causing premature timeout of the address. The bug is now
2436 PH/20 Added hosts_avoid_pipelining to the smtp transport.
2438 PH/21 Long custom messages for fakedefer and fakereject are now split up
2439 into multiline responses in the same way that messages for "deny" and
2440 other ACL rejections are.
2442 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
2443 with slight modification.
2445 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
2446 draft, changing the syntax and factoring some duplicate code".
2448 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
2449 for deliveries of the second and subsequent messages over the same SMTP
2452 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
2453 ${reduce, with only minor "tidies".
2455 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
2457 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
2458 expansion side effects.
2460 PH/27 When a message times out after an over-quota error from an Exim-imposed
2461 quota, the bounce message says "mailbox is full". This message was not
2462 being given when it was a system quota that was exceeded. It now should
2465 MH/03 Made $recipients available in local_scan(). local_scan() already has
2466 better access to the recipient list through recipients_list[], but
2467 $recipients can be useful in postmaster-provided expansion strings.
2469 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
2470 in the case of a MAIL command with additional options following the
2471 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
2472 were accidentally chopped off.
2474 PH/29 SMTP synchronization checks are implemented when a command is read -
2475 there is a check that no more input is waiting when there shouldn't be
2476 any. However, for some commands, a delay in an ACL can mean that it is
2477 some time before the response is written. In this time, more input might
2478 arrive, invalidly. So now there are extra checks after an ACL has run for
2479 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
2480 pipelining has not been advertised.
2482 PH/30 MH's patch to allow iscntrl() characters to be list separators.
2484 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
2485 returned in the SMTP response when smtp_return_error_details was false.
2486 This has been fixed.
2488 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
2489 instead of the C I/O that was originally supplied, because problems were
2490 reported on Solaris.
2492 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
2493 Exim which did not show up earlier: it was assuming that a call to
2494 SSL_CTX_set_info_callback() might give an error value. In fact, there is
2495 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
2496 was a macro that became an assignment, so it seemed to work. This has
2497 changed to a proper function call with a void return, hence the compile
2498 error. Exim's code has been fixed.
2500 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
2503 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
2505 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
2507 PH/37 If a message is not accepted after it has had an id assigned (e.g.
2508 because it turns out to be too big or there is a timeout) there is no
2509 "Completed" line in the log. When some messages of this type were
2510 selected by exigrep, they were listed as "not completed". Others were
2511 picked up by some special patterns. I have improved the selection
2512 criteria to be more general.
2514 PH/38 The host_find_failed option in the manualroute router can now be set
2515 to "ignore", to completely ignore a host whose IP address cannot be
2516 found. If all hosts are ignored, the behaviour is controlled by the new
2517 host_all_ignored option.
2519 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
2520 homing or because of multiple MX records with /mx) generated more than
2521 one IP address, and the following item turned out to be the local host,
2522 all the secondary addresses of the first item were incorrectly removed
2523 from the list, along with the local host and any following hosts (which
2524 is what is supposed to happen).
2526 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
2527 whoever called Exim into the -H file. In the case of the daemon it was
2528 behaving confusingly. When first started, it used values for whoever
2529 started the daemon, but after a SIGHUP it used the Exim user (because it
2530 calls itself on a restart). I have changed the code so that it now always
2533 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
2534 message are rejected with the same error (e.g. no authentication or bad
2535 sender address), and a DATA command is nevertheless sent (as can happen
2536 with PIPELINING or a stupid MUA), the error message that was given to the
2537 RCPT commands is included in the rejection of the DATA command. This is
2538 intended to be helpful for MUAs that show only the final error to their
2541 PH/42 Another patch from the Sieve maintainer.
2543 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
2544 Eximstats - Fixed some broken HTML links and added missing column headers
2546 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
2547 columns for Rejects, Temp Rejects, Ham, and Spam rows.
2549 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
2551 PH/43 Yet another patch from the Sieve maintainer.
2553 PH/44 I found a way to check for a TCP/IP connection going away before sending
2554 the response to the final '.' that terminates a message, but only in the
2555 case where the client has not sent further data following the '.'
2556 (unfortunately, this is allowed). However, in many cases there won't be
2557 any further data because there won't be any more messages to send. A call
2558 to select() can be used: if it shows that the input is "ready", there is
2559 either input waiting, or the socket has been closed. An attempt to read
2560 the next input character can distinguish the two cases. Previously, Exim
2561 would have sent an OK response which the client would never have see.
2562 This could lead to message repetition. This fix should cure that, at
2563 least in a lot of common cases.
2565 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
2566 advertised in response to EHLO.
2572 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
2573 fixed by 4.65/MH/01 (is this a record?) are fixed:
2575 (i) An empty string was always treated as zero by the numeric comparison
2576 operators. This behaviour has been restored.
2578 (ii) It is documented that the numeric comparison operators always treat
2579 their arguments as decimal numbers. This was broken in that numbers
2580 starting with 0 were being interpreted as octal.
2582 While fixing these problems I realized that there was another issue that
2583 hadn't been noticed. Values of message_size_limit (both the global option
2584 and the transport option) were treated as octal if they started with 0.
2585 The documentation was vague. These values are now always treated as
2586 decimal, and I will make that clear in the documentation.
2592 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
2593 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
2596 MH/01 Don't check that the operands of numeric comparison operators are
2597 integers when their expansion is in "skipping" mode (fixes bug
2598 introduced by 4.64-PH/07).
2600 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
2601 child addresses, Exim now panics and dies. Previously, because the count
2602 is held in a short int, deliveries were likely to be lost. As such a
2603 large number of recipients for a single message is ridiculous
2604 (performance will be very, very poor), I have chosen to impose a limit
2605 rather than extend the field.
2611 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
2612 leftover -K file (the existence of which was triggered by #402).
2613 While we were at it, introduced process PID as part of the -K
2614 filename. This should rule out race conditions when creating
2617 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
2618 processing considerably. Previous code took too long for large mails,
2619 triggering a timeout which in turn triggers #401.
2621 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
2622 in the DK code in transports.c. sendfile() is not really portable,
2623 hence the _LINUX specificness.
2625 TF/01 In the add_headers option to the mail command in an Exim filter,
2626 there was a bug that Exim would claim a syntax error in any
2627 header after the first one which had an odd number of characters
2630 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
2631 callout verification, Exim cached a "reject" for the entire domain. This
2632 is correct for most verifications, but it is not correct for a recipient
2633 verification with use_sender or use_postmaster set, because in that case
2634 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
2635 case of MAIL FROM:<> rejection from other early rejections (e.g.
2636 rejection of HELO). When verifying a recipient using a non-null MAIL
2637 address, the cache is ignored if it shows MAIL FROM:<> rejection.
2638 Whatever the result of the callout, the value of the domain cache is
2639 left unchanged (for any other kind of callout, getting as far as trying
2640 RCPT means that the domain itself is ok).
2642 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
2645 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
2646 manifest itself as EPIPE rather than ECONNECT. When tidying away a
2647 session, the daemon ignores ECONNECT errors and logs others; it now
2648 ignores EPIPE as well.
2650 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
2651 (quoted-printable decoding).
2653 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
2654 later the small subsequent patch to fix an introduced bug.
2656 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
2658 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
2660 PH/08 An error is now given if message_size_limit is specified negative.
2662 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
2663 to be given (somewhat) arbitrary names.
2665 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
2668 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
2669 miscellaneous code fixes
2671 PH/10 Added the log_reject_target ACL modifier to specify where to log
2674 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
2675 hostname. This is wrong, because it relates to the incoming message (and
2676 probably the interface on which it is arriving) and not to the outgoing
2677 callout (which could be using a different interface). This has been
2678 changed to use the value of the helo_data option from the smtp transport
2679 instead - this is what is used when a message is actually being sent. If
2680 there is no remote transport (possible with a router that sets up host
2681 addresses), $smtp_active_hostname is used.
2683 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
2684 tweaks were necessary in order to get it to work (see also 21 below):
2685 (a) The code assumed that strncpy() returns a negative number on buffer
2686 overflow, which isn't the case. Replaced with Exim's string_format()
2688 (b) There were several signed/unsigned issues. I just did the minimum
2689 hacking in of casts. There is scope for a larger refactoring.
2690 (c) The code used strcasecmp() which is not a standard C function.
2691 Replaced with Exim's strcmpic() function.
2692 (d) The code set only $1; it now sets $auth1 as well.
2693 (e) A simple test gave the error "authentication client didn't specify
2694 service in request". It would seem that Dovecot has changed its
2695 interface. Fortunately there's a specification; I followed it and
2696 changed what the client sends and it appears to be working now.
2698 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
2701 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
2702 address A is aliased to B and C, where B exists and C does not. Without
2703 -v the output is "A verified" because verification stops after a
2704 successful redirection if more than one address is generated. However,
2705 with -v the child addresses are also verified. Exim was outputting "A
2706 failed to verify" and then showing the successful verification for C,
2707 with its parentage. It now outputs "B failed to verify", showing B's
2708 parentage before showing the successful verification of C.
2710 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
2711 look up a TXT record in a specific list after matching in a combined
2714 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
2715 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
2716 they consult the DNS. I had assumed they would set it the way they
2717 wanted; and indeed my experiments on Linux seem to show that in some
2718 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
2719 To be on the safe side, however, I have now made the interface to
2720 host_find_byname() similar to host_find_bydns(), with an argument
2721 containing the DNS resolver options. The host_find_byname() function now
2722 sets these options at its start, just as host_find_bydns() does. The smtp
2723 transport options dns_qualify_single and dns_search_parents are passed to
2724 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
2725 of host_find_byname() use the default settings of RES_DEFNAMES
2726 (qualify_single) but not RES_DNSRCH (search_parents).
2728 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
2729 spool_read_header() do less string testing, by means of a preliminary
2730 switch on the second character of optional "-foo" lines. (This is
2731 overdue, caused by the large number of possibilities that now exist.
2732 Originally there were few.) While I was there, I also converted the
2733 str(n)cmp tests so they don't re-test the leading "-" and the first
2734 character, in the hope this might squeeze out yet more improvement.
2736 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
2737 flag allowing group syntax was set by the header_syntax check but not
2738 turned off, possible causing trouble later; (2) The flag was not being
2739 set at all for the header_verify test, causing "group"-style headers to
2740 be rejected. I have now set it in this case, and also caused header_
2741 verify to ignore an empty address taken from a group. While doing this, I
2742 came across some other cases where the code for allowing group syntax
2743 while scanning a header line wasn't quite right (mostly, not resetting
2744 the flag correctly in the right place). These bugs could have caused
2745 trouble for malformed header lines. I hope it is now all correct.
2747 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
2748 with the "reply" argument non-NULL. The code, however (which originally
2749 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
2750 but it didn't always do it. This confused somebody who was copying the
2751 code for some other use. I have removed all the tests.
2753 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
2754 feature that was used to support insecure browsers during the U.S. crypto
2755 embargo. It requires special client support, and Exim is probably the
2756 only MTA that supported it -- and would never use it because real RSA is
2757 always available. This code has been removed, because it had the bad
2758 effect of slowing Exim down by computing (never used) parameters for the
2759 RSA_EXPORT functionality.
2761 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
2762 authenticator to fail if there's a tab character in the incoming data
2763 (there should never be unless someone is messing about, as it's supposed
2764 to be base64-encoded). Also added, on Timo's advice, the "secured" option
2765 if the connection is using TLS or if the remote IP is the same as the
2766 local IP, and the "valid-client-cert option" if a client certificate has
2769 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
2770 authenticators. This can be used for authorization after authentication
2771 succeeds. (In the case of plaintext, it servers for both authentication
2774 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
2775 if any retry times were supplied.
2777 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
2778 connection, where there is no client IP address to check. In this
2779 situation, the verify now always succeeds.
2781 PH/25 Applied John Jetmore's -Mset patch.
2783 PH/26 Added -bem to be like -Mset, but loading a message from a file.
2785 PH/27 In a string expansion for a processed (not raw) header when multiple
2786 headers of the same name were present, leading whitespace was being
2787 removed from all of them, but trailing whitespace was being removed only
2788 from the last one. Now trailing whitespace is removed from each header
2789 before concatenation. Completely empty headers in a concatenation (as
2790 before) are ignored.
2792 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
2793 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
2795 PH/29 [Removed. This was a change that I later backed out, and forgot to
2796 correct the ChangeLog entry (that I had efficiently created) before
2797 committing the later change.]
2799 PH/30 Exim was sometimes attempting to deliver messages that had suffered
2800 address errors (4xx response to RCPT) over the same connection as other
2801 messages routed to the same hosts. Such deliveries are always "forced",
2802 so retry times are not inspected. This resulted in far too many retries
2803 for the affected addresses. The effect occurred only when there were more
2804 hosts than the hosts_max_try setting in the smtp transport when it had
2805 the 4xx errors. Those hosts that it had tried were not added to the list
2806 of hosts for which the message was waiting, so if all were tried, there
2807 was no problem. Two fixes have been applied:
2809 (i) If there are any address or message errors in an SMTP delivery, none
2810 of the hosts (tried or untried) are now added to the list of hosts
2811 for which the message is waiting, so the message should not be a
2812 candidate for sending over the same connection that was used for a
2813 successful delivery of some other message. This seems entirely
2814 reasonable: after all the message is NOT "waiting for some host".
2815 This is so "obvious" that I'm not sure why it wasn't done
2816 previously. Hope I haven't missed anything, but it can't do any
2817 harm, as the worst effect is to miss an optimization.
2819 (ii) If, despite (i), such a delivery is accidentally attempted, the
2820 routing retry time is respected, so at least it doesn't keep
2821 hammering the server.
2823 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
2824 in ${readsocket because some servers need this prod.
2826 PH/32 Added some extra debug output when updating a wait-xxx database.
2828 PH/33 The hint "could be header name not terminated by colon", which has been
2829 given for certain expansion errors for a long time, was not being given
2830 for the ${if def:h_colon_omitted{... case.
2832 PH/34 The spec says: "With one important exception, whenever a domain list is
2833 being scanned, $domain contains the subject domain." There was at least
2834 one case where this was not true.
2836 PH/35 The error "getsockname() failed: connection reset by peer" was being
2837 written to the panic log as well as the main log, but it isn't really
2838 panic-worthy as it just means the connection died rather early on. I have
2839 removed the panic log writing for the ECONNRESET error when getsockname()
2842 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
2843 runs only) independently of the message's sender address. This meant
2844 that, if the 4xx error was in fact related to the sender, a different
2845 message to the same recipient with a different sender could confuse
2846 things. In particular, this can happen when sending to a greylisting
2847 server, but other circumstances could also provoke similar problems.
2848 I have changed the default so that the retry time for these errors is now
2849 based a combination of the sender and recipient addresses. This change
2850 can be overridden by setting address_retry_include_sender=false in the
2853 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
2854 remote server are returned as part of bounce messages. This was not
2855 happening for LMTP over a pipe (the lmtp transport), but now it is the
2856 same for both kinds of LMTP.
2858 PH/38 Despite being documented as not happening, Exim was rewriting addresses
2859 in header lines that were in fact CNAMEs. This is no longer the case.
2861 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
2862 and queue runs started by the daemon processed all messages. This has
2863 been fixed so that -R and -S can now usefully be given with -q<time>.
2865 PH/40 Import PCRE release 6.7 (fixes some bugs).
2867 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
2869 PH/42 Give an error if -q is specified more than once.
2871 PH/43 Renamed the variables $interface_address and $interface_port as
2872 $received_ip_address and $received_port, to make it clear that these
2873 values apply to message reception, and not to the outgoing interface when
2874 a message is delivered. (The old names remain recognized, of course.)
2876 PH/44 There was no timeout on the connect() call when using a Unix domain
2877 socket in the ${readsocket expansion. There now is.
2879 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
2880 be meaningful with "accept".
2882 SC/01 Eximstats V1.43
2883 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
2885 SC/02 Eximstats V1.44
2886 Use a glob alias rather than an array ref in the generated
2887 parser. This improves both readability and performance.
2889 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
2890 Collect SpamAssassin and rejection statistics.
2891 Don't display local sender or destination tables unless
2892 there is data to show.
2893 Added average volumes into the top table text output.
2895 SC/04 Eximstats V1.46
2896 Collect data on the number of addresses (recipients)
2897 as well as the number of messages.
2899 SC/05 Eximstats V1.47
2900 Added 'Message too big' to the list of mail rejection
2901 reasons (thanks to Marco Gaiarin).
2903 SC/06 Eximstats V1.48
2904 Mainlog lines which have GMT offsets and are too short to
2905 have a flag are now skipped.
2907 SC/07 Eximstats V1.49 (Alain Williams)
2908 Added the -emptyok flag.
2910 SC/08 Eximstats V1.50
2911 Fixes for obtaining the IP address from reject messages.
2913 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
2914 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
2915 whitespace changes from 4.64-PH/27
2917 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
2920 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
2921 are found, allow negative numbers in numeric criteria)
2923 JJ/06 exipick.20061117.2, added new $message_body_missing variable
2925 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
2926 to match changes made in 4.64-PH/43
2928 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
2930 PH/47 Put in an explicit test for a DNS lookup of an address record where the
2931 "domain" is actually an IP address, and force a failure. This locks out
2932 those revolvers/nameservers that support "A-for-A" lookups, in
2933 contravention of the specifications.
2935 PH/48 When a host name was looked up from an IP address, and the subsequent
2936 forward lookup of the name timed out, the host name was left in
2937 $sender_host_name, contrary to the specification.
2939 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
2940 restricted to single-key lookups, Exim was not diagnosing an error if
2941 * or *@ was used with a query-style lookup.
2943 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
2945 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
2946 long ago, but noone interested enough thought of it. Let's just say that
2947 the "1.1" means that there are some new functions that weren't there at
2948 some point in the past.
2950 PH/51 Error processing for expansion failure of helo_data from an smtp
2951 transport during callout processing was broken.
2953 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
2954 tested/used via the -bh/-bhc/-bs options.
2956 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
2957 bug, fixed in subsequent PCRE releases).
2959 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
2960 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
2962 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
2968 SC/01 Use a glob alias rather than an array ref in eximstats generated
2969 parser. This improves both readability and performance.
2971 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
2972 Don't display local sender or destination tables in eximstats unless
2973 there is data to show.
2974 Added average volumes into the eximstats top table text output.
2976 SC/03 Collect data on the number of addresses (recipients) as well
2977 as the number of messages in eximstats.
2979 TF/01 Correct an error in the documentation for the redirect router. Exim
2980 does (usually) call initgroups() when daemonizing.
2982 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
2983 with consistent privilege compared to when running as a daemon.
2985 TF/03 Note in the spec that $authenticated_id is not set for local
2986 submissions from trusted users.
2988 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
2989 Thanks to Dean Brooks <dean@iglou.com> for the patch.
2991 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
2992 by adding some example configuration directives to the default
2993 configuration file. A little bit of work is required to uncomment the
2994 directives and define how usernames and passwords are checked, but
2995 there is now a framework to start from.
2997 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
2998 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
2999 without this. I don't know how relevant this is to other LDAP libraries.
3001 PH/02 Add the verb name to the "unknown ACL verb" error.
3003 PH/03 Magnus Holmgren's patch for filter_prepend_home.
3005 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
3007 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
3008 directory not expanded when it should be if an expanded home directory
3009 was set for the address (which is overridden by the transport).
3011 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
3014 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
3015 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
3016 because it is too late at that time, and has no effect.
3018 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
3019 security issue with \' (bugzilla #107). I could not use the
3020 PQescapeStringConn() function, because it needs a PGconn value as one of
3023 PH/08 When testing addresses using -bt, indicate those final addresses that
3024 are duplicates that would not cause an additional delivery. At least one
3025 person was confused, thinking that -bt output corresponded to deliveries.
3026 (Suppressing duplicates isn't a good idea as you lose the information
3027 about possibly different redirections that led to the duplicates.)
3029 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
3030 systems where poll() doesn't work, in particular OS X.
3032 PH/10 Added more information to debugging output for retry time not reached.
3034 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
3035 operations in malware.c.
3037 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
3040 PH/13 If write_rejectlog was set false when logging was sent to syslog with
3041 syslog_duplication set false, log lines that would normally be written
3042 both the the main log and to the reject log were not written to syslog at
3045 PH/14 In the default configuration, change the use of "message" in ACL warn
3046 statements to "add_header".
3048 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
3049 not followed by a command (e.g. "seen endif").
3051 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
3052 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
3055 PH/17 Added extra conditions to the default value of delay_warning_condition
3059 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
3060 { match{$h_precedence:}{(?i)bulk|list|junk} } \
3061 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
3064 The Auto-Submitted: and various List- headers are standardised, whereas I
3065 don't think Precedence: ever was.
3067 PH/18 Refactored debugging code in route_finduser() to show more information,
3068 in particular, the error code if getpwnam() issues one.
3070 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
3071 This is apparently needed in addition to the PH/07 change above to avoid
3072 any possible encoding problems.
3074 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
3075 but not after initializing Perl.
3077 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
3078 output them only if debugging. By default they are written stderr,
3079 apparently, which is not desirable.
3081 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
3084 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
3087 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
3089 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
3090 authenticated or an ident call has been made. Suppress the default
3091 values for $authenticated_id and $authenticated_sender (but permit -oMai
3092 and -oMas) when testing with -bh.
3094 PH/24 Re-jigged the order of the tests in the default configuration so that the
3095 tests for valid domains and recipients precede the DNS black list and CSA
3096 tests, on the grounds that those ones are more expensive.
3098 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
3099 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
3100 This bug exists in every version of Exim that I still have, right back to
3103 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
3104 However, an attempt to turn on case-sensitivity in a regex key by
3105 including (?-i) didn't work because the subject string was already
3106 lowercased, and the effects were non-intuitive. It turns out that a
3107 one-line patch can be used to allow (?-i) to work as expected.
3113 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
3114 other effects) broke the use of negated acl sub-conditions.
3116 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
3119 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
3120 "Deny" causes Exim to reject the incoming connection with a 554 error.
3121 Unfortunately, if there is a major crisis, such as a disk failure,
3122 tcp-wrappers gives "deny", whereas what one would like would be some
3123 kind of temporary error. A kludge has been added to help with this.
3124 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
3125 554 error is used if errno is still zero or contains ENOENT (which occurs
3126 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
3129 PH/03 Add -lutil to the default FreeBSD LIBS setting.
3131 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
3132 errors. Otherwise a message that provokes a temporary error (when other
3133 messages do not) can cause a whole host to time out.
3135 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
3136 addresses were routed directly to files or pipes from a redirect router.
3137 File deliveries just didn't batch; pipe deliveries might have suffered
3140 PH/06 A failure to get a lock for a hints database would erroneously always say
3141 "Failed to get write lock", even when it was really a read lock.
3143 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
3144 of 0600. This has been changed to use the value of the lockfile_mode
3145 option (which defaults to 0600).
3147 PH/08 Applied small patch from the Sieve maintainer.
3149 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
3150 folder from quota calculations, a direct delivery into this folder messed
3151 up the contents of the maildirsize file. This was because the regex was
3152 used only to exclude .Trash (or whatever) when the size of the mailbox
3153 was calculated. There was no check that a delivery was happening into an
3154 excluded directory. This bug has been fixed by ignoring all quota
3155 processing for deliveries into excluded directories.
3157 PH/10 Added the maildirfolder_create_regex option to appendfile.
3163 PH/01 The code for finding all the local interface addresses on a FreeBSD
3164 system running IPv6 was broken. This may well have applied to all BSD
3165 systems, as well as to others that have similar system calls. The broken
3166 code found IPv4 interfaces correctly, but gave incorrect values for the
3167 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
3168 that it would not match correctly against @[] and not recognize the IPv6
3171 PH/02 The ipliteral router was not recognizing addresses of the form user@
3172 [ipv6:....] because it didn't know about the "ipv6:" prefix.
3174 PH/03 Added disable_ipv6.
3176 PH/04 Changed $reply_address to use the raw form of the headers instead of the
3177 decoded form, because it is most often used to construct To: headers
3178 lines in autoreplies, and the decoded form may well be syntactically
3179 invalid. However, $reply_address has leading white space removed, and all
3180 newlines turned into spaces so that the autoreply transport does not
3183 PH/05 If group was specified without a user on a router, and no group or user
3184 was specified on a transport, the group from the router was ignored.
3186 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
3187 for visible compile-time settings that can be used to change these
3188 numbers, for those that want even more. Backwards compatibility with old
3189 spool files has been maintained. However, going back to a previous Exim
3190 release will lost any variables that are in spool files.
3192 PH/07 Two small changes when running in the test harness: increase delay when
3193 passing a TCP/IP connection to a new process, in case the original
3194 process has to generate a bounce, and remove special handling of
3195 127.0.0.2 (sic), which is no longer necessary.
3197 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
3198 be the same on different OS.
3200 PH/09 Moved a debug statement in filter processing to avoid a race problem when
3203 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
3204 whether --show-vars was specified or not
3206 JJ/02 exipick: Added support for new ACL variable spool format introduced
3209 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
3210 syntactically invalid From: or Reply-to: line, and a filter used this to
3211 generate an autoreply, and therefore failed to obtain an address for the
3212 autoreply, Exim could try to deliver to a non-existent relative file
3213 name, causing unrelated and misleading errors. What now happens is that
3214 it logs this as a hard delivery error, but does not attempt to create a
3217 PH/11 The exinext utility has a -C option for testing purposes, but although
3218 the given file was scanned by exinext itself; it wasn't being passed on
3219 when Exim was called.
3221 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
3222 an end-of-file indication when reading a command response.
3224 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
3225 compiled. In many other places in Exim, IPv6 addresses are always
3226 recognized, so I have changed this. It also means that IPv4 domain
3227 literals of the form [IPV4:n.n.n.n] are now always recognized.
3229 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
3230 used if the router is not running as root, for example, when verifying at
3231 ACL time, or when using -bh. The debugging output from this situation was
3232 non-existent - all you got was a failure to exec. I have made two
3235 (a) Failures to set uid/gid, the current directory, or a process leader
3236 in a subprocess such as that created by queryprogram now generate
3237 suitable debugging output when -d is set.
3239 (b) The queryprogram router detects when it is not running as root,
3240 outputs suitable debugging information if -d is set, and then runs
3241 the subprocess without attempting to change uid/gid.
3243 PH/15 Minor change to Makefile for building test_host (undocumented testing
3246 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
3247 additional section of a DNS packet that returns MX or SRV records.
3248 Instead, it always explicitly searches for A/AAAA records. This avoids
3249 major problems that occur when a DNS server includes only records of one
3250 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
3251 fixed another bug: if SRV records were looked up and the corresponding
3252 address records were *not* found in the additional section, the port
3253 values from the SRV records were lost.
3255 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
3256 using the correct key (the original address) when searching the retry
3257 rules in order to find which one to use for generating the retry hint.
3259 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
3260 adding the default one. Similarly, if it contains a Reply-To: header, the
3261 errors_reply_to option, if set, is not used.
3263 PH/19 When calculating a retry time, Exim used to measure the "time since
3264 failure" by looking at the "first failed" field in the retry record. Now
3265 it does not use this if it is later than than the arrival time of the
3266 message. Instead it uses the arrival time. This makes for better
3267 behaviour in cases where some deliveries succeed, thus re-setting the
3268 "first failed" field. An example is a quota failure for a huge message
3269 when small messages continue to be delivered. Without this change, the
3270 "time since failure" will always be short, possible causing more frequent
3271 delivery attempts for the huge message than are intended.
3272 [Note: This change was subsequently modified - see PH/04 for 4.62.]
3274 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
3275 $1, $2, $3) because the numerical variables can be reset during some
3276 expansion items (e.g. "match"), thereby losing the authentication data.
3278 PH/21 Make -bV show the size of off_t variables so that the test suite can
3279 decide whether to run tests for quotas > 2G.
3281 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
3282 mailbox_size, and mailbox_filecount in the appendfile transport. If a
3283 filecount value is greater than 2G or if a quota value is greater than 2G
3284 on a system where the size of off_t is not greater than 4, a panic error
3287 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
3288 never match. The debug and -bh output now contains an explicit error
3289 message indicating a malformed IPv4 address or mask.
3291 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
3292 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
3293 PH/23 above applies.
3295 PH/25 Do not write to syslog when running in the test harness. The only
3296 occasion when this arises is a failure to open the main or panic logs
3297 (for which there is an explicit test).
3299 PH/26 Added the /no_tell option to "control=freeze".
3301 PH/27 If a host name lookup failed very early in a connection, for example, if
3302 the IP address matched host_lookup and the reverse lookup yielded a name
3303 that did not have a forward lookup, an error message of the form "no IP
3304 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
3305 could be logged. Now it outputs the IP address instead of "NULL".
3307 PH/28 An enabling patch from MH: add new function child_open_exim2() which
3308 allows the sender and the authenticated sender to be set when
3309 submitting a message from within Exim. Since child_open_exim() is
3310 documented for local_scan(), the new function should be too.
3312 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
3313 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
3314 results in an empty string is now treated as unset.
3316 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
3318 PH/31 Added sender_verify_fail logging option.
3320 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
3321 needed by RFC 822 but not by RFC 2822 was commented out. I have now
3322 tidied the source and removed it altogether.
3324 PH/33 When a queue run was abandoned because the load average was too high, a
3325 log line was always written; now it is written only if the queue_run log
3326 selector is set. In addition, the log line for abandonment now contains
3327 information about the queue run such as the pid. This is always present
3328 in "start" and "stop" lines but was omitted from the "abandon" line.
3330 PH/34 Omit spaces between a header name and the colon in the error message that
3331 is given when verify = headers_syntax fails (if there are lots of them,
3332 the message gets confusing).
3334 PH/35 Change the default for dns_check_names_pattern to allow slashes within
3335 names, as there are now some PTR records that contain slashes. This check
3336 is only to protect against broken name servers that fall over on strange
3337 characters, so the fact that it applies to all lookups doesn't matter.
3339 PH/36 Now that the new test suite is complete, we can remove some of the
3340 special code in Exim that was needed for the old test suite. For example,
3341 sorting DNS records because real resolvers return them in an arbitrary
3342 order. The new test suite's fake resolver always returns records in the
3345 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
3346 bounces) except when queue_only is set, to avoid logging races between
3347 the different processes.
3349 PH/38 Panic-die if .include specifies a non-absolute path.
3351 PH/39 A tweak to the "H" retry rule from its user.
3353 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
3354 a label. They prevented compilation on older perls.
3356 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
3357 a warning to be raised on newish perls.
3359 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
3360 on queue. Changes to match documented behaviour of showing count of
3361 messages matching specified criteria.
3363 PH/40 Changed the default ident timeout from 30s to 5s.
3365 PH/41 Added support for the use of login_cap features, on those BSD systems
3366 that have them, for controlling the resources used by pipe deliveries.
3368 PH/42 The content-scanning code uses fopen() to create files in which to put
3369 message data. Previously it was not paying any attention to the mode of
3370 the files. Exim runs with umask(0) because the rest of the code creates
3371 files with open(), and sets the required mode explicitly. Thus, these
3372 files were ending up world-writeable. This was not a big issue, because,
3373 being within the spool directory, they were not world-accessible. I have
3374 created a function called modefopen, which takes an additional mode
3375 argument. It sets umask(777), creates the file, chmods it to the required
3376 mode, then resets the umask. All the relevant calls to fopen() in the
3377 content scanning code have been changed to use this function.
3379 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
3380 to 24 hours. This avoids potential overflow problems when processing G
3381 and H retry rules. I suspect nobody ever tinkers with this value.
3383 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
3385 PH/45 When the plaintext authenticator is running as a client, the server's
3386 challenges are checked to ensure they are valid base64 strings. By
3387 default, the authentication attempt is cancelled if an invalid string is
3388 received. Setting client_ignore_invalid_base64 true ignores these errors.
3389 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
3390 they are received. Thus, the responses can be made to depend on the
3391 challenges. If an invalid string is ignored, an empty string is placed in
3394 PH/46 Messages that are created by the autoreply transport now contains a
3395 References: header, in accordance with RFCs 2822 and 3834.
3397 PH/47 Added authenticated_sender_force to the smtp transport.
3399 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
3401 PH/49 Installed latest patch from the Sieve maintainer.
3403 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
3404 was also set, the appendfile transport was unnecessarily scanning a
3405 directory of message files (e.g. for maildir delivery) to find the count
3406 of files (along with the size), even though it did not need this
3407 information. It now does the scan only if it needs to find either the
3408 size of the count of files.
3410 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
3412 PH/52 Two bugs concerned with error handling when the smtp transport is
3415 (i) Exim was not creating retry information for temporary errors given
3416 for individual recipients after the DATA command when the smtp transport
3417 was used in LMTP mode. This meant that they could be retried too
3418 frequently, and not timed out correctly.
3420 (ii) Exim was setting the flag that allows error details to be returned
3421 for LMTP errors on RCPT commands, but not for LMTP errors for individual
3422 recipients that were returned after the DATA command.
3424 PH/53 This is related to PH/52, but is more general: for any failing address,
3425 when detailed error information was permitted to be returned to the
3426 sender, but the error was temporary, then after the final timeout, only
3427 "retry timeout exceeded" was returned. Now it returns the full error as
3428 well as "retry timeout exceeded".
3430 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
3431 do this, and (what is worse) MTAs that accept it.
3433 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
3434 will now be deprecated.
3436 PH/56 New os.c-cygwin from the Cygwin maintainer.
3438 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
3439 formats (previously only available in exim formats via -bpr, -bpru,
3440 and -bpra. Now also available in native and exiqgrep formats)
3442 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
3443 with very large, slow to parse queues
3445 JJ/08 exipick: added ! as generic prefix to negate any criteria format
3447 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
3449 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
3450 responses to authentication challenges, though it was showing the
3451 challenges; (ii) I've removed the CR characters from the debug output for
3454 PH/58 Allow for the insertion of a newline as well as a space when a string
3455 is turned into more than one encoded-word during RFC 2047 encoding. The
3456 Sieve code now uses this.
3458 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
3459 data_4xx, lost_connection, tls_required.
3461 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
3462 message was being sent as an SMTP response.
3464 PH/61 Add -l and -k options to exicyclog.
3466 PH/62 When verifying, if an address was redirected to one new address, so that
3467 verification continued, and the new address failed or deferred after
3468 having set something in $address_data, the value of $address_data was not
3469 passed back to the ACL. This was different to the case when no
3470 redirection occurred. The value is now passed back in both cases.
3472 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
3473 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
3474 use login_cap.h, so on its own it isn't the distinguishing feature. The
3475 new name refers directly to the setclassresources() function.
3477 PH/65 Added configuration files for NetBSD3.
3479 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
3481 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
3482 is preferred over IPv4.
3484 PH/68 The bounce_return_message and bounce_return_body options were not being
3485 honoured for bounces generated during the reception of non-SMTP messages.
3486 In particular, this applied to messages rejected by the ACL. This bug has
3487 been fixed. However, if bounce_return_message is true and bounce_return_
3488 body is false, the headers that are returned for a non-SMTP message
3489 include only those that have been read before the error was detected.
3490 (In the case of an ACL rejection, they have all been read.)
3492 PH/69 The HTML version of the specification is now built in a directory called
3493 spec_html instead of spec.html, because the latter looks like a path with
3494 a MIME-type, and this confuses some software.
3496 PH/70 Catch two compiler warnings in sieve.c.
3498 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
3499 function verify_get_ident() calls ip_connect() to connect a socket, but
3500 if the "connect()" function timed out, ip_connect() used to close the
3501 socket. However, verify_get_ident() also closes the socket later, and in
3502 between Exim writes to the log, which may get opened at this point. When
3503 the socket was closed in ip_connect(), the log could get the same file
3504 descriptor number as the socket. This naturally causes chaos. The fix is
3505 not to close the socket in ip_connect(); the socket should be closed by
3506 the function that creates it. There was only one place in the code where
3507 this was missing, in the iplookup router, which I don't think anybody now
3508 uses, but I've fixed it anyway.
3510 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
3511 well as to direct DNS lookups. Otherwise the handling of names in host
3512 lists is inconsistent and therefore confusing.
3518 PH/01 Two changes to the default runtime configuration:
3520 (1) Move the checks for relay_from_hosts and authenticated clients from
3521 after to before the (commented out) DNS black list checks.
3523 (2) Add control=submission to the relay_from_hosts and authenticated
3524 clients checks, on the grounds that messages accepted by these
3525 statements are most likely to be submissions.
3527 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
3529 (1) Generate an error if the third argument for the ${prvs expansion is
3532 (2) Treat a missing third argument of ${prvscheck as if it were an empty
3535 (3) Reset the variables that are obtained from the first argument of
3536 ${prvscheck and used in the second argument before leaving the code,
3537 because their memory is reclaimed, so using them afterwards may do
3540 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
3541 one (it's much easier than Tom thought :-).
3543 (5) Because of (4), we can now allow for the use of $prvscheck_result
3544 inside the third argument.
3546 PH/03 For some reason, the default setting of PATH when running a command from
3547 a pipe transport was just "/usr/bin". I have changed it to
3550 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
3551 anything to be listed in the output from -bV.
3553 PH/05 When a filter generated an autoreply, the entire To: header line was
3554 quoted in the delivery log line, like this:
3556 => >A.N.Other <ano@some.domain> <original@ddress> ...
3558 This has been changed so that it extracts the operative address. There
3559 may be more than one such address. If so, they are comma-separated, like
3562 => >ano@some.domain,ona@other.domain <original@ddress> ...
3564 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
3565 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
3566 not being looked up in the rDNS to get a host name, Exim was showing the
3567 IP address twice in Received: lines, even though the IP addresses were
3568 identical. For example:
3570 Received: from [1.2.3.4] (helo=[1.2.3.4])
3572 However, if the real host name was known, it was omitting the HELO data
3573 if it matched the actual IP address. This has been tidied up so that it
3574 doesn't show the same IP address twice.
3576 PH/07 When both +timestamp and +memory debugging was on, the value given by
3577 $tod_xxx expansions could be wrong, because the tod_stamp() function was
3578 called by the debug printing, thereby overwriting the timestamp buffer.
3579 Debugging no longer uses the tod_stamp() function when +timestamp is set.
3581 PH/08 When the original message was included in an autoreply transport, it
3582 always said "this is a copy of the message, including all the headers",
3583 even if body_only or headers_only was set. It now gives an appropriate
3586 PH/09 Applied a patch from the Sieve maintainer which:
3588 o fixes some comments
3589 o adds the (disabled) notify extension core
3590 o adds some debug output for the result of if/elsif tests
3591 o points to the current vacation draft in the documentation
3592 and documents the missing references header update
3596 o fixes a bug in processing the envelope test (when testing
3597 multiple envelope elements, the last element determined the
3600 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
3601 Electronic Mail") by including:
3603 Auto-submitted: auto-generated
3605 in the messages that it generates (bounce messages and others, such as
3606 warnings). In the case of bounce messages for non-SMTP messages, there was
3607 also a typo: it was using "Auto_submitted" (underscore instead of
3608 hyphen). Since every message generated by Exim is necessarily in response
3609 to another message, thes have all been changed to:
3611 Auto-Submitted: auto-replied
3613 in accordance with these statements in the RFC:
3615 The auto-replied keyword:
3617 - SHOULD be used on messages sent in direct response to another
3618 message by an automatic process,
3620 - MUST NOT be used on manually-generated messages,
3622 - MAY be used on Delivery Status Notifications (DSNs) and Message
3623 Disposition Notifications (MDNs),
3625 - MUST NOT be used on messages generated by automatic or periodic
3626 processes, except for messages which are automatic responses to
3629 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
3630 to the default Received: header definition.
3632 PH/12 Added log selector acl_warn_skipped (default on).
3634 PH/13 After a successful wildlsearch lookup, discard the values of numeric
3635 variables because (a) they are in the wrong storage pool and (b) even if
3636 they were copied, it wouldn't work properly because of the caching.
3638 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
3639 checking when decoding. Apparently there are clients that generate
3640 overlong encoded strings. Why am I not surprised?
3642 PH/15 If the first argument of "${if match_address" was not empty, but did not
3643 contain an "@" character, Exim crashed. Now it writes a panic log message
3644 and treats the condition as false.
3646 PH/16 In autoreply, treat an empty string for "once" the same as unset.
3648 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
3649 extension "envelope-auth". The code is finished and in agreement with
3650 other implementations, but there is no documentation so far and in fact,
3651 nobody wrote the draft yet. This extension is currently #undef'ed, thus
3652 not changing the active code.
3654 Print executed "if" and "elsif" statements when debugging is used. This
3655 helps a great deal to understand what a filter does.
3657 Document more things not specified clearly in RFC3028. I had all this
3658 sorted out, when out of a sudden new issues came to my mind. Oops."
3660 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
3663 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
3664 canonical form (as documented). However, after a host name lookup from
3665 the IP address, check_host() was doing a simple string comparison with
3666 addresses acquired from the DNS when checking that the found name did
3667 have the original IP as one of its addresses. Since any found IPv6
3668 addresses are likely to be in abbreviated form, the comparison could
3669 fail. Luckily, there already exists a function for doing the comparison
3670 by converting both addresses to binary, so now that is used instead of
3671 the text comparison.
3673 PH/20 There was another similar case to PH/19, when a complete host name was
3674 given in a host list; looking up its IP address could give an abbreviated
3675 form, whereas the current host's name might or might not be abbreviated.
3676 The same fix has been applied.
3682 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
3683 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
3686 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
3687 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
3689 PH/03 Typo: missing ".o" in src/pcre/Makefile.
3691 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
3692 header line, restrict the check to what is listed in RFCs 2369 and 2929.
3693 Also, for "Auto-Submitted", treat anything other than "no" as
3694 non-personal, in accordance with RFC 3834. (Previously it treated
3695 anything starting "auto-" as non-personal.)
3697 TF/01 The control=submission/name=... option had a problem with syntax
3698 errors if the name included a slash character. The /name= option
3699 now slurps the rest of the string, so it can include any characters
3700 but it must come last in the list of options (after /sender_retain
3703 PH/05 Some modifications to the interface to the fake nameserver for the new
3711 TK/01 Added the "success_on_redirect" address verification option. See
3712 NewStuff for rationale and an example.
3714 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
3716 PH/02 Patch to exigrep to allow it to work on syslog lines.
3718 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
3719 fread() to skip over the body file's header line, because in Cygwin the
3720 header line is locked and is inaccessible.
3722 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
3723 co-exist for some time) to make it clear that it is the Exim ID that is
3724 referenced, not the Message-ID: header line.
3726 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
3727 string_format() function, because snprintf() does not exist on all
3730 PH/06 The use of forbid_filter_existstest now also locks out the use of the
3731 ${stat: expansion item.
3733 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
3734 protocol synchronization error", to keep the pedants happy.
3736 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
3737 well as for IRIX systems, when gcc is being used. See the host.c source
3740 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
3742 PH/10 Named domain lists were not working if used in a queue_smtp_domains
3745 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
3746 transport and to the smtp transport in LMTP mode.
3748 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
3750 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
3751 run a filter in a subprocess. This could lead to confusion in subsequent
3752 lookups in the parent process. There should also be a search_tidyup() at
3753 the end of the subprocess.
3755 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
3756 only if the host matched helo_try_verify_hosts, which caused the
3757 verification to occur when the EHLO/HELO command was issued. The ACL just
3758 tested the remembered result. Now, if a previous verification attempt has
3759 not happened, "verify = helo" does it there and then.
3761 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
3763 TK/03 Fix log output including CR from clamd.
3765 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
3766 exist provoked a memory error which could cause a segfault.
3768 PH/15 Installed PCRE 6.2
3770 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
3772 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
3773 of the problem. Specifically, suggested +O2 rather than +O1 for the
3776 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
3778 PH/20 If a delivery was routed to a non-standard port by means of an SRV
3779 record, the port was not correctly logged when the outgoing_port log
3780 selector was set (it logged the transort's default port).
3782 PH/21 Added support for host-specific ports to manualroute, queryprogram,
3783 fallback_hosts, and "hosts" in the smtp transport.
3785 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
3786 host errors such as "Connection refused".
3788 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
3789 authentication with radiusclient 0.4.9:
3791 - Error returned from rc_read_config was caught wrongly
3792 - Username/password not passed on to radius server due to wrong length.
3794 The presumption is that some radiusclient API changes for 4.51/PH/17
3795 were not taken care of correctly. The code is still untested by me (my
3796 Linux distribution still has 0.3.2 of radiusclient), but it was
3797 contributed by a Radius user.
3799 PH/24 When doing a callout, the value of $domain wasn't set correctly when
3800 expanding the "port" option of the smtp transport.
3802 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
3803 while reading a MIME header. Thanks to Tom Hughes for a patch.
3805 PH/24 Include config.h inside local_scan.h so that configuration settings are
3808 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
3809 that in an ACL for RCPT (for example), you can examine exactly what was
3812 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
3813 commands, but it was not correctly comparing the address with the actual
3814 client host address. Thus, it would show the EHLO address in Received:
3815 header lines when this was not necessary.
3817 PH/27 Added the % operator to ${eval:}.
3819 PH/28 Exim tries to create and chdir to its spool directory when it starts;
3820 it should be ignoring failures (because with -C, for example, it has lost
3821 privilege). It wasn't ignoring creation failures other than "already
3824 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
3827 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
3828 because an input line was too long, either on its own, or by virtue of
3829 too many continuations, the temporary file was not being removed, and the
3830 return code was incorrect.
3832 PH/31 Missing "BOOL" in function definition in filtertest.c.
3834 PH/32 Applied Sieve patches from the maintainer.
3836 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
3838 PH/33 Added "verify = not_blind".
3840 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
3841 Local/Makefile (with some defaults set). These are used in built scripts
3842 such as exicyclog, but they have never been used in the exim_install
3843 script (though there are many overriding facilities there). I have
3844 arranged that the exim_install script now takes note of these two
3847 PH/35 Installed configuration files for Dragonfly.
3849 PH/36 When a locally submitted message by a trusted user did not contain a
3850 From: header, and the sender address was obtained from -f or from an SMTP
3851 MAIL command, and the trusted user did not use -F to supply a sender
3852 name, $originator_name was incorrectly used when constructing a From:
3853 header. Furthermore, $originator_name was used for submission mode
3854 messages from external hosts without From: headers in a similar way,
3855 which is clearly wrong.
3857 PH/37 Added control=suppress_local_fixups.
3859 PH/38 When log_selector = +received_sender was set, and the addition of the
3860 sender made the log line's construction buffer exactly full, or one byte
3861 less than full, an overflow happened when the terminating "\n" was
3864 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
3865 when the result of a list match is failure because a DNS lookup failed.
3867 PH/40 RM_COMMAND is now used in the building process.
3869 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
3870 the "build-* directories that it finds.
3872 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
3873 address in a domain literal was a prefix of an interface address.
3875 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
3876 when verifying a sender address, unless rewrite_headers is false.
3878 PH/44 Wrote a long comment about why errors_to addresses are verified as
3879 recipients, not senders.
3881 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
3882 the ratelimit ACL was added.
3884 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
3886 PH/46 Added extra information about PostgreSQL errors to the error string.
3888 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
3889 avoiding the need to install special zones in a real server. This is
3890 backwards compatible; if it can't find the fake resolver, it drops back.
3891 Thus, both old and new test suites can be run.
3893 TF/02 Added util/ratelimit.pl
3895 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
3896 clock is set back in time.
3898 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
3899 Candler <B.Candler@pobox.com>.
3901 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
3902 OK for addresses that are the result of redirections.
3904 PH/48 A number of further additions for the benefit of the new test suite,
3905 including a fake gethostbyname() that interfaces to the fake DNS resolver
3908 TF/06 The fix for widen_domains has also been applied to qualify_single and
3909 search_parents which are the other dnslookup options that can cause
3912 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
3915 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
3917 TF/07 Exim produced the error message "an SRV record indicated no SMTP
3918 service" if it encountered an MX record with an empty target hostname.
3919 The message is now "an MX or SRV record indicated no SMTP service".
3921 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
3922 if the DNS of the sending site is misconfigured. This is quite a
3923 common situation. This change restores the behaviour of treating a
3924 helo verification defer as a failure.
3926 PH/51 If self=fail was set on a router, the bounce message did not include the
3927 actual error message.
3933 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
3935 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
3936 command itself ended in error, the underlying message about the transport
3937 filter timeout was being overwritten with the pipe command error. Now the
3938 underlying error message should be appended to the second error message.
3940 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
3942 PH/02 Reduce the amount of output that "make" produces by default. Full output
3943 can still be requested.
3945 PH/03 The warning log line about a condition test deferring for a "warn" verb
3946 was being output only once per connection, rather than after each
3947 occurrence (because it was using the same function as for successful
3948 "warn" verbs). This seems wrong, so I have changed it.
3950 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
3951 it should not have, which might have caused a crash in the right
3952 circumstances, but probably never did.
3954 PH/04 Installed a modified version of Tony Finch's patch to make submission
3955 mode fix the return path as well as the Sender: header line, and to
3956 add a /name= option so that you can make the user's friendly name appear
3959 TF/03 Added the control = fakedefer ACL modifier.
3961 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
3962 Mark Lowes for thorough testing.
3964 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
3966 TK/03 Merged latest SRS patch from Miles Wilton.
3968 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
3969 with the definition in sysexits.h (which is #included earlier).
3970 Fortunately, Exim does not actually use EX_OK. The code used to try to
3971 preserve the sysexits.h value, by assuming that macro definitions were
3972 scanned for macro replacements. I have been disabused of this notion,
3973 so now the code just undefines EX_OK before #including unistd.h.
3975 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
3976 in the smtp transport. When a block could not be written in a single
3977 write() function, the timeout was being re-applied to each part-write.
3978 This seems wrong - if the receiver was accepting one byte at a time it
3979 would take for ever. The timeout is now adjusted when this happens. It
3980 doesn't have to be particularly precise.
3982 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
3983 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
3985 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
3986 without a domain if the check to <postmaster@domain> fails.
3988 SC/01 Eximstats: added -xls and the ability to specify output files
3989 (patch written by Frank Heydlauf).
3991 SC/02 Eximstats: use FileHandles for outputting results.
3993 SC/03 Eximstats: allow any combination of xls, txt, and html output.
3995 SC/04 Eximstats: fixed display of large numbers with -nvr option
3997 SC/05 Eximstats: fixed merging of reports with empty tables.
3999 SC/06 Eximstats: added the -include_original_destination flag
4001 SC/07 Eximstats: removed tabs and trailing whitespace.
4003 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
4005 TK/06 MBOX spool code: Add real "From " MBOX separator line
4006 so the .eml file is really in mbox format (even though
4007 most programs do not really care). Patch from Alex Miller.
4009 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
4010 The latter is generated from $received_to and is only set if the
4011 message has one envelope recipient. SA can use these headers,
4012 obviously out-of-the-box. Patch from Alex Miller.
4014 PH/08 The ${def test on a variable was returning false if the variable's
4015 value was "0", contrary to what the specification has always said!
4016 The result should be true unless the variable is empty.
4018 PH/09 The syntax error of a character other than { following "${if
4019 def:variable_name" (after optional whitespace) was not being diagnosed.
4020 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
4021 accidental colon was present, for example, could give incorrect results.
4023 PH/10 Tidied the code in a number of places where the st_size field of a stat()
4024 result is used (not including appendfile, where other changes are about
4027 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
4028 This involved changing a lot of size variables from int to off_t. It
4029 should work with maildirs and everything.
4031 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
4032 spamd dying while we are connected to it.
4034 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
4037 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
4038 function for BDB 4.3.
4040 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
4042 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
4043 the sources for PCRE have been re-arranged and more files are now
4046 PH/15 The code I had for printing potentially long long variables in PH/11
4047 above was not the best (it lost precision). The length of off_t variables
4048 is now inspected at build time, and an appropriate printing format (%ld
4049 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
4050 to be "long long int" or "long int". This is needed for the internal
4051 formatting function string_vformat().
4053 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
4054 the configuration file to be ":syslog", then the script "guesses" where
4055 the logs files are, rather than using the compiled in default. In our
4056 case the guess is not the same as the compiled default, so the script
4057 suddenly stopped working when I started to use syslog. The patch checks
4058 to see if log_file_path is "". If so, it attempts to read it from exim
4059 with no configuration file to get the compiled in version, before it
4060 falls back to the previous guessing code."
4062 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
4063 implementing BATV in an Exim configuration. See NewStuff for the gory
4066 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
4067 Makefile that are specific to HP-UX.
4069 PH/18 If the "use_postmaster" option was set for a recipient callout together
4070 with the "random" option, the postmaster address was used as the MAIL
4071 FROM address for the random test, but not for the subsequent recipient
4072 test. It is now used for both.
4074 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
4075 patch removes a few documentation additions to RFC 3028, because the
4076 latest draft now contains them. It adds the new en;ascii-case comparator
4077 and a new error check for 8bit text in MIME parts. Comparator and
4078 require names are now matched exactly. I enabled the subaddress
4079 extension, but it is not well tested yet (read: it works for me)."
4081 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
4082 rework some of the code of TK/09 above to avoid the hardwired use of
4083 "%lld" and "long long". Replaced the call to snprintf() with a call to
4086 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
4087 records point to non-existent hosts", "retry timeout exceeded", and
4088 "retry time not reached for any host after a long failure period".
4090 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
4091 experimental DomainKeys support:
4093 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
4094 (2) On an error such as an illegally used "control", the wrong name for
4095 the control was given.
4097 These problems did NOT occur unless DomainKeys support was compiled.
4099 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
4101 PH/24 Added ${if match_ip condition.
4103 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
4104 databases so that it will be absolutely obvious if a crash occurs in the
4105 DB library. This is a regular occurrence (often caused by mis-matched
4108 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
4109 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
4110 picked up on a user's system that detects such things. There doesn't seem
4111 to be a gcc warning option for this - only an attribute that has to be
4112 put on the function's prototype. It seems that in Fedora Core 4 they have
4113 set this on a number of new functions. No doubt there will be more in due
4116 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
4117 specify a transport. However, if an address that was verified by such a
4118 router was the subject of a callout, Exim crashed because it tried to
4119 read the rcpt_include_affixes from the non-existent transport. Now it
4120 just assumes that the setting of that option is false. This bug was
4121 introduced by 4.51/PH/31.
4123 PH/28 Changed -d+all to exclude +memory, because that information is very
4124 rarely of interest, but it makes the output a lot bigger. People tend to
4125 do -d+all out of habit.
4127 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
4128 code in os-type was giving problems when libc.so lives in lib64, like on
4131 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
4132 aren't the modern standard, and it seems that some systems' include files
4133 don't always have them. Exim was already checking for some of the newer
4134 ones like T_AAAA, and defining it itself. I've added checks for all the
4135 record types that Exim uses.
4137 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
4138 not automatically generating a new one, as it is supposed to. This
4139 prevented TLS from working. If the file did exist, but contained invalid
4140 data, a new version was generated, as expected. It was only the case of a
4141 non-existent file that was broken.
4143 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
4144 with a change in libdomainkeys > 0.64.
4146 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
4147 from DNS. If the selector record carries the flag, it now has
4148 precedence over the domain-wide flag.
4150 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
4152 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
4153 the use of a port name that isn't defined in /etc/services) occurred, the
4154 message was deferred as in a normal delivery, and thus remained on the
4155 spool, instead of being failed because of the mua_wrapper setting. This
4156 is now fixed, and I tidied up some of the mua_wrapper messages at the
4159 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
4160 the messages in a hash of arrays rather than using individual hashes.
4161 This is a bit cleaner and results in dramatic memory savings, albeit
4162 at a slight CPU cost.
4164 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
4165 as requested by Marc Sherman.
4167 SC/10 Eximstats: added histograms for user specified patterns as requested
4170 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
4172 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
4173 fopen() in the content-scanning modules that did not already have it.
4179 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
4180 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
4182 TK/02 Fix ACL "control" statement not being available in MIME ACL.
4184 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
4186 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
4187 to test Sieve filters that use "vacation".
4189 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
4190 that changes the way the GnuTLS parameters are stored in the cache file.
4191 The new format can be generated externally. For backward compatibility,
4192 if the data in the cache doesn't make sense, Exim assumes it has read an
4193 old-format file, and it generates new data and writes a new file. This
4194 means that you can't go back to an older release without removing the
4197 PH/03 A redirect router that has both "unseen" and "one_time" set does not
4198 work if there are any delivery delays because "one_time" forces the
4199 parent to be marked "delivered", so its unseen clone is never tried
4200 again. For this reason, Exim now forbids the simultaneous setting of
4203 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
4204 redirected to themselves ("homonym" addresses). Read the long ChangeLog
4205 entry if you want to know the details. The fix, however, neglected to
4206 consider the case when local delivery batching is involved. The test for
4207 "previously delivered" was not happening when checking to see if an
4208 address could be batched with a previous (undelivered) one; under
4209 certain circumstances this could lead to multiple deliveries to the same
4212 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
4213 in its include files, and this causes problems building Exim.
4215 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
4216 header_syntax) but Exim was just ignoring anything given after a slash.
4217 In particular, this caused confusion with an attempt to use "verify =
4218 reverse_host_lookup/defer_ok". An error is now given when options are
4219 supplied for verify items that do not have them. (Maybe reverse_host_
4220 lookup should have a defer_ok option, but that's a different point.)
4222 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
4223 defined by RFC 821) to 2048, because there were problems with some AUTH
4224 commands, and RFC 1869 says the size should be increased for extended
4225 SMTP commands that take arguments.
4227 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
4230 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
4231 "unknown" error; now it says that the functionality isn't in the binary.
4233 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
4234 an address' error message when a string expansion fails (syntax or
4235 whatever). Otherwise the password may appear in the log. Following change
4236 PH/42 below, there is no longer a chance of it appearing in a bounce
4239 PH/11 Installed exipick version 20050225.0 from John Jetmore.
4241 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
4242 of its addresses was ever tried. (Bugzilla bug #2.)
4244 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
4245 the result incorrectly in the debug output. (It correctly added a newline
4246 to what was transported.)
4248 TF/01 Added $received_time.
4250 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
4251 commented out examples of how to interface to a virus scanner and to
4252 SpamAssassin. Also added commented examples of av_scanner and
4253 spamd_address settings.
4255 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
4256 and controls are allowed in which ACLs. There were a couple of minor
4257 errors. Some of the entries in the conditions table (which is a table of
4258 where they are NOT allowed) were getting very unwieldy; rewrote them as a
4259 negation of where the condition IS allowed.
4261 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
4263 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
4264 header file does not have a version number, so I've had to invent a new
4265 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
4266 API. The code is untested by me (my Linux distribution still has 0.3.2 of
4267 radiusclient), but it was contributed by a Radius user.
4269 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
4270 files or directories, for OpenSSL.
4272 PH/19 When an Exim process that is running as root has to create an Exim log
4273 file, it does so in a subprocess that runs as exim:exim so as to get the
4274 ownership right at creation (otherwise, other Exim processes might see
4275 the file with the wrong ownership). There was no test for failure of this
4276 fork() call, which would lead to the process getting stuck as it waited
4277 for a non-existent subprocess. Forks do occasionally fail when resources
4278 run out. I reviewed all the other calls to fork(); they all seem to check
4281 PH/20 When checking for unexpected SMTP input at connect time (before writing
4282 the banner), Exim was not dealing correctly with a non-positive return
4283 from the read() function. If the client had disconnected by this time,
4284 the result was a log entry for a synchronization error with an empty
4285 string after "input=" when read() returned zero. If read() returned -1
4286 (an event I could not check), uninitialized data bytes were printed.
4287 There were reports of junk text (parts of files, etc) appearing after
4290 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
4292 PH/22 Added support for macro redefinition, and (re)definition in between
4293 driver and ACL definitions.
4295 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
4296 forgetting to use the resulting value; it was using the unexpanded value.
4298 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
4299 hadn't been configured. The fix is from Juergen Kreileder, who
4300 understands it better than I do:
4302 "Here's what I see happening with three configured cyrus_sasl
4303 authenticators configured (plain, login, cram-md5):
4305 On startup auth_cyrus_sasl_init() gets called for each of these.
4306 This means three calls to sasl_listmech() without a specified mech_list.
4307 => SASL tests which mechs of all available mechs actually work
4308 => three warnings about OTP not working
4309 => the returned list contains: plain, login, cram-md5, digest-md5, ...
4311 With the patch, sasl_listmech() also gets called three times. But now
4312 SASL's mech_list option is set to the server_mech specified in the the
4313 authenticator. Or in other words, the answer from sasl_listmech()
4314 gets limited to just the mech you're testing for (which is different
4316 => the return list contains just 'plain' or 'login', 'cram-md5' or
4317 nothing depending on the value of ob->server_mech.
4319 I've just tested the patch: Authentication still works fine,
4320 unavailable mechs specified in the exim configuration are still
4321 caught, and the auth.log warnings about OTP are gone."
4323 PH/25 When debugging is enabled, the contents of the command line are added
4324 to the debugging output, even when log_selector=+arguments is not
4327 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
4328 answer is "GNU", and only if the return is "GNU/something" is the answer
4331 PH/27 $acl_verify_message is now set immediately after the failure of a
4332 verification in an ACL, and so is available in subsequent modifiers. In
4333 particular, the message can be preserved by coding like this:
4335 warn !verify = sender
4336 set acl_m0 = $acl_verify_message
4338 Previously, $acl_verify_message was set only while expanding "message"
4339 and "log_message" when a very denied access.
4341 PH/28 Modified OS/os.c-Linux with
4343 -#ifndef OS_LOAD_AVERAGE
4344 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
4346 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
4347 nomenclature these days.)
4349 PH/29 Installed patch from the Sieve maintainer that adds the options
4350 sieve_useraddress and sieve_subaddress to the redirect router.
4352 PH/30 In these circumstances:
4353 . Two addresses routed to the same list of hosts;
4354 . First host does not offer TLS;
4355 . First host accepts first address;
4356 . First host gives temporary error to second address;
4357 . Second host offers TLS and a TLS session is established;
4358 . Second host accepts second address.
4359 Exim incorrectly logged both deliveries with the TLS parameters (cipher
4360 and peerdn, if requested) that were in fact used only for the second
4363 PH/31 When doing a callout as part of verifying an address, Exim was not paying
4364 attention to any local part prefix or suffix that was matched by the
4365 router that accepted the address. It now behaves in the same way as it
4366 does for delivery: the affixes are removed from the local part unless
4367 rcpt_include_affixes is set on the transport.
4369 PH/32 Add the sender address, as F=<...>, to the log line when logging a
4370 timeout during the DATA phase of an incoming message.
4372 PH/33 Sieve envelope tests were broken for match types other than :is. I have
4373 applied a patch sanctioned by the Sieve maintainer.
4375 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
4376 the uid or gid is negative. A case of a negative gid caused this to be
4377 noticed. The fix allows for either to be negative.
4379 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
4380 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
4383 PH/36 The change PH/12 above was broken. Fixed it.
4385 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
4386 the grounds that routing the same address twice would always produce the
4387 same answer. This might have been true once, but it is certainly no
4388 longer true now. Routing a child address may depend on the previous
4389 routing that produced that child. Some complicated redirection strategies
4390 went wrong when messages had multiple recipients, and made Exim's
4391 behaviour dependent on the order in which the addresses were given.
4393 I have moved the duplicate checking until after the routing is complete.
4394 Exim scans the addresses that are assigned to local and remote
4395 transports, and removes any duplicates. This means that more work will be
4396 done, as duplicates will always all be routed, but duplicates are
4397 presumably rare, so I don't expect this is of any significance.
4399 For deliveries to pipes, files, and autoreplies, the duplicate checking
4400 still happens during the routing process, since they are not going to be
4403 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
4404 It corrects a timeout issue with spamd. This is Ian's comment: "The
4405 background is that sometimes spamd either never reads data from a
4406 connection it has accepted, or it never writes response data. The exiscan
4407 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
4408 blindly assumes that writes won't block so it may never time out."
4410 PH/39 Allow G after quota size as well as K and M.
4412 PH/40 The value set for $authenticated_id in an authenticator may not contain
4413 binary zeroes or newlines because the value is written to log lines and
4414 to spool files. There was no check on this. Now the value is run through
4415 the string_printing() function so that such characters are converted to
4416 printable escape sequences.
4418 PH/41 $message_linecount is a new variable that contains the total number of
4419 lines in the message. Compare $body_linecount, which is the count for the
4422 PH/42 Exim no longer gives details of delivery errors for specific addresses in
4423 bounce and delay warning messages, except in certain special cases, which
4426 (a) An SMTP error message from a remote host;
4427 (b) A message specified in a :fail: redirection;
4428 (c) A message specified in a "fail" command in a system filter;
4429 (d) A message specified in a FAIL return from the queryprogram router;
4430 (e) A message specified by the cannot_route_message router option.
4432 In these cases only, Exim does include the error details in bounce and
4433 warning messages. There are also a few cases where bland messages such
4434 as "unrouteable address" or "local delivery error" are given.
4436 PH/43 $value is now also set for the "else" part of a ${run expansion.
4438 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
4439 being worked on, but at least Exim now implements the latest version to
4442 PH/45 In a pipe transport, although a timeout while waiting for the pipe
4443 process to complete was treated as a delivery failure, a timeout while
4444 writing the message to the pipe was logged, but erroneously treated as a
4445 successful delivery. Such timeouts include transport filter timeouts. For
4446 consistency with the overall process timeout, these timeouts are now
4447 treated as errors, giving rise to delivery failures by default. However,
4448 there is now a new Boolean option for the pipe transport called
4449 timeout_defer, which, if set TRUE, converts the failures into defers for
4450 both kinds of timeout. A transport filter timeout is now identified in
4453 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
4454 systems where "make" and "gmake" are different, calling "gmake" at top
4455 level broke things. I've arranged for the value of $(MAKE) to be passed
4456 from the Makefile to this script so that it can call the same version of
4460 A note about Exim versions 4.44 and 4.50
4461 ----------------------------------------
4463 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
4464 changes of various kinds. As a consequence, a big documentation update was
4465 needed. This delayed the release for rather longer than seemed good, especially
4466 in the light of a couple of (minor) security issues. Therefore, the changes
4467 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
4468 release. So 4.44 and 4.50 are in effect two different branches that both start
4471 I have left the 4.50 change log unchanged; it contains all the changes since
4472 4.43. The change log for 4.44 is below; many of its items are identical to
4473 those for 4.50. This seems to be the most sensible way to preserve the
4474 historical information.
4480 1. Minor wording change to the doc/README.SIEVE file.
4482 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
4483 computation of the current number of files was incorrect.
4485 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4486 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4489 4. Give more explanation in the error message when the command for a transport
4490 filter fails to execute.
4492 5. There are several places where Exim runs a non-Exim command in a
4493 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4494 was being done only for the command run by the queryprogram router. It is
4495 now done for all such subprocesses. The other cases are: ${run, transport
4496 filters, and the commands run by the lmtp and pipe transports.
4498 6. Added CONFIGURE_GROUP build-time option.
4500 7. Some older OS have a limit of 256 on the maximum number of file
4501 descriptors. Exim was using setrlimit() to set 1000 as a large value
4502 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4503 systems. I've change it so that if it can't get 1000, it tries for 256.
4505 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4506 was an oversight, and furthermore, ever since the addition of extra
4507 controls (e.g. 4.43/32), the checks on when to allow different forms of
4508 "control" were broken. There should now be diagnostics for all cases when a
4509 control that does not make sense is encountered.
4511 9. Added the /retain_sender option to "control=submission".
4513 10. $recipients is now available in the predata ACL (oversight).
4515 11. Tidy the search cache before the fork to do a delivery from a message
4516 received from the command line. Otherwise the child will trigger a lookup
4517 failure and thereby defer the delivery if it tries to use (for example) a
4518 cached ldap connection that the parent has called unbind on.
4520 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4521 of $address_data from the recipient verification was clobbered by the
4522 sender verification.
4524 13. The value of address_data from a sender verification is now available in
4525 $sender_address_data in subsequent conditions in the ACL statement.
4527 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
4529 15. Added a new option "connect=<time>" to callout options, to set a different
4532 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4533 was its contents. (It was OK if the option was not defined at all.)
4535 17. A "Completed" log line is now written for messages that are removed from
4536 the spool by the -Mrm option.
4538 18. New variables $sender_verify_failure and $recipient_verify_failure contain
4539 information about exactly what failed.
4541 19. Added -dd to debug only the daemon process.
4543 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
4544 handles timeouts, both on the server side and network timeouts. Renamed the
4545 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
4547 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
4548 It is now set to "smtps".
4550 22. $host_address is now set to the target address during the checking of
4551 ignore_target_hosts.
4553 23. When checking ignore_target_hosts for an ipliteral router, no host name was
4554 being passed; this would have caused $sender_host_name to have been used if
4555 matching the list had actually called for a host name (not very likely,
4556 since this list is usually IP addresses). A host name is now passed as
4559 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4560 code that specifies a non-restarting handler (typically sigaction() in
4561 modern systems) in an attempt to fix a rare and obscure crash bug.
4563 25. Narrowed the window for a race in the daemon that could cause it to ignore
4564 SIGCHLD signals. This is not a major problem, because they are used only to
4565 wake it up if nothing else does.
4567 26. A malformed maildirsize file could cause Exim to calculate negative values
4568 for the mailbox size or file count. Odd effects could occur as a result.
4569 The maildirsize information is now recalculated if the size or filecount
4572 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4573 support for a long time. Removed HAVE_SYS_VFS_H.
4575 28. Installed the latest version of exipick from John Jetmore.
4577 29. In an address list, if the pattern was not a regular expression, an empty
4578 subject address (from a bounce message) matched only if the pattern was an
4579 empty string. Non-empty patterns were not even tested. This was the wrong
4580 because it is perfectly reasonable to use an empty address as part of a
4581 database query. An empty address is now tested by patterns that are
4582 lookups. However, all the other forms of pattern expect the subject to
4583 contain a local part and a domain, and therefore, for them, an empty
4584 address still always fails if the pattern is not itself empty.
4586 30. Exim went into a mad DNS loop when attempting to do a callout where the
4587 host was specified on an smtp transport, and looking it up yielded more
4588 than one IP address.
4590 31. Re-factored the code for checking spool and log partition space into a
4591 function that finds that data and another that does the check. The former
4592 is then used to implement four new variables: $spool_space, $log_space,
4593 $spool_inodes, and $log_inodes.
4595 32. The RFC2047 encoding function was originally intended for short strings
4596 such as real names; it was not keeping to the 75-character limit for
4597 encoded words that the RFC imposes. It now respects the limit, and
4598 generates multiple encoded words if necessary. To be on the safe side, I
4599 have increased the buffer size for the ${rfc2047: expansion operator from
4602 33. It is now permitted to omit both strings after an "if" condition; if the
4603 condition is true, the result is "true". As before, when the second string
4604 is omitted, a false condition yields an empty string. This makes it less
4605 cumbersome to write custom ACL and router conditions.
4607 34. Failure to deliver a bounce message always caused it to be frozen, even if
4608 there was an errors_to setting on the router. The errors_to setting is now
4611 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4612 canonical form (fully expanded) before being placed in
4613 $sender_host_address.
4615 36. The table in the code that translates DNS record types into text (T_A to
4616 "A" for instance) was missing entries for NS and CNAME. It is just possible
4617 that this could have caused confusion if both these types were looked up
4618 for the same domain, because the text type is used as part of Exim's
4619 per-process caching. But the chance of anyone hitting this buglet seems
4622 37. The dnsdb lookup has been extended in a number of ways.
4624 (1) There is a new type, "zns", which walks up the domain tree until it
4625 finds some nameserver records. It should be used with care.
4627 (2) There is a new type, "mxh", which is like "mx" except that it returns
4628 just the host names, not the priorities.
4630 (3) It is now possible to give a list of domains (or IP addresses) to be
4631 looked up. The behaviour when one of the lookups defers can be
4632 controlled by a keyword.
4634 (4) It is now possible to specify the separator character for use when
4635 multiple records are returned.
4637 38. The dnslists ACL condition has been extended: it is now possible to supply
4638 a list of IP addresses and/or domains to be looked up in a particular DNS
4641 39. Added log_selector=+queue_time_overall.
4643 40. When running the queue in the test harness, wait just a tad after forking a
4644 delivery process, to get repeatability of debugging output.
4646 41. Include certificate and key file names in error message when GnuTLS fails
4647 to set them up, because the GnuTLS error message doesn't include the name
4648 of the failing file when there is a problem reading it.
4650 42. Allow both -bf and -bF in the same test run.
4652 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
4654 44. The "Exiscan patch" is now merged into the mainline Exim source.
4656 45. Sometimes the final signoff response after QUIT could fail to get
4657 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4658 < 0 before doing a fflush(). This bug looks as though it goes back to the
4659 introduction of TLS in release 3.20, but "sometimes" must have been rare
4660 because the tests only now provoked it.
4662 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
4663 (this can affect the format of dates).
4665 47. exim_tidydb, when checking for the continued existence of a message for
4666 which it has found a message-specific retry record, was not finding
4667 messages that were in split spool directories. Consequently, it was
4668 deleting retry records that should have stayed in existence.
4670 48. Steve fixed some bugs in eximstats.
4672 49. The SPA authentication driver was not abandoning authentication and moving
4673 on to the next authenticator when an expansion was forced to fail,
4674 contradicting the general specification for all authenticators. Instead it
4675 was generating a temporary error. It now behaves as specified.
4677 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
4678 (the order specifies the preference for clients). The order is now AES256,
4679 AES128, 3DES, ARCFOUR128.
4681 51. Small patch to Sieve code - explicitly set From: when generating an
4684 52. Exim crashed if a remote delivery caused a very long error message to be
4685 recorded - for instance if somebody sent an entire SpamAssassin report back
4686 as a large number of 550 error lines. This bug was coincidentally fixed by
4687 increasing the size of one of Exim's internal buffers (big_buffer) that
4688 happened as part of the Exiscan merge. However, to be on the safe side, I
4689 have made the code more robust (and fixed the comments that describe what
4692 53. Now that there can be additional text after "Completed" in log lines (if
4693 the queue_time_overall log selector is set), a one-byte patch to exigrep
4694 was needed to allow it to recognize "Completed" as not the last thing in
4697 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
4698 patch that reportedly fixes this has been added. I am not expert enough to
4699 create a test for it. This is what the patch creator wrote:
4701 "I found a little strange behaviour of ldap code when working with
4702 Windows 2003 AD Domain, where users was placed in more than one
4703 Organization Units. When I tried to give exim partial DN, the exit code
4704 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
4705 But simultaneously result of request was absolutely normal ldap result,
4706 so I produce this patch..."
4708 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
4709 REFERENCE, so I have modified the code to exclude the patch when that macro
4712 55. Some experimental protocols are using DNS PTR records for new purposes. The
4713 keys for these records are domain names, not reversed IP addresses. The
4714 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4715 leaves it alone. Component reversal etc. now happens only for IP addresses.
4718 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
4720 57. Double the size of the debug message buffer (to 2048) so that more of very
4721 long debug lines gets shown.
4723 58. The exicyclog utility now does better if the number of log files to keep
4724 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
4726 59. Two changes related to the smtp_active_hostname option:
4728 (1) $smtp_active_hostname is now available as a variable.
4729 (2) The default for smtp_banner uses $smtp_active_hostname instead
4730 of $primary_hostname.
4732 60. The host_aton() function is supposed to be passed a string that is known
4733 to be a valid IP address. However, in the case of IPv6 addresses, it was
4734 not checking this. This is a hostage to fortune. Exim now panics and dies
4735 if the condition is not met. A case was found where this could be provoked
4736 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4737 components; fortuitously, this particular loophole had already been fixed
4738 by change 4.50/55 above.
4740 If there are any other similar loopholes, the new check in host_aton()
4741 itself should stop them being exploited. The report I received stated that
4742 data on the command line could provoke the exploit when Exim was running as
4743 exim, but did not say which command line option was involved. All I could
4744 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4745 running as the user.
4748 61. There was a buffer overflow vulnerability in the SPA authentication code
4749 (which came originally from the Samba project). I have added a test to the
4750 spa_base64_to_bits() function which I hope fixes it.
4753 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
4754 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
4756 63. The daemon start-up calls getloadavg() while still root for those OS that
4757 need the first call to be done as root, but it missed one case: when
4758 deliver_queue_load_max is set with deliver_drop_privilege. This is
4759 necessary for the benefit of the queue runner, because there is no re-exec
4760 when deliver_drop_privilege is set.
4762 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
4763 This has been fixed.
4765 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4766 was in use, was not putting the data itself into the right store pool;
4767 consequently, it could be overwritten for a subsequent message in the same
4768 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4771 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
4773 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
4774 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
4775 the function were treating the return as a boolean value, which happened to
4776 work because 0=false and not-0=true, but is not correct code.
4778 68. The host_aton() function was not handling scoped IPv6 addresses (those
4779 with, for example, "%eth0" on the end) correctly.
4781 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
4782 negated items (that is, ~something) in unsigned ints. Some compilers
4783 apparently mutter when there is no cast.
4785 70. If an address verification called from an ACL failed, and did not produce a
4786 user-specific message (i.e. there was only a "system" message), nothing was
4787 put in $acl_verify_message. In this situation, it now puts the system
4790 71. Change 4.23/11 added synchronization checking at the start of an SMTP
4791 session; change 4.31/43 added the unwanted input to the log line - except
4792 that it did not do this in the start of session case. It now does.
4794 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
4795 This is wrong and can cause the other end to generate a synchronization
4796 error if it is another Exim or anything else that does the synchronization
4797 check. A QUIT command is no longer sent after a timeout.
4799 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
4800 during host lookups.
4802 74. The defer_ok option of callout verification was not working if it was used
4803 when verifying addresses in header lines, that is, for this case:
4805 verify = header_sender/callout=defer_ok
4807 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
4808 those file descriptors could be used for SMTP connections. If anything
4809 wrote to stderr (the example that came up was "warn" in embedded Perl), it
4810 could be sent to the SMTP client, causing chaos. The daemon now opens
4811 stdin, stdout, and stderr to /dev/null when it puts itself into the
4814 76. Arrange for output from Perl's "warn" command to be written to Exim's main
4815 log by default. The user can override this with suitable Perl magic.
4817 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
4818 the log message when discard triggers, was not working for the DATA ACL or
4819 for the non-SMTP ACL.
4821 78. Error message wording change in sieve.c.
4823 79. If smtp_accept_max_per_host was set, the number of connections could be
4824 restricted to fewer than expected, because the daemon was trying to set up
4825 a new connection before checking whether the processes handling previous
4826 connections had finished. The check for completed processes is now done
4827 earlier. On busy systems, this bug wouldn't be noticed because something
4828 else would have woken the daemon, and it would have reaped the completed
4831 80. If a message was submitted locally by a user whose login name contained one
4832 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
4833 It caused a spool format error. I have fixed the spool reading code. A
4834 related problem was that the "from" clause in the Received: line became
4835 illegal because of the space(s). It is now covered by ${quote_local_part.
4837 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
4840 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
4842 83. Patch from Sieve maintainer to fix unterminated string problem in
4843 "vacation" handling.
4845 84. Some minor changes to the Linux configuration files to help with other
4846 OS variants using glibc.
4848 85. One more patch for Sieve to update vacation handling to latest spec.
4851 ----------------------------------------------------
4852 See the note above about the 4.44 and 4.50 releases.
4853 ----------------------------------------------------
4859 1. Change 4.43/35 introduced a bug that caused file counts to be
4860 incorrectly computed when quota_filecount was set in an appendfile
4863 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4864 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4867 3. Give more explanation in the error message when the command for a transport
4868 filter fails to execute.
4870 4. There are several places where Exim runs a non-Exim command in a
4871 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4872 was being done only for the command run by the queryprogram router. It is
4873 now done for all such subprocesses. The other cases are: ${run, transport
4874 filters, and the commands run by the lmtp and pipe transports.
4876 5. Some older OS have a limit of 256 on the maximum number of file
4877 descriptors. Exim was using setrlimit() to set 1000 as a large value
4878 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4879 systems. I've change it so that if it can't get 1000, it tries for 256.
4881 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4882 was an oversight, and furthermore, ever since the addition of extra
4883 controls (e.g. 4.43/32), the checks on when to allow different forms of
4884 "control" were broken. There should now be diagnostics for all cases when a
4885 control that does not make sense is encountered.
4887 7. $recipients is now available in the predata ACL (oversight).
4889 8. Tidy the search cache before the fork to do a delivery from a message
4890 received from the command line. Otherwise the child will trigger a lookup
4891 failure and thereby defer the delivery if it tries to use (for example) a
4892 cached ldap connection that the parent has called unbind on.
4894 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4895 of $address_data from the recipient verification was clobbered by the
4896 sender verification.
4898 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4899 was its contents. (It was OK if the option was not defined at all.)
4901 11. A "Completed" log line is now written for messages that are removed from
4902 the spool by the -Mrm option.
4904 12. $host_address is now set to the target address during the checking of
4905 ignore_target_hosts.
4907 13. When checking ignore_target_hosts for an ipliteral router, no host name was
4908 being passed; this would have caused $sender_host_name to have been used if
4909 matching the list had actually called for a host name (not very likely,
4910 since this list is usually IP addresses). A host name is now passed as
4913 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4914 code that specifies a non-restarting handler (typically sigaction() in
4915 modern systems) in an attempt to fix a rare and obscure crash bug.
4917 15. Narrowed the window for a race in the daemon that could cause it to ignore
4918 SIGCHLD signals. This is not a major problem, because they are used only to
4919 wake it up if nothing else does.
4921 16. A malformed maildirsize file could cause Exim to calculate negative values
4922 for the mailbox size or file count. Odd effects could occur as a result.
4923 The maildirsize information is now recalculated if the size or filecount
4926 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4927 support for a long time. Removed HAVE_SYS_VFS_H.
4929 18. Updated exipick to current release from John Jetmore.
4931 19. Allow an empty sender to be matched against a lookup in an address list.
4932 Previously the only cases considered were a regular expression, or an
4935 20. Exim went into a mad DNS lookup loop when doing a callout where the
4936 host was specified on the transport, if the DNS lookup yielded more than
4939 21. The RFC2047 encoding function was originally intended for short strings
4940 such as real names; it was not keeping to the 75-character limit for
4941 encoded words that the RFC imposes. It now respects the limit, and
4942 generates multiple encoded words if necessary. To be on the safe side, I
4943 have increased the buffer size for the ${rfc2047: expansion operator from
4946 22. Failure to deliver a bounce message always caused it to be frozen, even if
4947 there was an errors_to setting on the router. The errors_to setting is now
4950 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4951 canonical form (fully expanded) before being placed in
4952 $sender_host_address.
4954 24. Updated eximstats to version 1.33
4956 25. Include certificate and key file names in error message when GnuTLS fails
4957 to set them up, because the GnuTLS error message doesn't include the name
4958 of the failing file when there is a problem reading it.
4960 26. Expand error message when OpenSSL has problems setting up cert/key files.
4963 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
4964 (this can affect the format of dates).
4966 28. exim_tidydb, when checking for the continued existence of a message for
4967 which it has found a message-specific retry record, was not finding
4968 messages that were in split spool directories. Consequently, it was
4969 deleting retry records that should have stayed in existence.
4971 29. eximstats updated to version 1.35
4972 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
4973 1.35 - bugfix such that pie charts by volume are generated correctly
4975 30. The SPA authentication driver was not abandoning authentication and moving
4976 on to the next authenticator when an expansion was forced to fail,
4977 contradicting the general specification for all authenticators. Instead it
4978 was generating a temporary error. It now behaves as specified.
4980 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
4981 (the order specifies the preference for clients). The order is now AES256,
4982 AES128, 3DES, ARCFOUR128.
4984 31. Small patch to Sieve code - explicitly set From: when generating an
4987 32. Exim crashed if a remote delivery caused a very long error message to be
4988 recorded - for instance if somebody sent an entire SpamAssassin report back
4989 as a large number of 550 error lines. This bug was coincidentally fixed by
4990 increasing the size of one of Exim's internal buffers (big_buffer) that
4991 happened as part of the Exiscan merge. However, to be on the safe side, I
4992 have made the code more robust (and fixed the comments that describe what
4995 33. Some experimental protocols are using DNS PTR records for new purposes. The
4996 keys for these records are domain names, not reversed IP addresses. The
4997 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4998 leaves it alone. Component reversal etc. now happens only for IP addresses.
5001 34. The host_aton() function is supposed to be passed a string that is known
5002 to be a valid IP address. However, in the case of IPv6 addresses, it was
5003 not checking this. This is a hostage to fortune. Exim now panics and dies
5004 if the condition is not met. A case was found where this could be provoked
5005 from a dnsdb PTR lookup with an IPv6 address that had more than 8
5006 components; fortuitously, this particular loophole had already been fixed
5007 by change 4.50/55 or 4.44/33 above.
5009 If there are any other similar loopholes, the new check in host_aton()
5010 itself should stop them being exploited. The report I received stated that
5011 data on the command line could provoke the exploit when Exim was running as
5012 exim, but did not say which command line option was involved. All I could
5013 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
5014 running as the user.
5017 35. There was a buffer overflow vulnerability in the SPA authentication code
5018 (which came originally from the Samba project). I have added a test to the
5019 spa_base64_to_bits() function which I hope fixes it.
5022 36. The daemon start-up calls getloadavg() while still root for those OS that
5023 need the first call to be done as root, but it missed one case: when
5024 deliver_queue_load_max is set with deliver_drop_privilege. This is
5025 necessary for the benefit of the queue runner, because there is no re-exec
5026 when deliver_drop_privilege is set.
5028 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
5029 was in use, was not putting the data itself into the right store pool;
5030 consequently, it could be overwritten for a subsequent message in the same
5031 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
5034 38. Sometimes the final signoff response after QUIT could fail to get
5035 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
5036 < 0 before doing a fflush(). This bug looks as though it goes back to the
5037 introduction of TLS in release 3.20, but "sometimes" must have been rare
5038 because the tests only now provoked it.
5044 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
5045 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
5046 Now it returns an integer. A number of calls were still expecting a T/F
5047 return. Fortuitously, in all cases, the tests worked in OK situations,
5048 which is the norm. However, things would have gone wrong on any write
5049 failures on the smtp file descriptor. This function is used when sending
5050 messages over SMTP and also when doing verify callouts.
5052 2. When Exim is called to do synchronous delivery of a locally submitted
5053 message (the -odf or -odi options), it no longer closes stderr before doing
5056 3. Implemented the mua_wrapper option.
5058 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
5060 5. Implemented the functions header_remove(), header_testname(),
5061 header_add_at_position(), and receive_remove_recipient(), and exported them
5064 6. If an ACL "warn" statement specified the addition of headers, Exim already
5065 inserted X-ACL-Warn: at the start if there was no header name. However, it
5066 was not making this test for the second and subsequent header lines if
5067 there were newlines in the string. This meant that an invalid header could
5068 be inserted if Exim was badly configured.
5070 7. Allow an ACL "warn" statement to add header lines at the start or after all
5071 the Received: headers, as well as at the end.
5073 8. Added the rcpt_4xx retry error code.
5075 9. Added postmaster_mailfrom=xxx to callout verification option.
5077 10. Added mailfrom=xxxx to the callout verification option, for verify=
5080 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
5081 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
5083 12. Inserted some casts to stop certain compilers warning when using pointer
5084 differences as field lengths or precisions in printf-type calls (mostly
5085 affecting debugging statements).
5087 13. Added optional readline() support for -be (dynamically loaded).
5089 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
5090 same clock tick as a message's arrival, so that its received time was the
5091 same as the "first fail" time on the retry record, and that message
5092 remained on the queue past the ultimate address timeout, every queue runner
5093 would try a delivery (because it was past the ultimate address timeout) but
5094 after another failure, the ultimate address timeout, which should have then
5095 bounced the address, did not kick in. This was a "< instead of <=" error;
5096 in most cases the first failure would have been in the next clock tick
5097 after the received time, and all would be well.
5099 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
5100 being recognized when the domain list was tested by the match_domain
5101 condition in an expansion string.
5103 16. Added the ${str2b64: operator.
5105 17. Exim was always calling setrlimit() to set a large limit for the number of
5106 processes, without checking whether the existing limit was already
5107 adequate. (It did check for the limit on file descriptors.) Furthermore,
5108 errors from getrlimit() and setrlimit() were being ignored. Now they are
5109 logged to the main and panic logs, but Exim does carry on, to try to do its
5110 job under whatever limits there are.
5112 18. Imported PCRE 5.0.
5114 19. Trivial typo in log message " temporarily refused connection" (the leading
5117 20. If the log selector return_path_on_delivery was set and an address was
5118 redirected to /dev/null, the delivery process crashed because it assumed
5119 that a return path would always be set for a "successful" delivery. In this
5120 case, the whole delivery is bypassed as an optimization, and therefore no
5123 21. Internal re-arrangement: the function for sending a challenge and reading
5124 a response while authentication was assuming a zero-terminated challenge
5125 string. It's now changed to take a pointer and a length, to allow for
5126 binary data in such strings.
5128 22. Added the cyrus_sasl authenticator (code supplied by MBM).
5130 23. Exim was not respecting finduser_retries when seeking the login of the
5131 uid under which it was called; it was always trying 10 times. (The default
5132 setting of finduser_retries is zero.) Also, it was sleeping after the final
5133 failure, which is pointless.
5135 24. Implemented tls_on_connect_ports.
5137 25. Implemented acl_smtp_predata.
5139 26. If the domain in control=submission is set empty, Exim assumes that the
5140 authenticated id is a complete email address when it generates From: or
5141 Sender: header lines.
5143 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
5144 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
5145 chown and chgrp in /bin and hostname in /usr/bin.
5147 28. Exim was keeping the "process log" file open after each use, just as it
5148 does for the main log. This opens the possibility of it remaining open for
5149 long periods when the USR1 signal hits a daemon. Occasional processlog
5150 errors were reported, that could have been caused by this. Anyway, it seems
5151 much more sensible not to leave this file open at all, so that is what now
5154 29. The long-running daemon process does not normally write to the log once it
5155 has entered its main loop, and it closes the log before doing so. This is
5156 so that log files can straightforwardly be renamed and moved. However,
5157 there are a couple of unusual error situations where the daemon does write
5158 log entries, and I had neglected to close the log afterwards.
5160 30. The text of an SMTP error response that was received during a remote
5161 delivery was being truncated at 512 bytes. This is too short for some of
5162 the long messages that one sometimes sees. I've increased the limit to
5165 31. It is now possible to make retry rules that apply only when a message has a
5166 specific sender, in particular, an empty sender.
5168 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
5169 it possible to be selective about when SMTP synchronization is enforced.
5171 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
5173 32. Implemented hosts_connection_nolog.
5175 33. Added an ACL for QUIT.
5177 34. Setting "delay_warning=" to disable warnings was not working; it gave a
5180 35. Added mailbox_size and mailbox_filecount to appendfile.
5182 36. Added control = no_multiline_responses to ACLs.
5184 37. There was a bug in the logic of the code that waits for the clock to tick
5185 in the case where the clock went backwards by a substantial amount such
5186 that the microsecond fraction of "now" was more than the microsecond
5187 fraction of "then" (but the whole seconds number was less).
5189 38. Added support for the libradius Radius client library this is found on
5190 FreeBSD (previously only the radiusclient library was supported).
5196 1. When certain lookups returned multiple values in the form name=value, the
5197 quoting of the values was not always being done properly. Specifically:
5198 (a) If the value started with a double quote, but contained no whitespace,
5200 (b) If the value contained whitespace other than a space character (i.e.
5201 tabs or newlines or carriage returns) it was not quoted.
5202 This fix has been applied to the mysql and pgsql lookups by writing a
5203 separate quoting function and calling it from the lookup code. The fix
5204 should probably also be applied to nisplus, ibase and oracle lookups, but
5205 since I cannot test any of those, I have not disturbed their existing code.
5207 2. A hit in the callout cache for a specific address caused a log line with no
5208 reason for rejecting RCPT. Now it says "Previous (cached) callout
5209 verification failure".
5211 3. There was an off-by-one bug in the queryprogram router. An over-long
5212 return line was truncated at 256 instead of 255 characters, thereby
5213 overflowing its buffer with the terminating zero. As well as fixing this, I
5214 have increased the buffer size to 1024 (and made a note to document this).
5216 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
5217 when Exim is waiting for an SMTP response from a remote server, Exim
5218 restarts its select() call on the socket, thereby resetting its timeout.
5219 This is not a problem when such interrupts are rare. Somebody set up a cron
5220 job to run exiwhat every 2 minutes, which is less than the normal select()
5221 timeout (5 or 10 minutes). This meant that the select() timeout never
5222 kicked in because it was always reset. I have fixed this by comparing the
5223 time when an interrupt arrives with the time at the start of the first call
5224 to select(). If more time than the timeout has elapsed, the interrupt is
5225 treated as a timeout.
5227 5. Some internal re-factoring in preparation for the addition of Sieve
5228 extensions (by MH). In particular, the "personal" test is moved to a
5229 separate function, and given an option for scanning Cc: and Bcc: (which is
5230 not set for Exim filters).
5232 6. When Exim created an email address using the login of the caller as the
5233 local part (e.g. when creating a From: or Sender: header line), it was not
5234 quoting the local part when it contained special characters such as @.
5236 7. Installed new OpenBSD configuration files.
5238 8. Reworded some messages for syntax errors in "and" and "or" conditions to
5239 try to make them clearer.
5241 9. Callout options, other than the timeout value, were being ignored when
5242 verifying sender addresses in header lines. For example, when using
5244 verify = header_sender/callout=no_cache
5246 the cache was (incorrectly) being used.
5248 10. Added a missing instance of ${EXE} to the exim_install script; this affects
5249 only the Cygwin environment.
5251 11. When return_path_on_delivery was set as a log selector, if different remote
5252 addresses in the same message used different return paths and parallel
5253 remote delivery occurred, the wrong values would sometimes be logged.
5254 (Whenever a remote delivery process finished, the return path value from
5255 the most recently started remote delivery process was logged.)
5257 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
5258 lines when AUTH and/or TLS are in use. This is the "received protocol"
5259 field. Exim used to use "asmtp" for authenticated SMTP, without any
5260 indication (in the protocol name) for TLS use. Now it follows the RFC and
5261 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
5262 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
5263 names appear in log lines as well as in Received: header lines.
5265 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
5266 extensions, and comparison tests, and to fix some bugs.
5268 14. Changes to the "personal" filter test:
5270 (1) The test was buggy in that it was just doing the equivalent of
5271 "contains" tests on header lines. For example, if a user's address was
5272 anne@some.where, the "personal" test would incorrectly be true for
5274 To: susanne@some.where
5276 This test is now done by extracting each address from the header in turn,
5277 and checking the entire address. Other tests that are part of "personal"
5278 are now done using regular expressions (for example, to check local parts
5279 of addresses in From: header lines).
5281 (2) The list of non-personal local parts in From: addresses has been
5282 extended to include "listserv", "majordomo", "*-request", and "owner-*",
5283 taken from the Sieve specification recommendations.
5285 (3) If the message contains any header line starting with "List-" it is
5286 treated as non-personal.
5288 (4) The test for "circular" in the Subject: header line has been removed
5289 because it now seems ill-conceived.
5291 15. Minor typos in src/EDITME comments corrected.
5293 16. Installed latest exipick from John Jetmore.
5295 17. If headers_add on a router specified a text string that was too long for
5296 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
5297 of string_sprintf() is now avoided.
5299 18. $message_body_size was not set (it was always zero) when running the DATA
5300 ACL and the local_scan() function.
5302 19. For the "mail" command in an Exim filter, no default was being set for
5303 the once_repeat time, causing a random time value to be used if "once" was
5304 specified. (If the value happened to be <= 0, no repeat happened.) The
5305 default is now 0s, meaning "never repeat". The "vacation" command was OK
5306 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
5307 (I found it when inspecting the code).
5309 20. There is now an overall timeout for performing a callout verification. It
5310 defaults to 4 times the callout timeout, which applies to individual SMTP
5311 commands during the callout. The overall timeout applies when there is more
5312 than one host that can be tried. The timeout is checked before trying the
5313 next host. This prevents very long delays if there are a large number of
5314 hosts and all are timing out (e.g. when the network connections are timing
5315 out). The value of the overall timeout can be changed by specifying an
5316 additional sub-option for "callout", called "maxwait". For example:
5318 verify = sender/callout=5s,maxwait=20s
5320 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
5321 to the end before writing, but this should make it even safer).
5323 22. Exim was forgetting that it had advertised PIPELINING for the second and
5324 subsequent messages on an SMTP connection. It was also not resetting its
5325 memory on STARTTLS and an internal HELO.
5327 23. When Exim logs an SMTP synchronization error within a session, it now
5328 records whether PIPELINING has been advertised or not.
5330 24. Added 3 instances of "(long int)" casts to time_t variables that were being
5331 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
5332 rather than long int.
5334 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
5336 26. Added the never_mail option to autoreply.
5342 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
5343 crash if the getsockname() call failed; this can happen if a connection is
5344 closed very soon after it is established. The problem was simply in the
5345 order in which certain operations were done, causing Exim to try to write
5346 to the SMTP stream before it had set up the file descriptor. The bug has
5347 been fixed by making things happen in the correct order.
5353 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
5354 before the connection was closed, thus losing the rejection response.
5356 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
5357 some early Solaris releases, but causes trouble in current releases where
5358 socklen_t is defined.
5360 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
5363 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
5366 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
5367 that releasing the top of it at the end releases what was used for sub-
5368 expansions (unless the block got too big). However, discard this block if
5369 the first thing is a variable or header, so that we can use its block when
5370 it is dynamic (useful for very large $message_headers, for example).
5372 6. Lookups now cache *every* query, not just the most recent. A new, separate
5373 store pool is used for this. It can be recovered when all lookup caches are
5374 flushed. Lookups now release memory at the end of their result strings.
5375 This has involved some general refactoring of the lookup sources.
5377 7. Some code has been added to the store_xxx() functions to reduce the amount
5378 of flapping under certain conditions.
5380 8. log_incoming_interface used to affect only the <= reception log lines. Now
5381 it causes the local interface and port to be added to several more SMTP log
5382 lines, for example "SMTP connection from", and rejection lines.
5384 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
5386 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
5388 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
5389 could be overwritten at the end of the current message (or the start of a
5390 new message if it was set in a HELO ACL). The value is now preserved for
5391 the duration of the SMTP connection.
5393 12. If a transport had a headers_rewrite setting, and a matching header line
5394 contained an unqualified address, that address was qualified, even if it
5395 did not match any rewriting rules. The underlying bug was that the values
5396 of the flags that permit the existence of unqualified sender and recipient
5397 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
5398 non-local messages, and by -bnq for local messages) were not being
5399 preserved with the message after it was received.
5401 13. When Exim was logging an SMTP synchronization error, it could sometimes log
5402 "next input=" as part of the text comprising the host identity instead of
5403 the correct text. The code was using the same buffer for two different
5404 strings. However, depending on which order the printing function evaluated
5405 its arguments, the bug did not always show up. Under Linux, for example, my
5406 test suite worked just fine.
5408 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
5409 doesn't work with some older versions of Perl. It has been changed to "my",
5410 which in any case is probably the better facility to use.
5412 15. A really picky compiler found some instances of statements for creating
5413 error messages that either had too many or two few arguments for the format
5416 16. The size of the buffer for calls to the DNS resolver has been increased
5417 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
5418 for addresses that have a lot of PTR records. This alleviates a problem; it
5419 does not fully solve it.
5421 17. A dnsdb lookup for PTR records that receives more data than will fit in the
5422 buffer now truncates the list and logs the incident, which is the same
5423 action as happens when Exim is looking up a host name and its aliases.
5424 Previously in this situation something unpredictable would happen;
5425 sometimes it was "internal error: store_reset failed".
5427 18. If a server dropped the connection unexpectedly when an Exim client was
5428 using GnuTLS and trying to read a response, the client delivery process
5429 crashed while trying to generate an error log message.
5431 19. If a "warn" verb in an ACL added multiple headers to a message in a single
5432 string, for example:
5434 warn message = H1: something\nH2: something
5436 the text was added as a single header line from Exim's point of view
5437 though it ended up OK in the delivered message. However, searching for the
5438 second and subsequent header lines using $h_h2: did not work. This has been
5439 fixed. Similarly, if a system filter added multiple headers in this way,
5440 the routers could not see them.
5442 20. Expanded the error message when iplsearch is called with an invalid key to
5443 suggest using net-iplsearch in a host list.
5445 21. When running tests using -bh, any delays imposed by "delay" modifiers in
5446 ACLs are no longer actually imposed (and a message to that effect is
5449 22. If a "gecos" field in a passwd entry contained escaped characters, in
5450 particular, if it contained a \" sequence, Exim got it wrong when building
5451 a From: or a Sender: header from that name. A second bug also caused
5452 incorrect handling when an unquoted " was present following a character
5453 that needed quoting.
5455 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
5456 was not being matched caselessly.
5458 24. Arranged for all hyphens in the exim.8 source to be escaped with
5461 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
5462 properly. Recipient callout cache records were still being keyed to include
5463 the sender, even when use_sender was set false. This led to far more
5464 callouts that were necessary. The sender is no longer included in the key
5465 when use_sender is false.
5467 26. Added "control = submission" modifier to ACLs.
5469 27. Added the ${base62d: operator to decode base 62 numbers.
5471 28. dnsdb lookups can now access SRV records.
5473 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
5474 the configuration file.
5476 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
5477 (-v) mode. This makes the output for a verbose queue run more intelligible.
5479 31. Added a use_postmaster feature to recipient callouts.
5481 32. Added the $body_zerocount variable, containing the number of binary zero
5482 bytes in the message body.
5484 33. The time of last modification of the "new" subdirectory is now used as the
5485 "mailbox time last read" when there is a quota error for a maildir
5488 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
5490 35. Added +ignore_unknown as a special item in host lists.
5492 36. Code for decoding IPv6 addresses in host lists is now included, even if
5493 IPv6 support is not being compiled. This fixes a bug in which an IPv6
5494 address was recognized as an IP address, but was then not correctly decoded
5495 into binary, causing unexpected and incorrect effects when compared with
5502 1. Very minor rewording of debugging text in manualroute to say "list of
5503 hosts" instead of "hostlist".
5505 2. If verify=header_syntax was set, and a header line with an unqualified
5506 address (no domain) and a large number of spaces between the end of the
5507 name and the colon was received, the reception process suffered a buffer
5508 overflow, and (when I tested it) crashed. This was caused by some obsolete
5509 code that should have been removed. The fix is to remove it!
5511 3. When running in the test harness, delay a bit after writing a bounce
5512 message to get a bit more predictability in the log output.
5514 4. Added a call to search_tidyup() just before forking a reception process. In
5515 theory, someone could use a lookup in the expansion of smtp_accept_max_
5516 per_host which, without the tidyup, could leave open a database connection.
5518 5. Added the variables $recipient_data and $sender_data which get set from a
5519 lookup success in an ACL "recipients" or "senders" condition, or a router
5520 "senders" option, similar to $domain_data and $local_part_data.
5522 6. Moved the writing of debug_print from before to after the "senders" test
5525 7. Change 4.31/66 (moving the time when the Received: is generated) caused
5526 problems for message scanning, either using a data ACL, or using
5527 local_scan() because the Received: header was not generated till after they
5528 were called (in order to set the time as the time of reception completion).
5529 I have revised the way this works. The header is now generated after the
5530 body is received, but before the ACL or local_scan() are called. After they
5531 are run, the timestamp in the header is updated.
5537 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
5538 before starting a queue runner without re-exec. This happened only when
5539 deliver_drop_privilege was set or when the Exim user was set to root. The
5540 effect of the bug was that timeouts during subsequent deliveries caused
5541 crashes instead of being properly handled. The handler is now left at its
5542 default (and expected) setting.
5544 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
5545 message, again when deliver_drop_privilege is set or Exim is run as root.
5546 The bug described in (1) was not present in this case, but the tidying up
5547 of the other signals was missing. I have made the two cases consistent.
5549 3. The ignore_target_hosts setting on a manualroute router was being ignored
5550 for hosts that were looked up using the /MX notation.
5552 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
5555 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
5556 operated on the sender address. After changing the $sender_address to <>
5557 for the sender address verify, Exim was re-instated it as the original
5558 (before rewriting) address, but remembering that it had rewritten it, so it
5559 wasn't rewriting it again. This bug also had the effect of breaking the
5560 sender address verification caching when the sender address was rewritten.
5562 6. The ignore_target_hosts option was being ignored by the ipliteral router.
5563 This has been changed so that if the ip literal address matches
5564 ignore_target_hosts, the router declines.
5566 7. Added expansion conditions match_domain, match_address, and match_local_
5567 part (NOT match_host).
5569 8. The placeholder for the Received: header didn't have a length field set.
5571 9. Added code to Exim itself and to exim_lock to test for a specific race
5572 condition that could lead to file corruption when using MBX delivery. The
5573 issue is with the lockfile that is created in /tmp. If this file is removed
5574 after a process has opened it but before that process has acquired a lock,
5575 there is the potential for a second process to recreate the file and also
5576 acquire a lock. This could lead to two Exim processes writing to the file
5577 at the same time. The added code performs the same test as UW imapd; it
5578 checks after acquiring the lock that its file descriptor still refers to
5579 the same named file.
5581 10. The buffer for building added header lines was of fixed size, 8192 bytes.
5582 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
5585 11. Added the smtp_active_hostname option. If used, this will typically be made
5586 to depend on the incoming interface address. Because $interface_address is
5587 not set up until the daemon has forked a reception process, error responses
5588 that can happen earlier (such as "too many connections") no longer contain
5591 12. If an expansion in a condition on a "warn" statement fails because a lookup
5592 defers, the "warn" statement is abandoned, and the next ACL statement is
5593 processed. Previously this caused the whole ACL to be aborted.
5595 13. Added the iplsearch lookup type.
5597 14. Added ident_timeout as a log selector.
5599 15. Added tls_certificate_verified as a log selector.
5601 16. Added a global option tls_require_ciphers (compare the smtp transport
5602 option of the same name). This controls incoming TLS connections.
5604 17. I finally figured out how to make tls_require_ciphers do a similar thing
5605 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
5606 before starting the TLS session.
5608 18. Tabs are now shown as \t in -bP output.
5610 19. If the log selector return_path_on_delivery was set, Exim crashed when
5611 bouncing a message because it had too many Received: header lines.
5613 20. If two routers both had headers_remove settings, and the first one included
5614 a superfluous trailing colon, the final name in the first list and the
5615 first name in the second list were incorrectly joined into one item (with a
5616 colon in the middle).
5622 1. Added -C and -D options to the exinext utility, mainly to make it easier
5623 to include in the automated testing, but these could be helpful when
5624 multiple configurations are in use.
5626 2. The exinext utility was not formatting the output nicely when there was
5627 an alternate port involved in the retry record key, nor when there was a
5628 message id as well (for retries that were specific to a specific message
5629 and a specific host). It was also confused by IPv6 addresses, because of
5630 the additional colons they contain. I have fixed the IPv4 problem, and
5631 patched it up to do a reasonable job for IPv6.
5633 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
5634 delivery, the log line now contains "pipelined" if PIPELINING was used.
5636 4. An SMTP transport process used to panic and die if the bind() call to set
5637 an explicit outgoing interface failed. This has been changed; it is now
5638 treated in the same way as a connect() failure.
5640 5. A reference to $sender_host_name in the part of a conditional expansion
5641 that was being skipped was still causing a DNS lookup. This no longer
5644 6. The def: expansion condition was not recognizing references to header lines
5645 that used bh_ and bheader_.
5647 7. Added the _cache feature to named lists.
5649 8. The code for checking quota_filecount in the appendfile transport was
5650 allowing one more file than it should have been.
5652 9. For compatibility with Sendmail, the command line option
5660 and sets the incoming protocol and host name (for trusted callers). The
5661 host name and its colon can be omitted when only the protocol is to be set.
5662 Note the Exim already has two private options, -pd and -ps, that refer to
5663 embedded Perl. It is therefore impossible to set a protocol value of "d" or
5664 "s", but I don't think that's a major issue.
5666 10. A number of refactoring changes to the code, none of which should affect
5669 (a) The number of logging options was getting close to filling up the
5670 32-bit word that was used as a bit map. I have split them into two classes:
5671 those that are passed in the argument to log_write(), and those that are
5672 only ever tested independently outside of that function. These are now in
5673 separate 32-bit words, so there is plenty of room for expansion again.
5674 There is no change in the user interface or the logging behaviour.
5676 (b) When building, for example, log lines, the code previously used a
5677 macro that called string_cat() twice, in order to add two strings. This is
5678 not really sufficiently general. Furthermore, there was one instance where
5679 it was actually wrong because one of the argument was used twice, and in
5680 one call a function was used. (As it happened, calling the function twice
5681 did not affect the overall behaviour.) The macro has been replaced by a
5682 function that can join an arbitrary number of extra strings onto a growing
5685 (c) The code for expansion conditions now uses a table and a binary chop
5686 instead of a serial search (which was left over from when there were very
5687 few conditions). Also, it now recognizes conditions like "pam" even when
5688 the relevant support is not compiled in: a suitably worded error message is
5689 given if an attempt is made to use such a condition.
5691 11. Added ${time_interval:xxxxx}.
5693 12. A bug was causing one of the ddress fields not to be passed back correctly
5694 from remote delivery subprocesses. The field in question was not being
5695 subsequently used, so this caused to problems in practice.
5697 13. Added new log selectors queue_time and deliver_time.
5699 14. Might have fixed a bug in maildirsizefile handling that threw up
5700 "unexpected character" debug warnings, and recalculated the data
5701 unnecessarily. In any case, I expanded the warning message to give more
5704 15. Added the message "Restricted characters in address" to the statements in
5705 the default ACL that block characters like @ and % in local parts.
5707 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
5708 Three changes have been made:
5710 (a) There was a serious bug; a negative response to MAIL caused the whole
5711 recipient domain to be cached as invalid, thereby blocking all messages
5712 to all local parts at the same domain, from all senders. This bug has
5713 been fixed. The domain is no longer cached after a negative response to
5714 MAIL if the sender used is not empty.
5716 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
5719 (c) A new callout option, "use_sender" has been added for people who want
5720 the modified behaviour.
5726 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
5729 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
5730 indeed breaks things for older releases.
5732 3. Added additional logging to the case where there is a problem reading data
5733 from a filter that is running in a subprocess using a pipe, in order to
5734 try to track down a specific problem.
5736 4. Testing facility fudge: when running in the test harness and attempting
5737 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
5738 getting "No route to host". Convert this to a timeout.
5740 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
5743 6. Some OS don't have socklen_t but use size_t instead. This affects the
5744 fifth argument of getsockopt() amongst other things. This is now
5745 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
5746 can be set for individual OS. I have set it for SunOS5, OSF1, and
5747 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
5748 some earlier ones do not.
5750 7. Change 4.30/15 was not doing the test caselessly.
5752 8. The standard form for an IPv6 address literal was being rejected by address
5753 parsing in, for example, MAIL and RCPT commands. An example of this kind of
5754 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5755 this, as well as the form without the "IPv6" on the front (but only when
5756 address literals are enabled, of course).
5758 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
5760 10. Exim crashed if a message with an empty sender address specified by -f
5761 encountered a router with an errors_to setting. This could be provoked only
5762 by a command such as
5766 where an empty string was supplied; "<>" did not hit this bug.
5768 11. Installed PCRE release 4.5.
5770 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
5771 remained set. It is now erased.
5773 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
5774 times from message ids (which are base 36 rather than the normal 62).
5776 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
5777 were being counted as actual protocol errors, and logged if the log
5778 selector +smtp_protocol_error was set. One cannot be perfect in this test,
5779 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
5780 and DATA following a set of rejected RCPTs do not count as protocol errors.
5781 In other words, Exim assumes they were pipelined, though this may not
5782 actually be the case. Of course, in all cases the client gets an
5783 appropriate error code.
5785 15. If a lookup fails in an ACL condition, a message about the failure may
5786 be available; it is used if testing the ACL cannot continue, because most
5787 such messages specify what the cause of the deferral is. However, some
5788 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
5789 that caused an old message to be retained and used if a later statement
5790 caused a defer, replacing the real cause of the deferral.
5792 16. If an IP address had so many PTR records that the DNS lookup buffer
5793 was not large enough to hold them, Exim could crash while trying to process
5794 the truncated data. It now detects and logs this case.
5796 17. Further to 4.21/58, another change has been made: if (and only if) the
5797 first line of a message (the first header line) ends with CRLF, a bare LF
5798 in a subsequent header line has a space inserted after it, so as not to
5799 terminate the header.
5801 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
5802 unnecessarily, used atoi() instead of strtol(), and didn't check the
5803 termination when getting file sizes from file names by regex.
5805 19. Completely re-implemented the support for maildirsize files, in the light
5806 of a number of problems with the previous contributed implementation
5807 (4.30/29). In particular:
5809 . If the quota is zero, the maildirsize file is maintained, but no quota is
5812 . If the maildir directory does not exist, it is created before any attempt
5813 to write a maildirsize file.
5815 . The quota value in the file is just a cache; if the quota is changed in
5816 the transport, the new value overrides.
5818 . A regular expression is available for excluding directories from the
5821 20. The autoreply transport checks the characters in options that define the
5822 message's headers; it allows continued headers, but it was checking with
5823 isspace() after an embedded newline instead of explicitly looking for a
5826 21. If all the "regular" hosts to which an address was routed had passed their
5827 expiry times, and had not reached their retry times, the address was
5828 bounced, even if fallback hosts were defined. Now Exim should go on to try
5831 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
5832 equivalent code in the SMTP transport. Some hosts send humungous responses
5833 to HELO/EHLO, more than 1024 it seems.
5835 23. Refactoring: code in filter.c used (void *) for "any old type" but this
5836 gives compiler warnings in some environments. I've now done it "properly",
5839 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
5840 (because of problems with the built-in one) was declared to return uschar *
5841 instead of char *, causing compiler failure.
5843 25. Fixed a file descriptor leak when processing alias/forward files.
5845 26. Fixed a minor format string issue in dbfn.c.
5847 27. Typo in exim.c: ("dmbnz" for "dbmnz").
5849 28. If a filter file refered to $h_xxx or $message_headers, and the headers
5850 contained RFC 2047 "words", Exim's memory could, under certain conditions,
5853 29. When a sender address is verified, it is cached, to save repeating the test
5854 when there is more than one recipient in a message. However, when the
5855 verification involves a callout, it is possible for different callout
5856 options to be set for different recipients. It is too complicated to keep
5857 track of this in the cache, so now Exim always runs a verification when a
5858 callout is required, relying on the callout cache for the optimization.
5859 The overhead is duplication of the address routing, but this should not be
5862 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
5863 to be verified with callout=postmaster, and the main callout worked but the
5864 postmaster check failed, the verification correctly failed. However, if a
5865 subsequent RCPT command asked for sender verification *without* the
5866 postmaster check, incorrect caching caused this verification also to fail,
5869 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
5870 it was not caching the DNS options (qualify_single, search_parents) that
5871 were used when the lookup failed. A subsequent lookup with different
5872 options therefore always gave the same answer, though there were cases
5873 where it should not have. (Example: a "domains = !$mx_any" option on a
5874 dnslookup router: the "domains" option is always processed without any
5875 widening, but the router might have qualify_single set.) Now Exim uses the
5876 cached value only when the same options are set.
5878 32. Added John Jetmore's "exipick" utility to the distribution.
5880 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
5881 than a timeout (e.g. a certificate is required, and is not provided), an
5882 Exim server now closes the connection immediately. Previously it waited for
5883 the client to close - but if the client is SSL, it seems that they each
5884 wait for each other, leading to a delay before one of them times out.
5886 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
5887 maintained 0.8.x compatibility because I don't think many are using it, and
5888 it is clearly obsolete.
5890 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
5893 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
5894 client certificate was expired. A simple patch fixes this, though I don't
5895 understand the full logic of why the verify callback is called multiple
5898 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
5899 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
5900 which causes problems with some clients (such as the Certicom SSL Plus
5901 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
5902 disables the coutermeasure allowing Eudora to connect."
5904 38. Exim was not checking that a write() to a log file succeeded. This could
5905 lead to Bad Things if a log got too big, in particular if it hit a file
5906 size limit. Exim now panics and dies if it cannot write to a log file, just
5907 as it does if it cannot open a log file.
5909 39. Modified OS/Makefile-Linux so that it now contains
5911 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
5913 The two -D definitions ensure that Exim is compiled with large file
5914 support, which makes it possible to handle log files that are bigger than
5917 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
5918 instance) a domain was checked against a named list that involved a lookup,
5919 causing $domain_data to be set, then another domain was checked against the
5920 same list, then the first domain was re-checked, the value of $domain_data
5921 after the final check could be wrong. In particular, if the second check
5922 failed, it could be set empty. This bug probably also applied to
5925 41. The strip_trailing_dot option was not being applied to the address given
5926 with the -f command-line option.
5928 42. The code for reading a message's header from the spool was incrementing
5929 $received_count, but never initializing it. This meant that the value was
5930 incorrect (doubled) while delivering a message in the same process in which
5931 it was received. In the most common configuration of Exim, this never
5932 happens - a fresh exec is done - but it can happen when
5933 deliver_drop_privilege is set.
5935 43. When Exim logs an SMTP synchronization error - client data sent too soon -
5936 it now includes up to 150 characters of the unexpected data in the log
5939 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
5940 and building data strings. The size of both of these buffers was 10 000
5941 bytes - far larger than anybody would *ever* want, thought I. Needless to
5942 say, somebody hit the limit. I have increased the maximum line length to
5943 20 000 and the maximum data length of concatenated lines to 100 000. I have
5944 also fixed two bugs, because there was no checking on these buffers. Tsk,
5945 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
5946 buffer is too small.
5948 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
5949 lsearch lookups. Now it does.
5951 46. When parsing a route_list item in a manualroute router, a fixed-length
5952 buffer was used for the list of hosts. I made this 1024 bytes long,
5953 thinking that nobody would ever have a list of hosts that long. Wrong.
5954 Somebody had a whole pile of complicated expansion conditions, and the
5955 string was silently truncated, leading to an expansion error. It turns out
5956 that it is easier to change to an unlimited length (owing to other changes
5957 that have happened since this code was originally written) than to build
5958 structure for giving a limitation error. The length of the item that
5959 expands into the list of hosts is now unlimited.
5961 47. The lsearch lookup could not handle data where the length of text line was
5962 more than 4095 characters. Such lines were truncated, leading to shortened
5963 data being returned. It should now handle lines of any length.
5965 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
5966 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
5969 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
5970 replacing strings such as BIN_DIRECTORY in a source file: the replacement
5971 no longer happens in comment lines. A list of replacements is now placed
5972 at the head of all of the source files, except those whose only change is
5973 to replace PERL_COMMAND in the very first #! line.
5975 50. Replaced the slow insertion sort in queue.c, for sorting the list of
5976 messages on the queue, with a bottom-up merge sort, using code contributed
5977 by Michael Haardt. This should make operations like -bp somewhat faster on
5978 large queues. It won't affect queue runners, except when queue_run_in_order
5981 51. Installed eximstats 1.31 in the distribution.
5983 52. Added support for SRV lookups to the dnslookup router.
5985 53. If an ACL referred to $message_body or $message_body_end, the value was not
5986 reset for any messages that followed in the same SMTP session.
5988 54. The store-handling optimization for building very long strings was not
5989 differentiating between the different store pools. I don't think this
5990 actually made any difference in practice, but I've tidied it.
5992 55. While running the routers to verify a sender address, $sender_address
5993 was still set to the sender address. This is wrong, because when routing to
5994 send a bounce to the sender, it would be empty. Therefore, I have changed
5995 it so that, while verifying a sender address, $sender_address is set to <>.
5996 (There is no change to what happens when verifying a recipient address.)
5998 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
5999 target A or AAAA records (if not already returned) without resetting the
6000 qualify_single or search_parents options of the DNS resolver. These are
6001 inappropriate in this case because the targets of MX and SRV records must
6002 be FQDNs. A broken DNS record could cause trouble if it happened to have a
6003 target that, when qualified, matched something in the local domain. These
6004 two options are now turned off when doing these lookups.
6006 57. It seems that at least some releases of Reiserfs (which does not have the
6007 concept of a fixed number of inodes) returns zero and not -1 for the
6008 number of available inodes. This interacted badly with check_spool_inodes,
6009 which assumed that -1 was the "no such thing" setting. What I have done is
6010 to check that the total number of inodes is greater than zero before doing
6011 the test of how many are available.
6013 58. When a "warn" ACL statement has a log_message modifier, the message is
6014 remembered, and not repeated. This is to avoid a lot of repetition when a
6015 message has many recipients that cause the same warning to be written.
6016 However, Exim was preserving the list of already written lines for an
6017 entire SMTP session, which doesn't seem right. The memory is now reset if a
6018 new message is started.
6020 59. The "rewrite" debugging flag was not showing the result of rewriting in the
6021 debugging output unless log_rewrite was also set.
6023 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
6024 of (int)(handle) when we know that handle contains (void *)(-1).
6026 61. The Exim daemon panic-logs an error return when it closes the incoming
6027 connection. However "connection reset by peer" seems to be common, and
6028 isn't really an error worthy of noting specially, so that particular error
6031 62. When Exim is trying to find all the local interfaces, it used to panic and
6032 die if the ioctl to get the interface flags failed. However, it seems that
6033 on at least one OS (Solaris 9) it is possible to have an interface that is
6034 included in the list of interfaces, but for which you get a failure error
6035 for this call. This happens when the interface is not "plumbed" into a
6036 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
6037 failure of the "get flags" call assumes that the interface is down.
6039 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
6040 makes life easier for people who are doing arithmetic on fields extracted
6041 from dates, where you often get leading zeros that should not be
6042 interpreted as octal.
6044 64. Added qualify_domain to the redirect router, to override the global
6047 65. If a pathologically long header line contained very many addresses (the
6048 report of this problem mentioned 10 000) and each of them was rewritten,
6049 Exim could use up a very large amount of memory. (It kept on making new
6050 copies of the header line as it rewrote, and never released the old ones.)
6051 At the expense of a bit more processing, the header rewriting function has
6052 been changed so that it no longer eats memory in this way.
6054 66. The generation of the Received: header has been moved from the time that a
6055 message starts to be received, to the time that it finishes. The timestamp
6056 in the Received: header should now be very close to that of the <= log
6057 line. There are two side-effects of this change:
6059 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
6060 logged header lines no longer include the local Received: line, because
6061 it has not yet been created. The same applies to a copy of the message
6062 that is returned to a non-SMTP sender when a message is rejected.
6064 (b) When a filter file is tested using -bf, no additional Received: header
6065 is added to the test message. After some thought, I decided that this
6068 This change does not affect the value of $received_for. It is still set
6069 after address rewriting, but before local_scan() is called.
6071 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
6073 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
6074 gave an unhelpful panic error message, and a defer error. I have managed to
6075 change this behaviour so that it now rejects any supplied certificate,
6076 which seems right, as the list of acceptable certificates is empty.
6078 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
6079 gave an unhelpful defer error. I have not managed to make this reject any
6080 supplied certificates, but the error message it gives is "no certificate
6081 supplied", which is not helpful.
6083 70. exigrep's output now also includes lines that are not associated with any
6084 message, but which match the given pattern. Implemented by a patch from
6085 Martin Sluka, which also tidied up the Perl a bit.
6087 71. Recipient callout verification, like sender verification, was using <> in
6088 the MAIL FROM command. This isn't really the right thing, since the actual
6089 sender may affect whether the remote host accepts the recipient or not. I
6090 have changed it to use the actual sender in the callout; this means that
6091 the cache record is now keyed on a recipient/sender pair, not just the
6092 recipient address. There doesn't seem to be a real danger of callout loops,
6093 since a callout by the remote host to check the sender would use <>.
6094 [SEE ABOVE: changed after hitting problems.]
6096 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
6097 temporary errors. However, in the case of such a code being given after
6098 the end of a data transmission (i.e. after ".") Exim was failing to write
6099 a retry record for the message. (Yes, there was some broken host that was
6100 actually sending 8xx at this point.)
6102 73. An unknown lookup type in a host list could cause Exim to panic-die when
6103 the list was checked. (An example that provoked this was putting <; in the
6104 middle of a list instead of at the start.) If this happened during a DATA
6105 ACL check, a -D file could be left lying around. This kind of configuration
6106 error no longer causes Exim to die; instead it causes a defer error. The
6107 incident is still logged to the main and panic logs.
6109 74. Buglet left over from Exim 3 conversion. The message "too many messages
6110 in one connection" was written to the rejectlog but not the mainlog, except
6111 when address rewriting (yes!) was being logged.
6113 75. Added write_rejectlog option.
6115 76. When a system filter was run not as root (that is, when system_filter_user
6116 was set), the values of the $n variables were not being returned to the
6117 main process; thus, they were not subsequently available in the $sn
6120 77. Added +return_path_on_delivery log selector.
6122 78. A connection timeout was being treated differently from recipients deferred
6123 when testing hosts_max_try with a message that was older than the host's
6124 retry timeout. (The host should not be counted, thus allowing all hosts to
6125 be tried at least once before bouncing.) This may have been the cause of an
6126 occasionally reported bug whereby a message would remain on the queue
6127 longer than the retry timeout, but would be bounced if a delivery was
6128 forced. I say "may" because I never totally pinned down the problem;
6129 setting up timeout/retry tests is difficult. See also the next item.
6131 79. The ultimate address timeout was not being applied to errors that involved
6132 a combination of host plus message (for example, a timeout on a MAIL
6133 command). When an address resolved to a number of possible hosts, and they
6134 were not all tried for each delivery (e.g. because of hosts_max_try), a
6135 message could remain on the queue longer than the retry timeout.
6137 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
6140 81. Fixed an obscure SMTP outgoing bug which required at least the following
6141 conditions: (a) there was another message waiting for the same server;
6142 (b) the server returned 5xx to all RCPT commands in the first message so
6143 that the message was not completed; (c) the server dropped the connection
6144 or gave a negative response to the RSET that Exim sends to abort the
6145 transaction. The observed case was a dropped connection after DATA that had
6146 been sent in pipelining mode. That is, the server had advertised PIPELINING
6147 but was not implementing it correctly. The effect of the bug was incorrect
6148 behaviour, such as trying another host, and this could lead to a crash.
6154 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
6155 and daemon.c were passed as pointers to ints; they should have been
6156 pointers to socklen_t variables (which are typically unsigned ints).
6158 2. Some signed/unsigned type warnings in the os.c file for Linux have been
6161 3. Fixed a really odd bug that affected only the testing scheme; patching a
6162 certain fixed string in the binary changed the value of another string that
6163 happened to be identical to the end of the original first string.
6165 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
6166 name", it returns that address as the IP address. On some operating
6167 systems (e.g. Solaris), it also passes back the IP address string as the
6168 "host name". However, on others (e.g. Linux), it passes back an empty
6169 string. Exim wasn't checking for this, and was changing the host name to an
6170 empty string, assuming it had been canonicalized.
6172 5. Although rare, it is permitted to have more than one PTR record for a given
6173 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
6174 all the names associated with an address, because they do in Solaris.
6175 However, it seems that they do not in Linux for data that comes from the
6176 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
6177 I found this out when I moved to a new Linux workstation and tried to run
6178 the Exim test suite.
6180 To get round this problem I have changed the code so that it now does its
6181 own call to the DNS to look up PTR records when searching for a host name.
6182 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
6183 addresses that are only in /etc/hosts are still found.
6185 This behaviour is, however, controlled by an option called host_lookup_
6186 order, which defaults to "bydns:byaddr". If people want to use the other
6187 order, or indeed, just use one or the other means of lookup, they can
6188 specify it in this variable.
6190 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
6191 some operating systems, this comes back from gethostbyaddr() as an empty
6192 string, and this is what Exim used to test for. However, it seems that in
6193 other systems, "." is yielded. Exim now tests for this case too.
6195 7. The values of check_spool_space and check_log_space are now held internally
6196 as a number of kilobytes instead of an absolute number of bytes. If a
6197 numbers is specified without 'K' or 'M', it is rounded up to the nearest
6198 kilobyte. This means that much larger values can be stored.
6200 8. Exim monitor: an attempt to get the action menu when not actually pointing
6201 at a message produces an empty menu entitled "No message selected". This
6202 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
6203 no entries in it ("Shell widget menu has zero width and/or height"). So I
6204 have added a single, blank menu entry in this case.
6206 9. Added ${quote_local_part.
6208 10. MIME decoding is now applied to the contents of Subject: header lines when
6211 11. Now that a reference to $sender_host_address automatically causes a reverse
6212 lookup to occur if necessary (4.13/18), there is no need to arrange for a
6213 host lookup before query-style lookups in lists that might use this
6214 variable. This has therefore been abolished, and the "net-" prefix is no
6215 longer necessary for query-style lookups.
6217 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
6218 have been a typo for LFLAGS, so it has been changed.
6220 13. The install script calls Exim with "-C /dev/null" in order to find the
6221 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
6222 to be output. However, since Exim outputs its version number before the
6223 error, it didn't break the script. It just looked ugly. I fixed this by
6224 always allowing "-C /dev/null" if the caller is root.
6226 14. Ignore overlarge ACL variable number when reading spool file - insurance
6227 against a later release with more variables having written the file.
6229 15. The standard form for an IPv6 address literal was being rejected by EHLO.
6230 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
6231 this, as well as the form without the "IPv6" on the front.
6233 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
6234 OS/Makefile-Darwin file.
6236 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
6237 only to LDAP libraries that do not have LDAP_OPT_DEREF.
6239 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
6240 variable. However, some OS use ints for inodes. I've added cast to long int
6241 to get rid of the compiler warning.
6243 19. I had forgotten to lock out "/../" in configuration file names when
6244 ALT_CONFIG_PREFIX was set.
6246 20. Routers used for verification do not need to specify transports. However,
6247 if such a router generated a host list, and callout was configured, Exim
6248 crashed, because it could not find a port number from the (non-existent)
6249 transport. It now assumes port 25 in this circumstance.
6251 21. Added the -t option to exigrep.
6253 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
6254 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
6255 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
6256 them is set, LOOKUP_LSEARCH is forced.
6258 23. "exim -bV" now outputs a list of lookups that are included in the binary.
6260 24. Added sender and host information to the "rejected by local_scan()" log
6261 line; previously there was no indication of these.
6263 25. Added .include_if_exists.
6265 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
6266 Linux. It turns out that not all file systems support this. Apparently some
6267 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
6268 do it.) To cope with this, the error EINVAL, which means that sync-ing is
6269 not supported on the file descriptor, is now ignored when Exim is trying to
6270 sync a directory. This applies only to Linux.
6272 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
6274 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
6275 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
6278 29. Added support for maildirsize files from supplied patch (modified a bit).
6280 30. The use of :fail: followed by an empty string could lead Exim to respond to
6281 sender verification failures with (e.g.):
6283 550 Verification failed for <xxx>
6284 550 Sender verify failed
6286 where the first response line was missing the '-' that indicates it is not
6287 the final line of the response.
6289 31. The loop for finding the name of the user that called Exim had a hardwired
6290 limit of 10; it now uses the value of finduser_retries, which is used for
6291 all other user lookups.
6293 32. Added $received_count variable, available in data and not_smtp ACLs, and at
6296 33. Exim was neglecting to zero errno before one call of strtol() when
6297 expanding a string and expecting an integer value. On some systems this
6298 resulted in spurious "integer overflow" errors. Also, it was casting the
6299 result into an int without checking.
6301 34. Testing for a connection timeout using "timeout_connect" in the retry rules
6302 did not work. The code looks as if it has *never* worked, though it appears
6303 to have been documented since at least release 1.62. I have made it work.
6305 35. The "timeout_DNS" error in retry rules, also documented since at least
6306 1.62, also never worked. As it isn't clear exactly what this means, and
6307 clearly it isn't a major issue, I have abolished the feature by treating it
6308 as "timeout", and writing a warning to the main and panic logs.
6310 36. The display of retry rules for -brt wasn't always showing the error code
6313 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
6314 timeout_connect_A, timeout_connect_MX.
6316 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
6317 to the empty sender.
6319 39. The daemon was not analysing the content of -oX till after it had closed
6320 stderr and disconnected from the controlling terminal. This meant that any
6321 syntax errors were only noted on the panic log, and the return code from
6322 the command was 0. By re-arranging the code a little, I've made the
6323 decoding happen first, so such errors now appear on stderr, and the return
6324 code is 1. However, the actual setting up of the sockets still happens in
6325 the disconnected process, so errors there are still only recorded on the
6328 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
6329 connections (as happens on some IP stacks) was logged at start up time as
6330 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
6331 it from "IPv6 and IPv4", which means that two separate sockets are being
6334 41. The debug output for gethostbyname2() or getipnodebyname() failures now
6335 says whether AF_INET or AF_INET6 was passed as an argument.
6337 42. Exiwhat output was messed up when time zones were included in log
6340 43. Exiwhat now gives more information about the daemon's listening ports,
6341 and whether -tls-on-connect was used.
6343 44. The "port" option of the smtp transport is now expanded.
6345 45. A "message" modifier in a "warn" statement in a non-message ACL was being
6346 silently ignored. Now an error message is written to the main and panic
6349 46. There's a new ACL modifier called "logwrite" which writes to a log file
6350 as soon as it is encountered.
6352 47. Added $local_user_uid and $local_user_gid at routing time.
6354 48. Exim crashed when trying to verify a sender address that was being
6357 49. Exim was recognizing only a space character after ".include". It now also
6358 recognizes a tab character.
6360 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
6361 extracting the relevant information from the specification. The man page no
6362 longer contains scrambled data for the -d option, and I've added a section
6363 at the front about calling Exim under different names.
6365 51. Added "extra_headers" argument to the "mail" command in filter files.
6367 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
6370 53. Installed eximstats 1.29.
6372 54. Added transport_filter_timeout as a generic transport option.
6374 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
6375 Cc: header lines. This was required by RFC 822, but it not required by RFC
6378 56. Exim used to add From:, Date:, and Message-Id: header lines to any
6379 incoming messages that did not have them. Now it does so only if the
6380 message originates locally, that is, if there is no associated remote host
6381 address. When Resent- header lines are present, this applies to the Resent-
6382 lines rather than the non-Resent- lines.
6384 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
6385 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
6387 58. Messages for configuration errors now include the name of the main
6388 configuration file - useful now that there may be more than one file in a
6389 list (.included file names were always shown).
6391 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
6392 for those rare installations that do not start the daemon as root or run it
6393 setuid root. I've cut out the call to initgroups() if the daemon is not
6396 60. The Exim user and group can now be bound into the binary as text strings
6397 that are looked up at the start of Exim's processing.
6399 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
6401 62. Added $mailstore_basename variable.
6403 63. Installed patch to sieve.c from Michael Haardt.
6405 64. When Exim failed to open the panic log after failing to open the main log,
6406 the original message it was trying to log was written to stderr and debug
6407 output, but if they were not available (the usual case in production), it
6408 was lost. Now it is written to syslog before the two lines that record the
6409 failures to open the logs.
6411 65. Users' Exim filters run in subprocesses under the user's uid. It is
6412 possible for a "deliver" command or an alias in a "personal" command to
6413 provoke an address rewrite. If logging of address rewriting is configured,
6414 this fails because the process is not running as root or exim. There may be
6415 a better way of dealing with this, but for the moment (because 4.30 needs
6416 to be released), I have disabled address rewrite logging when running a
6417 filter in a non-root, non-exim process.
6423 1. The buildconfig auxiliary program wasn't quoting the value set for
6424 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
6425 not defined. This bug was masked in 4.22 by the effect that was fixed in
6428 2. Some messages that were rejected after a message id was allocated were
6429 shown as "incomplete" by exigrep. It no longer does this for messages that
6430 are rejected by local_scan() or the DATA or non-SMTP ACLs.
6432 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
6433 have allow_domain_literals set, the ID did not get logged in the <= line.
6434 Domain literals are now always recognized in Message-ID: header lines.
6436 4. The first argument for a ${extract expansion item is the key name or field
6437 number. Leading and trailing spaces in this item were not being ignored,
6438 causing some misleading effects.
6440 5. When deliver_drop_privilege was set, single queue runner processes started
6441 manually (i.e. by the command "exim -q") or by the daemon (which uses the
6442 same command in the process it spins off) were not dropping privilege.
6444 6. When the daemon running as "exim" started a queue runner, it always
6445 re-executed Exim in the spun-off process. This is a waste of effort when
6446 deliver_drop_privilege is set. The new process now just calls the
6447 queue-runner function directly.
6453 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
6456 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
6457 ignored. Though the use of -oP was forcing the writing of a pid file, it
6458 was always written to the default place.
6460 3. If the message "no IP address found for host xxxx" is generated during
6461 incoming verification, it is now followed by identification of the incoming
6462 connection (so you can more easily find what provoked it).
6464 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
6466 5. Added some features to "harden" Exim a bit more against certain attacks:
6468 (a) There is now a build-time option called FIXED_NEVER_USERS that can
6469 be put in Local/Makefile. This is like the never_users runtime option,
6470 but it cannot be overridden. The default setting is "root".
6472 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
6473 prefix string with which any file named in a -C command line option
6476 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
6477 is retained for -C and -D only if the caller of Exim is root. Without
6478 it, the exim user may also use -C and -D and retain privilege.
6480 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
6481 command line option is disabled.
6483 6. Macro names set by the -D option must start with an upper case letter, just
6484 like macro names defined in the configuration file.
6486 7. Added "dereference=" facility to LDAP.
6488 8. Two instances of the typo "uknown" in the source files are fixed.
6490 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
6491 the Configure-Makefile script screwed up while processing it.
6493 10. Incorporated PCRE 4.4.
6495 11. The SMTP synchronization check was not operating right at the start of an
6496 SMTP session. For example, it could not catch a HELO sent before the client
6497 waited for the greeting. There is now a check for outstanding input at the
6498 point when the greeting is written. Because of the duplex, asynchronous
6499 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
6500 way, but not yet received, when the check is performed.
6502 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
6503 on TCP/IP sockets, because this apparently causes some broken clients to
6506 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
6507 unchanged) from the Cygwin maintainer.
6509 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
6510 was supported instead of testing for mbx. Effectively a typo.
6512 15. The spa authenticator server code was not checking that the input it
6513 received was valid base64.
6515 16. The debug output line for the "set" modifier in ACLs was not showing the
6516 name of the variable that was being set.
6518 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
6520 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
6521 lookup on its own. Something else was needed to trigger the lookup. For
6522 example, a match in host_lookup or the need for a host name in a host list.
6523 Now, if $sender_host_name is referenced and the host name has not yet been
6524 looked up, a lookup is performed. If the lookup fails, the variable remains
6525 empty, and $host_lookup_failed is set to "1".
6527 19. Added "eqi" as a case-independent comparison operator.
6529 20. The saslauthd authentication condition could segfault if neither service
6530 nor realm was specified.
6532 21. If an overflowing value such as "2048M" was set for message_size_limit, the
6533 error message that was logged was misleading, and incoming SMTP
6534 connections were dropped. The message is now more accurate, and temporary
6535 errors are given to SMTP connections.
6537 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
6538 (except RSET) with a 421 error, until QUIT is received. However, it was
6539 failing to send a response to QUIT.
6541 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
6542 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
6543 ACL is now run after the helo_try_verify_hosts code.
6545 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
6546 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
6547 case-independent, so other case variants are also recognized). Apparently
6548 some systems use these upper case variants.
6550 25. If more than two messages were waiting for the same host, and a transport
6551 filter was specified for the transport, Exim sent two messages over the
6552 same TCP/IP connection, and then failed with "socket operation on non-
6553 socket" when it tried to send the third.
6555 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
6557 27. The extern definition of crypt16() in expand.c was not being excluded when
6558 the OS had its own crypt16() function.
6560 28. Added bounce_return_body as a new option, and bounce_return_size_limit
6561 as a preferred synonym for return_size_limit, both as an option and as an
6564 29. Added LIBS=-liconv to OS/Makefile-OSF1.
6566 30. Changed the default configuration ACL to relax the local part checking rule
6567 for addresses that are not in any local domains. For these addresses,
6568 slashes and pipe symbols are allowed within local parts, but the sequence
6569 /../ is explicitly forbidden.
6571 31. SPA server authentication was not clearing the challenge buffer before
6574 32. log_message in a "warn" ACL statement was writing to the reject log as
6575 well as to the main log, which contradicts the documentation and doesn't
6576 seem right (because no rejection is happening). So I have stopped it.
6578 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
6579 However, I am unable to do any testing of this.
6581 34. Fixed an infelicity in the appendfile transport. When checking directories
6582 for a mailbox, to see if any needed to be created, it was accidentally
6583 using path names with one or more superfluous leading slashes; tracing
6584 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
6586 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
6587 contents are added to the log line that is written for every discarded
6588 recipient. (Previously a log_message setting was ignored.)
6590 36. The ${quote: operator now quotes the string if it is empty.
6592 37. The install script runs exim in order to find its version number. If for
6593 some reason other than non-existence or emptiness, which it checks, it
6594 could not run './exim', it was installing it with an empty version number,
6595 i.e. as "exim-". This error state is now caught, and the installation is
6598 38. An argument was missing from the function that creates an error message
6599 when Exim fails to connect to the socket for saslauthd authentication.
6600 This could cause Exim to crash, or give a corrupted message.
6602 39. Added isip, isip4, and isip6 to ${if conditions.
6604 40. The ACL variables $acl_xx are now saved with the message, and can be
6605 accessed later in routers, transports, and filters.
6607 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
6608 strings in the file are not string-expanded.
6610 42. If a MAIL command specified a SIZE value that was too large to fit into an
6611 int variable, the check against message_size_limit failed. Such values are
6612 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
6613 day this will have to be increased, but I don't think I want to be around
6614 when emails are that large.
6621 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
6622 iconv() is not standard in FreeBSD.
6624 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
6625 IPv6 enabled. The observed symptom was a segmentation fault on return from
6626 the function os_common_find_running_interfaces() in src/os.c.
6628 3. In the check_special_case() function in daemon.c I had used "errno" as an
6629 argument name, which causes warnings on some systems. This was basically a
6630 typo, since it was named "eno" in the comments!
6632 4. The code that waits for the clock to tick (at a resolution of some fraction
6633 of a second) so as to ensure message-id uniqueness was always waiting for
6634 at least one whole tick, when it could have waited for less. [This is
6635 almost certainly not relevant at current processor speeds, where it is
6636 unlikely to ever wait at all. But we try to future-proof.]
6638 5. The function that sleeps for a time interval that includes fractions of a
6639 second contained a race. It did not block SIGALRM between setting the
6640 timer, and suspending (a couple of lines later). If the interval was short
6641 and the sigsuspend() was delayed until after it had expired, the suspension
6642 never ended. On busy systems this could lead to processes getting stuck for
6645 6. Some uncommon configurations may cause a lookup to happen in a queue runner
6646 process, before it forks any delivery processes. The open lookup caching
6647 mechanism meant that the open file or database connection was passed into
6648 the delivery process. The problem was that delivery processes always tidy
6649 up cached lookup data. This could cause a problem for the next delivery
6650 process started by the queue runner, because the external queue runner
6651 process does not know about the closure. So the next delivery process
6652 still has data in the lookup cache. In the case of a file lookup, there was
6653 no problem because closing a file descriptor in a subprocess doesn't affect
6654 the parent. However, if the lookup was caching a connection to a database,
6655 the connection was closed, and the second delivery process was likely to
6656 see errors such as "PGSQL: query failed: server closed the connection
6657 unexpectedly". The problem has been fixed by closing all cached lookups
6658 in a queue runner before running a delivery process.
6660 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
6661 seem to have the "const" qualifier which it has on other OS. I've
6664 8. Change 4.21/2 was too strict. It is only if there are two authenticators
6665 *of the same type* (client or server) with the same public name that an
6666 error should be diagnosed.
6668 9. When Exim looked up a host name for an IP address, but failed to find the
6669 original IP address when looking up the host name (a safety check), it
6670 output the message "<ip address> does not match any IP for NULL", which was
6671 confusing, to say the least. The bug was that the host name should have
6672 appeared instead of "NULL".
6674 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
6675 user that is built into the binary, and the -C or -D options is used, root
6676 privilege is dropped before the configuration file is read. In addition,
6677 logging is switched to stderr instead of the normal log files. If the
6678 configuration then re-defines the Exim user, the unprivileged environment
6679 is probably not what is expected, so Exim logs a panic warning message (but
6682 However, if deliver_drop_privilege is set, the unprivileged state may well
6683 be exactly what is intended, so the warning has been cut out in that case,
6684 and Exim is allowed to try to write to its normal log files.
6690 1. smtp_return_error_details was not giving details for temporary sender
6691 or receiver verification errors.
6693 2. Diagnose a configuration error if two authenticators have the same public
6696 3. Exim used not to create the message log file for a message until the first
6697 delivery attempt. This could be confusing when incoming messages were held
6698 for policy or load reasons. The message log file is now created at the time
6699 the message is received, and an initial "Received" line is written to it.
6701 4. The automatically generated man page for command line options had a minor
6702 bug that caused no ill effects; however, a more serious problem was that
6703 the procedure for building the man page automatically didn't always
6704 operate. Consequently, release 4.20 contains an out-of-date version. This
6705 shouldn't happen again.
6707 5. When building Exim with embedded Perl support, the script that builds the
6708 Makefile was calling 'perl' to find its compile-time parameters, ignoring
6709 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
6711 6. The freeze_tell option was not being used for messages that were frozen on
6712 arrival, either by an ACL or by local_scan().
6714 7. Added the smtp_incomplete_transaction log selector.
6716 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
6717 was accepting AUTH without a new EHLO.
6719 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
6720 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
6723 10. Logging of TCP/IP connections (when configured) now happens in the main
6724 daemon process instead of the child process, so that the TCP/IP connection
6725 count is more accurate (but it can never be perfect).
6727 11. The use of "drop" in a nested ACL was not being handled correctly in the
6728 outer ACL. Now, if condition failure induced by the nested "drop" causes
6729 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
6730 or "require"), the connection is dropped.
6732 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
6733 that yield "discard" can now be used with an "accept" or a "discard" verb,
6734 but an error is generated for any others (because I can't see a useful way
6735 to define what should happen).
6737 13. When an ACL is read dynamically from a file (or anywhere else), the lines
6738 are now processed in the same way as lines in the Exim configuration file.
6739 In particular, continuation lines are supported.
6741 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
6743 15. Added -ti meaning -t -i.
6745 16. Check for letters, digits, hyphens, and dots in the names of dnslist
6746 domains, and warn by logging if others are found.
6748 17. At least on BSD, alignment is not guaranteed for the array of ifreq's
6749 returned from GIFCONF when Exim is trying to find the list of interfaces on
6750 a host. The code in os.c has been modified to copy each ifreq to an aligned
6751 structure in all cases.
6753 Also, in some cases, the returned ifreq's were being copied to a 'struct
6754 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
6755 means the last couple of bytes of an IPv6 address could be chopped if the
6756 ifreq contained only a normal sockaddr (14 bytes storage).
6758 18. Named domain lists were not supported in the hosts_treat_as_local option.
6759 An entry such as +xxxx was not recognized, and was treated as a literal
6762 19. Ensure that header lines added by a DATA ACL are included in the reject log
6763 if the ACL subsequently rejects the message.
6765 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
6766 MD5 (which is deprecated).
6768 21. When testing a filter file using -bf, Exim was writing a message when it
6769 took the sender from a "From " line in the message, but it was not doing so
6770 when it took $return_path from a Return-Path: header line. It now does.
6772 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
6773 with a valid header line field name (a series of printing characters
6774 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
6776 23. Changed "disc" in the source to "disk" to conform to the documentation and
6777 the book and for uniformity.
6779 24. Ignore Sendmail's -Ooption=value command line item.
6781 25. When execve() failed while trying to run a command in a pipe transport,
6782 Exim was returning EX_UNAVAILABLE (69) from the subprocess. However, this
6783 could be confused with a return value of 69 from the command itself. This
6784 has been changed to 127, the value the shell returns if it is asked to run
6785 a non-existent command. The wording for the related log line suggests a
6786 non-existent command as the problem.
6788 26. If received_header_text expands to an empty string, do not add a Received:
6789 header line to the message. (Well, it adds a token one on the spool, but
6790 marks it "old" so that it doesn't get used or transmitted.)
6792 27. Installed eximstats 1.28 (addition of -nt option).
6794 28. There was no check for failure on the call to getsockname() in the daemon
6795 code. This can fail if there is a shortage of resources on the system, with
6796 ENOMEM, for example. A temporary error is now given on failure.
6798 29. Contrary to the C standard, it seems that in some environments, the
6799 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
6800 program. Exim now does this explicitly; it affects the formatting of
6801 timestamps using strftime().
6803 30. If exiqsumm was given junk data, it threw up some uninitialized variable
6804 complaints. I've now initialized all the variables, to avoid this.
6806 32. Header lines added by a system filter were not being "seen" during
6807 transport-time rewrites.
6809 33. The info_callback() function passed to OpenSSL is set up with type void
6810 (*)(SSL *, int, int), as described somewhere. However, when calling the
6811 function (actually a macro) that sets it up, the type void(*)() is
6812 expected. I've put in a cast to prevent warnings from picky compilers.
6814 34. If a DNS black list lookup found a CNAME record, but there were no A
6815 records associated with the domain it pointed at, Exim crashed.
6817 35. If a DNS black list lookup returned more than one A record, Exim ignored
6818 all but the first. It now scans all returned addresses if a particular IP
6819 value is being sought. In this situation, the contents of the
6820 $dnslist_value variable are a list of all the addresses, separated by a
6823 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
6824 to accept a host name and all its aliases if the forward lookup for any of
6825 them yielded the IP address of the incoming connection. Now it accepts only
6826 those names whose forward lookup yields the correct IP address. Any other
6827 names are discarded. This closes a loophole whereby a rogue DNS
6828 administrator could create reverse DNS records to break through a
6829 wildcarded host restriction in an ACL.
6831 37. If a user filter or a system filter that ran in a subprocess used any of
6832 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
6833 the wrong values were passed to the pipe command ($thisaddress had the
6834 value of $0, $0 had the value of $1, etc). This bug was introduced by
6835 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
6837 38. Improved the line breaking for long SMTP error messages from ACLs.
6838 Previously, if there was no break point between 40 and 75 characters, Exim
6839 left the rest of the message alone. Two changes have been made: (a) I've
6840 reduced the minimum length to 35 characters; (b) if it can't find a break
6841 point between 35 and 75 characters, it looks ahead and uses the first one
6842 that it finds. This may give the occasional overlong line, but at least the
6843 remaining text gets split now.
6845 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
6846 file descriptors might be low, and that setting 1000 would always raise it.
6847 It turns out that in some environments, the limit is already over 1000 and
6848 that lowering it causes trouble. So now Exim takes care not to decrease it.
6850 40. When delivering a message, the value of $return_path is set to $sender_
6851 address at the start of routing (routers may change the value). By an
6852 oversight, this default was not being set up when an address was tested by
6853 -bt or -bv, which affected the outcome if any router or filter referred to
6856 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
6857 log only when "message" or "log_message" are set. However, if one of the
6858 conditions was an address verification, or a call to a nested ACL, the
6859 messages generated by the underlying test were being passed through. This
6860 no longer happens. The underlying message is available in $acl_verify_
6861 message for both "message" and "log_message" expansions, so it can be
6862 passed through if needed.
6864 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
6865 new expansion $bh_ to give the encoded byte string without charset
6866 translation. Translation happens only if iconv() is available; HAVE_ICONV
6867 indicates this at build time. HEADERS_CHARSET gives the charset to
6868 translate to; headers_charset can change it in the configuration, and
6869 "headers charset" can change it in an individual filter file.
6871 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
6872 that creates RFC 2047 encoded "words" labels them as that charset instead
6873 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
6874 expansion operator; (ii) when Exim creates a From: line for a local
6875 message; (iii) when a header line is rewritten to include a "phrase" part.
6877 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
6878 buggy, causing it to skip the first lines of messages whose message ID
6879 ended in 'D'. This would not have bitten before Exim release 4.14, because
6880 message IDs were unlikely to end in 'D' before then. The effect was to have
6881 incorrect size information for certain domains.
6883 45. #include "config.h" was missing at the start of the crypt16.c module. This
6884 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
6887 46. If there was a timeout during a "random" callout check, Exim treated it as
6888 a failure of the random address, and carried on sending RSET and the real
6889 address. If the delay was just some slowness somewhere, the response to the
6890 original RCPT would be taken as a response to RSET and so on, causing
6891 mayhem of various kinds.
6893 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
6894 when I implemented it. It didn't allow for the fact that some option values
6895 may legitimately be negative (e.g. size_addition), and it didn't even do
6896 the right test for positive values.
6898 48. Domain names in DNS records are case-independent. Exim always looks them up
6899 in lower case. Some resolvers return domain names in exactly the case they
6900 appear in the zone file, that is, they may contain uppercase letters. Not
6901 all resolvers do this - some return always lower case. Exim was treating a
6902 change of case by a resolver as a change of domain, similar to a widening
6903 of a domain abbreviation. This triggered its re-routing code and so it was
6904 trying to route what was effectively the same domain again. This normally
6905 caused routing to fail (because the router wouldn't handle the domain
6906 twice). Now Exim checks for this case specially, and just changes the
6907 casing of the domain that it ultimately uses when it transmits the message
6910 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
6913 50. If a filter generated a file delivery with a non-absolute name (possible if
6914 no home directory exists for the router), the forbid_file option was not
6917 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
6918 the existing equality matching.
6920 52. Exim was using ints instead of ino_t variables in some places where it was
6921 dealing with inode numbers.
6923 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
6924 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
6925 called TMPDIR, and if it finds it is different, it changes its value.
6927 54. The smtp_printf() function is now made available to local_scan() so
6928 additional output lines can be written before returning. There is also an
6929 smtp_fflush() function to enable the detection of a dropped connection.
6930 The variables smtp_input and smtp_batched_input are exported to
6933 55. Changed the default runtime configuration: the message "Unknown user"
6934 has been removed from the ACL, and instead placed on the localuser router,
6935 using the cannot_route_message feature. This means that any verification
6936 failures that generate their own messages won't get overridden. Similarly,
6937 the "Unrouteable address" message that was in the ACL for unverifiable
6938 relay addresses has also been removed.
6940 56. Added hosts_avoid_esmtp to the smtp transport.
6942 57. The exicyclog script was not checking for the esoteric option
6943 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
6944 will work only if exicyclog is run under the appropriate euid.
6946 58. Following a discussion on the list, the rules by which Exim recognises line
6947 endings on incoming messages have been changed. The -dropcr and drop_cr
6948 options are now no-ops, retained only for backwards compatibility. The
6949 following line terminators are recognized: LF CRLF CR. However, special
6950 processing applies to CR:
6952 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
6953 nor a local message in the state where . is a terminator.
6955 (ii) If a bare CR is encountered in a header line, an extra space is added
6956 after the line terminator so as not to end the header. The reasoning
6957 behind this is that bare CRs in header lines are most likely either
6958 to be mistakes, or people trying to play silly games.
6960 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
6961 was being incorrectly given as 18 bytes larger than it should have been.
6962 This is a VOB (very old bug).
6964 60. This may never have affected anything current, but just in case it has:
6965 When the local host is found other than at the start of a list of hosts,
6966 the local host, those with the same MX, and any that follow, are discarded.
6967 When the list in question was part of a longer list of hosts, the following
6968 hosts (not currently being processed) were also being discarded. This no
6969 longer happens. I'm not sure if this situation could ever has previously
6972 61. Added the "/MX" feature to lists of hosts in the manualroute and query
6975 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
6976 header. This is something that is recommended in a new Internet Draft, and
6977 is something that is documented as being done by Sendmail. There are two
6978 possible values. For messages generated by the autoreply transport, Exim
6981 Auto-Submitted: auto-replied
6983 whereas for all other generated messages (e.g. bounces) it adds
6985 Auto-Submitted: auto-generated
6987 63. The "personal" condition in filters now includes a test for the
6988 Auto-Submitted: header. If it contains the string "auto-" the message it
6989 not considered personal.
6991 64. Added rcpt_include_affixes as a generic transport option.
6993 65. Added queue_only_override (default true).
6995 66. Added the syslog_duplication option.
6997 67. If what should have been the first header line of a message consisted of
6998 a space followed by a colon, Exim was mis-interpreting it as a header line.
6999 It isn't of course - it is syntactically invalid and should therefore be
7000 treated as the start of the message body. The misbehaviour could have
7001 caused a number of strange effects, including loss of data in subsequent
7002 header lines, and spool format errors.
7004 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
7005 client host had authenticated. This control can now be exercised by an ACL
7006 for more flexibility.
7008 69. By default, callouts do not happen when testing with -bh. There is now a
7009 variant, -bhc, which does actually run the callout code, including
7010 consulting and updating the callout cache.
7012 70. Added support for saslauthd authentication, courtesy of Alexander
7015 71. If statvfs() failed on the spool or log directories while checking their
7016 size for availability, Exim confusingly gave the error "space shortage".
7017 Furthermore, in debugging mode it crashed with a floating point exception.
7018 These checks are done if check_{spool,log}_{space,inodes} are set, and when
7019 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
7020 serious problem, Exim now writes to the main and panic logs when this
7021 happens, with details of the failure. It then refuses to accept the
7022 incoming message, giving the message "spool directory problem" or "log
7023 directory problem" with a 421 code for SMTP messages.
7025 72. When Exim is about to re-exec itself, it ensures that the file descriptors
7026 0, 1, and 2 exist, because some OS complain for execs without them (see
7027 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
7028 descriptors. However, the code omitted to check that the open succeeded,
7029 causing mysterious errors if for some reason the permissions on /dev/null
7030 got screwed. Now Exim writes a message to the main and panic logs, and
7031 bombs out if it can't open /dev/null.
7033 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
7034 interact so that it is all more flexible. It is supposed to remain
7035 backwards compatible. Also added extra_local_interfaces.
7037 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
7038 to bomb out with an assertion failure - to the client this appears as a
7039 connection drop. This problem occurs in the part of the code that was taken
7040 from the Samba project. Fortunately, the assertion is in a very simple
7041 function, so I have fixed this by reproducing the function inline in the
7042 one place where it is called, and arranging for authentication to fail
7043 instead of killing the process with assert().
7045 75. The SPA client code was not working when the server requested OEM rather
7046 than Unicode encoding.
7048 76. Added code to make require_files with a specific uid setting more usable in
7049 the case where statting the file as root fails - usually a non-root-mounted
7050 NFS file system. When this happens and the failure is EACCES, Exim now
7051 forks a subprocess and does the per-uid checking as the relevant uid.
7053 77. Added process_log_path.
7055 78. If log_file_path was not explicitly set, a setting of check_log_space or
7056 check_log_inodes was ignored.
7058 79. If a space check for the spool or log partitions fails, the incident is now
7059 logged. Of course, in the latter case the data may get lost...
7061 80. Added the %p formatting code to string_format() so that it can be used to
7062 print addresses in debug_print(). Adjusted all the address printing in the
7063 debugging in store.c to use %p rather than %d.
7065 81. There was a concern that a line of code in smtp_in.c could overflow a
7066 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
7067 initially expressed, the concern was not well-founded, because trailing
7068 spaces are removed early. However, if the trailing spaces were followed by
7069 a NULL, they did not get removed, so the overflow was possible. Two fixes
7072 (a) I re-wrote the offending code in a cleaner fashion.
7073 (b) If an incoming SMTP command contains a NULL character, it is rejected
7076 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
7077 runs initgroups(), so that if the Exim user is in any additional groups,
7078 they will be used during message reception.
7084 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / exim.git / blob