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.
127 JH/01 Rework error string handling in TLS interface so that the caller in
128 more cases is responsible for logging. This permits library-sourced
129 string to be attached to addresses during delivery, and collapses
130 pairs of long lines into single ones.
132 PP/01 Allow PKG_CONFIG_PATH to be set in Local/Makefile and use it correctly
133 during configuration. Wildcards are allowed and expanded.
135 JH/02 Rework error string handling in DKIM to pass more info back to callers.
136 This permits better logging.
138 JH/03 Rework the transport continued-connection mechanism: when TLS is active,
139 do not close it down and have the child transport start it up again on
140 the passed-on TCP connection. Instead, proxy the child (and any
141 subsequent ones) for TLS via a unix-domain socket channel. Logging is
142 affected: the continued delivery log lines do not have any DNSSEC, TLS
143 Certificate or OCSP information. TLS cipher information is still logged.
145 JH/04 Shorten the log line for daemon startup by collapsing adjacent sets of
146 identical IP addresses on different listening ports. Will also affect
149 PP/02 Bug 2070: uClibc defines __GLIBC__ without providing glibc headers;
150 add noisy ifdef guards to special-case this sillyness.
151 Patch from Bernd Kuhls.
153 JH/05 Tighten up the checking in isip4 (et al): dotted-quad components larger
154 than 255 are no longer allowed.
156 JH/06 Default openssl_options to include +no_ticket, to reduce load on peers.
157 Disable the session-cache too, which might reduce our load. Since we
158 currrectly use a new context for every connection, both as server and
159 client, there is no benefit for these.
160 GnuTLS appears to not support tickets server-side by default (we don't
161 call gnutls_session_ticket_enable_server()) but client side is enabled
162 by default on recent versions (3.1.3 +) unless the PFS priority string
165 PP/03 Add $SOURCE_DATE_EPOCH support for reproducible builds, per spec at
166 <https://reproducible-builds.org/specs/source-date-epoch/>.
168 JH/07 Fix smtp transport use of limited max_rcpt under mua_wrapper. Previously
169 the check for any unsuccessful recipients did not notice the limit, and
170 erroneously found still-pending ones.
172 JH/08 Pipeline CHUNKING command and data together, on kernels that support
173 MSG_MORE. Only in-clear (not on TLS connections).
175 JH/09 Avoid using a temporary file during transport using dkim. Unless a
176 transport-filter is involved we can buffer the headers in memory for
177 creating the signature, and read the spool data file once for the
178 signature and again for transmission.
180 JH/10 Enable use of sendfile in Linux builds as default. It was disabled in
181 4.77 as the kernel support then wasn't solid, having issues in 64bit
182 mode. Now, it's been long enough. Add support for FreeBSD also.
184 JH/11 Bug 2104: Fix continued use of a transport connection with TLS. In the
185 case where the routing stage had gathered several addresses to send to
186 a host before calling the transport for the first, we previously failed
187 to close down TLS in the old transport process before passing the TCP
188 connection to the new process. The new one sent a STARTTLS command
189 which naturally failed, giving a failed delivery and bloating the retry
190 database. Investigation and fix prototype from Wolfgang Breyha.
192 JH/12 Fix check on SMTP command input synchronisation. Previously there were
193 false-negatives in the check that the sender had not preempted a response
194 or prompt from Exim (running as a server), due to that code's lack of
195 awareness of the SMTP input buffering.
197 PP/04 Add commandline_checks_require_admin option.
198 Exim drops privileges sanely, various checks such as -be aren't a
199 security problem, as long as you trust local users with access to their
200 own account. When invoked by services which pass untrusted data to
201 Exim, this might be an issue. Set this option in main configuration
202 AND make fixes to the calling application, such as using `--` to stop
205 JH/13 Do pipelining under TLS. Previously, although safe, no advantage was
206 taken. Now take care to pack both (client) MAIL,RCPT,DATA, and (server)
207 responses to those, into a single TLS record each way (this usually means
208 a single packet). As a side issue, smtp_enforce_sync now works on TLS
211 PP/05 OpenSSL/1.1: use DH_bits() for more accurate DH param sizes. This
212 affects you only if you're dancing at the edge of the param size limits.
213 If you are, and this message makes sense to you, then: raise the
214 configured limit or use OpenSSL 1.1. Nothing we can do for older
217 JH/14 For the "sock" variant of the malware scanner interface, accept an empty
218 cmdline element to get the documented default one. Previously it was
221 JH/15 Fix a crash in the smtp transport caused when two hosts in succession
222 are unsuable for non-message-specific reasons - eg. connection timeout,
223 banner-time rejection.
225 JH/16 Fix logging of delivery remote port, when specified by router, under
228 PP/06 Repair manualroute's ability to take options in any order, even if one
229 is the name of a transport.
232 HS/01 Cleanup, prevent repeated use of -p/-oMr (CVE-2017-1000369)
234 JH/17 Change the list-building routines interface to use the expanding-string
235 triplet model, for better allocation and copying behaviour.
237 JH/18 Prebuild the data-structure for "builtin" macros, for faster startup.
238 Previously it was constructed the first time a possibly-matching string
239 was met in the configuration file input during startup; now it is done
242 JH/19 Bug 2141: Use the full-complex API for Berkeley DB rather than the legacy-
243 compatible one, to avoid the (poorly documented) possibility of a config
244 file in the working directory redirecting the DB files, possibly correpting
245 some existing file. CVE-2017-10140 assigned for BDB.
247 JH/20 Bug 2147: Do not defer for a verify-with-callout-and-random which is not
248 cache-hot. Previously, although the result was properly cached, the
249 initial verify call returned a defer.
251 JH/21 Bug 2151: Avoid using SIZE on the MAIL for a callout verify, on any but
252 the main verify for receipient in uncached-mode.
254 JH/22 Retire historical build files to an "unsupported" subdir. These are
255 defined as "ones for which we have no current evidence of testing".
257 JH/23 DKIM: enforce the DNS pubkey record "h" permitted-hashes optional field,
258 if present. Previously it was ignored.
260 JH/24 Start using specified-initialisers in C structure init coding. This is
261 a C99 feature (it's 2017, so now considered safe).
263 JH/25 Use one-bit bitfields for flags in the "addr" data structure. Previously
264 if was a fixed-sized field and bitmask ops via macros; it is now more
267 PP/07 GitHub PR 56: Apply MariaDB build fix.
268 Patch provided by Jaroslav Škarvada.
270 PP/08 Bug 2161: Fix regression in sieve quoted-printable handling introduced
271 during Coverity cleanups [4.87 JH/47]
272 Diagnosis and fix provided by Michael Fischer v. Mollard.
274 JH/26 Fix DKIM bug: when the pseudoheader generated for signing was exactly
275 the right size to place the terminating semicolon on its own folded
276 line, the header hash was calculated to an incorrect value thanks to
277 the (relaxed) space the fold became.
279 HS/02 Fix Bug 2130: large writes from the transport subprocess where chunked
280 and confused the parent.
282 JH/27 Fix SOCKS bug: an unitialized pointer was deref'd by the transport process
283 which could crash as a result. This could lead to undeliverable messages.
285 JH/28 Logging: "next input sent too soon" now shows where input was truncated
288 JH/29 Fix queue_run_in_order to ignore the PID portion of the message ID. This
289 matters on fast-turnover and PID-randomising systems, which were getting
290 out-of-order delivery.
292 JH/30 Fix a logging bug on aarch64: an unsafe routine was previously used for
293 a possibly-overlapping copy. The symptom was that "Remote host closed
294 connection in response to HELO" was logged instead of the actual 4xx
297 JH/31 Fix CHUNKING code to properly flush the unwanted chunk after an error.
298 Previously only that bufferd was discarded, resulting in SYMTP command
301 JH/32 DKIM: when a message has multiple signatures matching an identity given
302 in dkim_verify_signers, run the dkim acl once for each. Previously only
303 one run was done. Bug 2189.
305 JH/33 Downgrade an unfound-list name (usually a typo in the config file) from
306 "panic the current process" to "deliberately defer". The panic log is
307 still written with the problem list name; the mail and reject logs now
308 get a temp-reject line for the message that was being handled, saying
309 something like "domains check lookup or other defer". The SMTP 451
310 message is still "Temporary local problem".
312 JH/34 Bug 2199: Fix a use-after-free while reading smtp input for header lines.
313 A crafted sequence of BDAT commands could result in in-use memory beeing
314 freed. CVE-2017-16943.
316 HS/03 Bug 2201: Fix checking for leading-dot on a line during headers reading
317 from SMTP input. Previously it was always done; now only done for DATA
318 and not BDAT commands. CVE-2017-16944.
320 JH/35 Bug 2201: Flush received data in BDAT mode after detecting an error fatal
321 to the message (such as an overlong header line). Previously this was
322 not done and we did not exit BDAT mode. Followon from the previous item
323 though a different problem.
329 JH/01 Bug 1922: Support IDNA2008. This has slightly different conversion rules
330 than -2003 did; needs libidn2 in addition to libidn.
332 JH/02 The path option on a pipe transport is now expanded before use.
334 PP/01 GitHub PR 50: Do not call ldap_start_tls_s on ldapi:// connections.
335 Patch provided by "Björn", documentation fix added too.
337 JH/03 Bug 2003: fix Proxy Protocol v2 handling: the address size field was
338 missing a wire-to-host endian conversion.
340 JH/04 Bug 2004: fix CHUNKING in non-PIPELINEING mode. Chunk data following
341 close after a BDAT command line could be taken as a following command,
342 giving a synch failure. Fix by only checking for synch immediately
343 before acknowledging the chunk.
345 PP/02 GitHub PR 52: many spelling fixes, which include fixing parsing of
346 no_require_dnssec option and creation of _HAVE_TRANSPORT_APPEND_MAILDIR
347 macro. Patches provided by Josh Soref.
349 JH/05 Have the EHLO response advertise VRFY, if there is a vrfy ACL defined.
350 Previously we did not; the RFC seems ambiguous and VRFY is not listed
351 by IANA as a service extension. However, John Klensin suggests that we
354 JH/06 Bug 2017: Fix DKIM verification in -bh test mode. The data feed into
355 the dkim code may be unix-mode line endings rather than smtp wire-format
356 CRLF, so prepend a CR to any bare LF.
358 JH/07 Rationalise the coding for callout smtp conversations and transport ones.
359 As a side-benfit, callouts can now use PIPELINING hence fewer round-trips.
361 JH/08 Bug 2016: Fix DKIM verification vs. CHUNKING. Any BDAT commands after
362 the first were themselves being wrongly included in the feed into dkim
363 processing; with most chunk sizes in use this resulted in an incorrect
364 body hash calculated value.
366 JH/09 Bug 2014: permit inclusion of a DKIM-Signature header in a received
367 DKIM signature block, for verification. Although advised against by
368 standards it is specifically not ruled illegal.
370 JH/10 Bug 2025: Fix reception of (quoted) local-parts with embedded spaces.
372 JH/11 Bug 2029: Fix crash in DKIM verification when a message signature block is
373 missing a body hash (the bh= tag).
375 JH/12 Bug 2018: Re-order Proxy Protocol startup versus TLS-on-connect startup.
376 It seems that HAProxy sends the Proxy Protocol information in clear and
377 only then does a TLS startup, so do the same.
379 JH/13 Bug 2027: Avoid attempting to use TCP Fast Open for non-transport client
380 TCP connections (such as for Spamd) unless the daemon successfully set
381 Fast Open mode on its listening sockets. This fixes breakage seen on
382 too-old kernels or those not configured for Fast Open, at the cost of
383 requiring both directions being enabled for TFO, and TFO never being used
384 by non-daemon-related Exim processes.
386 JH/14 Bug 2000: Reject messages recieved with CHUNKING but with malformed line
387 endings, at least on the first header line. Try to canonify any that get
388 past that check, despite the cost.
390 JH/15 Angle-bracket nesting (an error inserted by broken sendmails) levels are
391 now limited to an arbitrary five deep, while parsing addresses with the
392 strip_excess_angle_brackets option enabled.
394 PP/03 Bug 2018: For Proxy Protocol and TLS-on-connect, do not over-read and
395 instead leave the unprompted TLS handshake in socket buffer for the
396 TLS library to consume.
398 PP/04 Bug 2018: Also handle Proxy Protocol v2 safely.
400 PP/05 FreeBSD compat: handle that Ports no longer create /usr/bin/perl
402 JH/16 Drop variables when they go out of scope. Memory management drops a whole
403 region in one operation, for speed, and this leaves assigned pointers
404 dangling. Add checks run only under the testsuite which checks all
405 variables at a store-reset and panics on a dangling pointer; add code
406 explicitly nulling out all the variables discovered. Fixes one known
407 bug: a transport crash, where a dangling pointer for $sending_ip_address
408 originally assigned in a verify callout, is re-used.
410 PP/06 Drop '.' from @INC in various Perl scripts.
412 PP/07 Switch FreeBSD iconv to always use the base-system libc functions.
414 PP/08 Reduce a number of compilation warnings under clang; building with
415 CC=clang CFLAGS+=-Wno-dangling-else -Wno-logical-op-parentheses
416 should be warning-free.
418 JH/17 Fix inbound CHUNKING when DKIM disabled at runtime.
420 HS/01 Fix portability problems introduced by PP/08 for platforms where
421 realloc(NULL) is not equivalent to malloc() [SunOS et al].
423 HS/02 Bug 1974: Fix missing line terminator on the last received BDAT
424 chunk. This allows us to accept broken chunked messages. We need a more
425 general solution here.
427 PP/09 Wrote util/chunking_fixqueue_finalnewlines.pl to help recover
428 already-broken messages in the queue.
430 JH/18 Bug 2061: Fix ${extract } corrupting an enclosing ${reduce } $value.
432 JH/19 Fix reference counting bug in routing-generated-address tracking.
438 JH/01 Use SIZE on MAIL FROM in a cutthrough connection, if the destination
439 supports it and a size is available (ie. the sending peer gave us one).
441 JH/02 The obsolete acl condition "demime" is removed (finally, after ten
442 years of being deprecated). The replacements are the ACLs
443 acl_smtp_mime and acl_not_smtp_mime.
445 JH/03 Upgrade security requirements imposed for hosts_try_dane: previously
446 a downgraded non-dane trust-anchor for the TLS connection (CA-style)
447 or even an in-clear connection were permitted. Now, if the host lookup
448 was dnssec and dane was requested then the host is only used if the
449 TLSA lookup succeeds and is dnssec. Further hosts (eg. lower priority
450 MXs) will be tried (for hosts_try_dane though not for hosts_require_dane)
451 if one fails this test.
452 This means that a poorly-configured remote DNS will make it incommunicado;
453 but it protects against a DNS-interception attack on it.
455 JH/04 Bug 1810: make continued-use of an open smtp transport connection
456 non-noisy when a race steals the message being considered.
458 JH/05 If main configuration option tls_certificate is unset, generate a
459 self-signed certificate for inbound TLS connections.
461 JH/06 Bug 165: hide more cases of password exposure - this time in expansions
462 in rewrites and routers.
464 JH/07 Retire gnutls_require_mac et.al. These were nonfunctional since 4.80
465 and logged a warning sing 4.83; now they are a configuration file error.
467 JH/08 Bug 1836: Fix crash in VRFY handling when handed an unqualified name
468 (lacking @domain). Apply the same qualification processing as RCPT.
470 JH/09 Bug 1804: Avoid writing msglog files when in -bh or -bhc mode.
472 JH/10 Support ${sha256:} applied to a string (as well as the previous
475 JH/11 Cutthrough: avoid using the callout hints db on a verify callout when
476 a cutthrough deliver is pending, as we always want to make a connection.
477 This also avoids re-routing the message when later placing the cutthrough
478 connection after a verify cache hit.
479 Do not update it with the verify result either.
481 JH/12 Cutthrough: disable when verify option success_on_redirect is used, and
482 when routing results in more than one destination address.
484 JH/13 Cutthrough: expand transport dkim_domain option when testing for dkim
485 signing (which inhibits the cutthrough capability). Previously only
486 the presence of an option was tested; now an expansion evaluating as
487 empty is permissible (obviously it should depend only on data available
488 when the cutthrough connection is made).
490 JH/14 Fix logging of errors under PIPELINING. Previously the log line giving
491 the relevant preceding SMTP command did not note the pipelining mode.
493 JH/15 Fix counting of empty lines in $body_linecount and $message_linecount.
494 Previously they were not counted.
496 JH/16 DANE: treat a TLSA lookup response having all non-TLSA RRs, the same
497 as one having no matching records. Previously we deferred the message
498 that needed the lookup.
500 JH/17 Fakereject: previously logged as a normal message arrival "<="; now
501 distinguished as "(=".
503 JH/18 Bug 1867: make the fail_defer_domains option on a dnslookup router work
504 for missing MX records. Previously it only worked for missing A records.
506 JH/19 Bug 1850: support Radius libraries that return REJECT_RC.
508 JH/20 Bug 1872: Ensure that acl_smtp_notquit is run when the connection drops
509 after the data-go-ahead and data-ack. Patch from Jason Betts.
511 JH/21 Bug 1846: Send DMARC forensic reports for reject and quarantine results,
512 even for a "none" policy. Patch from Tony Meyer.
514 JH/22 Fix continued use of a connection for further deliveries. If a port was
515 specified by a router, it must also match for the delivery to be
518 JH/23 Bug 1874: fix continued use of a connection for further deliveries.
519 When one of the recipients of a message was unsuitable for the connection
520 (has no matching addresses), we lost track of needing to mark it
521 deferred. As a result mail would be lost.
523 JH/24 Bug 1832: Log EHLO response on getting conn-close response for HELO.
525 JH/25 Decoding ACL controls is now done using a binary search; the source code
526 takes up less space and should be simpler to maintain. Merge the ACL
527 condition decode tables also, with similar effect.
529 JH/26 Fix problem with one_time used on a redirect router which returned the
530 parent address unchanged. A retry would see the parent address marked as
531 delivered, so not attempt the (identical) child. As a result mail would
534 JH/27 Fix a possible security hole, wherein a process operating with the Exim
535 UID can gain a root shell. Credit to http://www.halfdog.net/ for
536 discovery and writeup. Ubuntu bug 1580454; no bug raised against Exim
539 JH/28 Enable {spool,log} filesystem space and inode checks as default.
540 Main config options check_{log,spool}_{inodes,space} are now
541 100 inodes, 10MB unless set otherwise in the configuration.
543 JH/29 Fix the connection_reject log selector to apply to the connect ACL.
544 Previously it only applied to the main-section connection policy
547 JH/30 Bug 1897: fix callouts connection fallback from TLS to cleartext.
549 PP/01 Changed default Diffie-Hellman parameters to be Exim-specific, created
550 by me. Added RFC7919 DH primes as an alternative.
552 PP/02 Unbreak build via pkg-config with new hash support when crypto headers
553 are not in the system include path.
555 JH/31 Fix longstanding bug with aborted TLS server connection handling. Under
556 GnuTLS, when a session startup failed (eg because the client disconnected)
557 Exim did stdio operations after fclose. This was exposed by a recent
558 change which nulled out the file handle after the fclose.
560 JH/32 Bug 1909: Fix OCSP proof verification for cases where the proof is
561 signed directly by the cert-signing cert, rather than an intermediate
562 OCSP-signing cert. This is the model used by LetsEncrypt.
564 JH/33 Bug 1914: Ensure socket is nonblocking before draining after SMTP QUIT.
566 HS/01 Fix leak in verify callout under GnuTLS, about 3MB per recipient on
567 an incoming connection.
569 HS/02 Bug 1802: Do not half-close the connection after sending a request
572 HS/03 Use "auto" as the default EC curve parameter. For OpenSSL < 1.0.2
573 fallback to "prime256v1".
575 JH/34 SECURITY: Use proper copy of DATA command in error message.
576 Could leak key material. Remotely exploitable. CVE-2016-9963.
582 JH/01 Bug 1664: Disable OCSP for GnuTLS library versions at/before 3.3.16
583 and 3.4.4 - once the server is enabled to respond to an OCSP request
584 it does even when not requested, resulting in a stapling non-aware
585 client dropping the TLS connection.
587 TF/01 Code cleanup: Overhaul the debug_selector and log_selector machinery to
588 support variable-length bit vectors. No functional change.
590 TF/02 Improve the consistency of logging incoming and outgoing interfaces.
591 The I= interface field on outgoing lines is now after the H= remote
592 host field, same as incoming lines. There is a separate
593 outgoing_interface log selector which allows you to disable the
596 JH/02 Bug 728: Close logfiles after a daemon-process "exceptional" log write.
597 If not running log_selector +smtp_connection the mainlog would be held
598 open indefinitely after a "too many connections" event, including to a
599 deleted file after a log rotate. Leave the per net connection logging
600 leaving it open for efficiency as that will be quickly detected by the
601 check on the next write.
603 HS/01 Bug 1671: Fix post transport crash.
604 Processing the wait-<transport> messages could crash the delivery
605 process if the message IDs didn't exist for some reason. When
606 using 'split_spool_directory=yes' the construction of the spool
607 file name failed already, exposing the same netto behaviour.
609 JH/03 Bug 425: Capture substrings in $regex1, $regex2 etc from regex &
610 mime_regex ACL conditions.
612 JH/04 Bug 1686: When compiled with EXPERIMENTAL_DSN_INFO: Add extra information
613 to DSN fail messages (bounces): remote IP, remote greeting, remote response
614 to HELO, local diagnostic string.
616 JH/05 Downgrade message for a TLS-certificate-based authentication fail from
617 log line to debug. Even when configured with a tls authenticator many
618 client connections are expected to not authenticate in this way, so
619 an authenticate fail is not an error.
621 HS/02 Add the Exim version string to the process info. This way exiwhat
622 gives some more detail about the running daemon.
624 JH/06 Bug 1395: time-limit caching of DNS lookups, to the TTL value. This may
625 matter for fast-change records such as DNSBLs.
627 JH/07 Bug 1678: Always record an interface option value, if set, as part of a
628 retry record, even if constant. There may be multiple transports with
629 different interface settings and the retry behaviour needs to be kept
632 JH/08 Bug 1586: exiqgrep now refuses to run if there are unexpected arguments.
634 JH/09 Bug 1700: ignore space & tab embedded in base64 during decode.
636 JH/10 Bug 840: fix log_defer_output option of pipe transport
638 JH/11 Bug 830: use same host for all RCPTS of a message, even under
639 hosts_randomize. This matters a lot when combined with mua_wrapper.
641 JH/12 Bug 1706: percent and underbar characters are no longer escaped by the
642 ${quote_pgsql:<string>} operator.
644 JH/13 Bug 1708: avoid misaligned access in cached lookup.
646 JH/14 Change header file name for freeradius-client. Relevant if compiling
647 with Radius support; from the Gentoo tree and checked under Fedora.
649 JH/15 Bug 1712: Introduce $prdr_requested flag variable
651 JH/16 Bug 1714: Permit an empty string as expansion result for transport
652 option transport_filter, meaning no filtering.
654 JH/17 Bug 1713: Fix non-PDKIM_DEBUG build. Patch from Jasen Betts.
656 JH/18 Bug 1709: When built with TLS support, the tls_advertise_hosts option now
657 defaults to "*" (all hosts). The variable is now available when not built
658 with TLS, default unset, mainly to enable keeping the testsuite sane.
659 If a server certificate is not supplied (via tls_certificate) an error is
660 logged, and clients will find TLS connections fail on startup. Presumably
661 they will retry in-clear.
662 Packagers of Exim are strongly encouraged to create a server certificate
663 at installation time.
665 HS/03 Add -bP config_file as a synonym for -bP configure_file, for consistency
666 with the $config_file variable.
668 JH/19 Two additional event types: msg:rcpt:defer and msg:rcpt:host:defer. Both
669 in transport context, after the attempt, and per-recipient. The latter type
670 is per host attempted. The event data is the error message, and the errno
671 information encodes the lookup type (A vs. MX) used for the (first) host,
672 and the trailing two digits of the smtp 4xx response.
674 GF/01 Bug 1715: Fix for race condition in exicyclog, where exim could attempt
675 to write to mainlog (or rejectlog, paniclog) in the window between file
676 creation and permissions/ownership being changed. Particularly affects
677 installations where exicyclog is run as root, rather than exim user;
678 result is that the running daemon panics and dies.
680 JH/20 Bug 1701: For MySQL lookups, support MySQL config file option group names.
682 JH/21 Bug 1720: Add support for priority groups and weighted-random proxy
683 selection for the EXPERIMENTAL_SOCKS feature, via new per-proxy options
684 "pri" and "weight". Note that the previous implicit priority given by the
685 list order is no longer honoured.
687 JH/22 Bugs 963, 1721: Fix some corner cases in message body canonicalization
690 JH/23 Move SOCKS5 support from Experimental to mainline, enabled for a build
691 by defining SUPPORT_SOCKS.
693 JH/26 Move PROXY support from Experimental to mainline, enabled for a build
694 by defining SUPPORT_PROXY. Note that the proxy_required_hosts option
695 is renamed to hosts_proxy, and the proxy_{host,target}_{address,port}.
696 variables are renamed to proxy_{local,external}_{address,port}.
698 JH/27 Move Internationalisation support from Experimental to mainline, enabled
699 for a build by defining SUPPORT_I18N
701 JH/28 Bug 1745: Fix redis lookups to handle (quoted) spaces embedded in parts
702 of the query string, and make ${quote_redis:} do that quoting.
704 JH/29 Move Events support from Experimental to mainline, enabled by default
705 and removable for a build by defining DISABLE_EVENT.
707 JH/30 Updated DANE implementation code to current from Viktor Dukhovni.
709 JH/31 Fix bug with hosts_connection_nolog and named-lists which were wrongly
710 cached by the daemon.
712 JH/32 Move Redis support from Experimental to mainline, enabled for a build
713 by defining LOOKUP_REDIS. The libhiredis library is required.
715 JH/33 Bug 1748: Permit ACL dnslists= condition in non-smtp ACLs if explicit
716 keys are given for lookup.
718 JH/34 Bug 1192: replace the embedded copy of PolarSSL RSA routines in the DKIM
719 support, by using OpenSSL or GnuTLS library ones. This means DKIM is
720 only supported when built with TLS support. The PolarSSL SHA routines
721 are still used when the TLS library is too old for convenient support.
723 JH/35 Require SINGLE_DH_USE by default in OpenSSL (main config option
724 openssl_options), for security. OpenSSL forces this from version 1.1.0
725 server-side so match that on older versions.
727 JH/36 Bug 1778: longstanding bug in memory use by the ${run } expansion: A fresh
728 allocation for $value could be released as the expansion processing
729 concluded, but leaving the global pointer active for it.
731 JH/37 Bug 1769: Permit a VRFY ACL to override the default 252 response,
732 and to use the domains and local_parts ACL conditions.
734 JH/38 Fix cutthrough bug with body lines having a single dot. The dot was
735 incorrectly not doubled on cutthrough transmission, hence seen as a
736 body-termination at the receiving system - resulting in truncated mails.
737 Commonly the sender saw a TCP-level error, and retransmitted the message
738 via the normal store-and-forward channel. This could result in duplicates
739 received - but deduplicating mailstores were liable to retain only the
740 initial truncated version.
742 JH/39 Bug 1781: Fix use of DKIM private-keys having trailing '=' in the base-64.
744 JH/40 Fix crash in queryprogram router when compiled with EXPERIMENTAL_SRS.
746 JH/41 Bug 1792: Fix selection of headers to sign for DKIM: bottom-up. While
747 we're in there, support oversigning also; bug 1309.
749 JH/42 Bug 1796: Fix error logged on a malware scanner connection failure.
751 HS/04 Add support for keep_environment and add_environment options.
753 JH/43 Tidy coding issues detected by gcc --fsanitize=undefined. Some remain;
754 either intentional arithmetic overflow during PRNG, or testing config-
757 JH/44 Bug 1800: The combination of a -bhc commandline option and cutthrough
758 delivery resulted in actual delivery. Cancel cutthrough before DATA
761 JH/45 Fix cutthrough, when connection not opened by verify and target hard-
762 rejects a recipient: pass the reject to the originator.
764 JH/46 Multiple issues raised by Coverity. Some were obvious or plausible bugs.
765 Many were false-positives and ignorable, but it's worth fixing the
768 JH/47 Fix build on HP-UX and older Solaris, which need (un)setenv now also
769 for the new environment-manipulation done at startup. Move the routines
770 from being local to tls.c to being global via the os.c file.
772 JH/48 Bug 1807: Fix ${extract } for the numeric/3-string case. While preparsing
773 an extract embedded as result-arg for a map, the first arg for extract
774 is unavailable so we cannot tell if this is a numbered or keyed
775 extraction. Accept either.
781 JH/01 Bug 1545: The smtp transport option "retry_include_ip_address" is now
784 JH/02 The smtp transport option "multi_domain" is now expanded.
786 JH/03 The smtp transport now requests PRDR by default, if the server offers
789 JH/04 Certificate name checking on server certificates, when exim is a client,
790 is now done by default. The transport option tls_verify_cert_hostnames
791 can be used to disable this per-host. The build option
792 EXPERIMENTAL_CERTNAMES is withdrawn.
794 JH/05 The value of the tls_verify_certificates smtp transport and main options
795 default to the word "system" to access the system default CA bundle.
796 For GnuTLS, only version 3.0.20 or later.
798 JH/06 Verification of the server certificate for a TLS connection is now tried
799 (but not required) by default. The verification status is now logged by
800 default, for both outbound TLS and client-certificate supplying inbound
803 JH/07 Changed the default rfc1413 lookup settings to disable calls. Few
806 JH/08 The EXPERIMENTAL_DSN compile option is no longer needed; all Delivery
807 Status Notification (bounce) messages are now MIME format per RFC 3464.
808 Support for RFC 3461 DSN options NOTIFY,ENVID,RET,ORCPT can be advertised
809 under the control of the dsn_advertise_hosts option, and routers may
810 have a dsn_lasthop option.
812 JH/09 A timeout of 2 minutes is now applied to all malware scanner types by
813 default, modifiable by a malware= option. The list separator for
814 the options can now be changed in the usual way. Bug 68.
816 JH/10 The smtp_receive_timeout main option is now expanded before use.
818 JH/11 The incoming_interface log option now also enables logging of the
819 local interface on delivery outgoing connections.
821 JH/12 The cutthrough-routing facility now supports multi-recipient mails,
822 if the interface and destination host and port all match.
824 JH/13 Bug 344: The verify = reverse_host_lookup ACL condition now accepts a
827 JH/14 Bug 1573: The spam= ACL condition now additionally supports Rspamd.
828 Patch from Andrew Lewis.
830 JH/15 Bug 670: The spamd_address main option (for the spam= ACL condition)
831 now supports optional time-restrictions, weighting, and priority
832 modifiers per server. Patch originally by <rommer@active.by>.
834 JH/16 The spamd_address main option now supports a mixed list of local
835 and remote servers. Remote servers can be IPv6 addresses, and
836 specify a port-range.
838 JH/17 Bug 68: The spamd_address main option now supports an optional
839 timeout value per server.
841 JH/18 Bug 1581: Router and transport options headers_add/remove can
842 now have the list separator specified.
844 JH/19 Bug 392: spamd_address, and clamd av_scanner, now support retry
847 JH/20 Bug 1571: Ensure that $tls_in_peerdn is set, when verification fails
850 JH/21 Support for the A6 type of dns record is withdrawn.
852 JH/22 Bug 608: The result of a QUIT or not-QUIT toplevel ACL now matters
853 rather than the verbs used.
855 JH/23 Bug 1572: Increase limit on SMTP confirmation message copy size
856 from 255 to 1024 chars.
858 JH/24 Verification callouts now attempt to use TLS by default.
860 HS/01 DNSSEC options (dnssec_require_domains, dnssec_request_domains)
861 are generic router options now. The defaults didn't change.
863 JH/25 Bug 466: Add RFC2322 support for MIME attachment filenames.
864 Original patch from Alexander Shikoff, worked over by JH.
866 HS/02 Bug 1575: exigrep falls back to autodetection of compressed
867 files if ZCAT_COMMAND is not executable.
869 JH/26 Bug 1539: Add timeout/retry options on dnsdb lookups.
871 JH/27 Bug 286: Support SOA lookup in dnsdb lookups.
873 JH/28 Bug 1588: Do not use the A lookup following an AAAA for setting the FQDN.
874 Normally benign, it bites when the pair was led to by a CNAME;
875 modern usage is to not canonicalize the domain to a CNAME target
876 (and we were inconsistent anyway for A-only vs AAAA+A).
878 JH/29 Bug 1632: Removed the word "rejected" from line logged for ACL discards.
880 JH/30 Check the forward DNS lookup for DNSSEC, in addition to the reverse,
881 when evaluating $sender_host_dnssec.
883 JH/31 Check the HELO verification lookup for DNSSEC, adding new
884 $sender_helo_dnssec variable.
886 JH/32 Bug 1397: Enable ECDHE on OpenSSL, just the NIST P-256 curve.
888 JH/33 Bug 1346: Note MAIL cmd seen in -bS batch, to avoid smtp_no_mail log.
890 JH/34 Bug 1648: Fix a memory leak seen with "mailq" and large queues.
892 JH/35 Bug 1642: Fix support of $spam_ variables at delivery time. Was
893 documented as working, but never had. Support all but $spam_report.
895 JH/36 Bug 1659: Guard checking of input smtp commands again pseudo-command
896 added for tls authenticator.
898 HS/03 Add perl_taintmode main config option
904 TL/01 When running the test suite, the README says that variables such as
905 no_msglog_check are global and can be placed anywhere in a specific
906 test's script, however it was observed that placement needed to be near
907 the beginning for it to behave that way. Changed the runtest perl
908 script to read through the entire script once to detect and set these
909 variables, reset to the beginning of the script, and then run through
910 the script parsing/test process like normal.
912 TL/02 The BSD's have an arc4random API. One of the functions to induce
913 adding randomness was arc4random_stir(), but it has been removed in
914 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
915 function when detected.
917 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
918 cause callback expansion.
920 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
921 syntax errors in an expansion can be treated as a string instead of
922 logging or causing an error, due to the internal use of bool_lax
923 instead of bool when processing it.
925 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
926 server certificates when making smtp deliveries.
928 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
930 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
932 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
934 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
935 Merged patch from Sebastian Wiedenroth.
937 JH/05 Fix results-pipe from transport process. Several recipients, combined
938 with certificate use, exposed issues where response data items split
939 over buffer boundaries were not parsed properly. This eventually
940 resulted in duplicates being sent. This issue only became common enough
941 to notice due to the introduction of connection certificate information,
942 the item size being so much larger. Found and fixed by Wolfgang Breyha.
944 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
945 size buffer was used, resulting in syntax errors when an expansion
948 JH/07 Add support for directories of certificates when compiled with a GnuTLS
949 version 3.3.6 or later.
951 JH/08 Rename the TPDA experimental facility to Event Actions. The #ifdef
952 is EXPERIMENTAL_EVENT, the main-configuration and transport options
953 both become "event_action", the variables become $event_name, $event_data
954 and $event_defer_errno. There is a new variable $verify_mode, usable in
955 routers, transports and related events. The tls:cert event is now also
956 raised for inbound connections, if the main configuration event_action
959 TL/06 In test suite, disable OCSP for old versions of openssl which contained
960 early OCSP support, but no stapling (appears to be less than 1.0.0).
962 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
963 server certificate names available under the smtp transport option
964 "tls_verify_cert_hostname" now do not permit multi-component wildcard
967 JH/10 Time-related extraction expansions from certificates now use the main
968 option "timezone" setting for output formatting, and are consistent
969 between OpenSSL and GnuTLS compilations. Bug 1541.
971 JH/11 Fix a crash in mime ACL when meeting a zero-length, quoted or RFC2047-
972 encoded parameter in the incoming message. Bug 1558.
974 JH/12 Bug 1527: Autogrow buffer used in reading spool files. Since they now
975 include certificate info, eximon was claiming there were spoolfile
978 JH/13 Bug 1521: Fix ldap lookup for single-attr request, multiple-attr return.
980 JH/14 Log delivery-related information more consistently, using the sequence
981 "H=<name> [<ip>]" wherever possible.
983 TL/07 Bug 1547: Omit RFCs from release. Draft and RFCs have licenses which
984 are problematic for Debian distribution, omit them from the release
987 JH/15 Updates and fixes to the EXPERIMENTAL_DSN feature.
989 JH/16 Fix string representation of time values on 64bit time_t architectures.
992 JH/17 Fix a null-indirection in certextract expansions when a nondefault
993 output list separator was used.
998 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
999 checkers that were complaining about end of non-void function with no
1002 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
1003 This was a regression introduced in 4.83 by another bugfix.
1005 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
1007 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
1008 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
1014 TF/01 Correctly close the server side of TLS when forking for delivery.
1016 When a message was received over SMTP with TLS, Exim failed to clear up
1017 the incoming connection properly after forking off the child process to
1018 deliver the message. In some situations the subsequent outgoing
1019 delivery connection happened to have the same fd number as the incoming
1020 connection previously had. Exim would try to use TLS and fail, logging
1021 a "Bad file descriptor" error.
1023 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
1024 utilities have not been installed.
1026 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
1027 temporary space as the ACL may create new global variables.
1029 TL/01 LDAP support uses per connection or global context settings, depending
1030 upon the detected version of the libraries at build time.
1032 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
1033 to extract and use the src ip:port in logging and expansions as if it
1034 were a direct connection from the outside internet. PPv2 support was
1035 updated based on HAProxy spec change in May 2014.
1037 JH/02 Add ${listextract {number}{list}{success}{fail}}.
1039 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
1040 Properly escape header and check for NULL return.
1042 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
1045 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
1047 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
1048 characters in header names, implemented as a verify condition.
1049 Contributed by Michael Fischer v. Mollard.
1051 TL/05 Rename SPF condition results err_perm and err_temp to standardized
1052 results permerror and temperror. Previous values are deprecated but
1053 still accepted. In a future release, err_perm and err_temp will be
1054 completely removed, which will be a backward incompatibility if the
1055 ACL tests for either of these two old results. Patch contributed by
1056 user bes-internal on the mailing list.
1058 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
1060 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
1061 selectors, in both main and reject logs.
1063 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
1066 JH/07 Add malware type "sock" for talking to simple daemon.
1068 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
1070 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
1071 routers/transports under cutthrough routing.
1073 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
1074 numbers. Touch up "bool" conditional to keep the same definition.
1076 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
1078 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
1080 JH/12 Expand items in router/transport headers_add or headers_remove lists
1081 individually rather than the list as a whole. Bug 1452.
1083 Required for reasonable handling of multiple headers_ options when
1084 they may be empty; requires that headers_remove items with embedded
1085 colons must have them doubled (or the list-separator changed).
1087 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
1088 view the policy declared in the DMARC record. Currently, $dmarc_status
1089 is a combined value of both the record presence and the result of the
1092 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
1094 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
1095 dnslookup router and the smtp transport (applying to the forward
1098 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
1099 of ldap servers used for a specific lookup. Patch provided by Heiko
1102 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
1103 New variable $lookup_dnssec_authenticated for observability.
1105 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
1106 Patch submitted by Lars Timman.
1108 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
1110 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
1111 Requires trusted mode and valid format message id, aborts otherwise.
1112 Patch contributed by Heiko Schlichting.
1114 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
1115 certextract with support for various fields. Bug 1358.
1117 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
1118 is requested by default, modifiable by smtp transport option
1121 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
1122 operate on certificate variables to give certificate fingerprints
1123 Also new ${sha256:cert_variable}.
1125 JH/23 The PRDR feature is moved from being Experimental into the mainline.
1127 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
1128 Christian Aistleitner.
1130 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
1132 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
1133 file. Patch from Wolfgang Breyha.
1135 JH/25 Expand the coverage of the delivery $host and $host_address to
1136 client authenticators run in verify callout. Bug 1476.
1138 JH/26 Port service names are now accepted for tls_on_connect_ports, to
1139 align with daemon_smtp_ports. Bug 72.
1141 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
1142 support and error reporting did not work properly.
1144 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
1145 and is readable. Patch from Andrew Colin Kissa.
1147 TL/14 Enhance documentation of ${run expansion and how it parses the
1148 commandline after expansion, particularly in the case when an
1149 unquoted variable expansion results in an empty value.
1151 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
1153 PP/02 Fix internal collision of T_APL on systems which support RFC3123
1154 by renaming away from it. Addresses GH issue 15, reported by
1157 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
1159 TL/15 SECURITY: prevent double expansion in math comparison functions
1160 (can expand unsanitized data). Not remotely exploitable.
1167 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
1169 PP/02 Make -n do something, by making it not do something.
1170 When combined with -bP, the name of an option is not output.
1172 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
1175 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
1176 $sender_host_name and config options to manage this, and basic check
1179 PP/05 DSCP support for outbound connections and control modifier for inbound.
1181 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
1182 (Only plugin which currently uses this is kerberos4, which nobody should
1183 be using, but we should make it available and other future plugins might
1184 conceivably use it, even though it would break NAT; stuff *should* be
1185 using channel bindings instead).
1187 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
1188 name; added for Sendmail compatibility; requires admin caller.
1189 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
1190 just ignore it); requires trusted caller.
1191 Also parse but ignore: -Ac -Am -X<logfile>
1194 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
1196 TL/02 Add +smtp_confirmation as a default logging option.
1198 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
1199 Patch by Magnus Holmgren from 2007-02-20.
1201 TL/04 Bugzilla 1281 - Spec typo.
1202 Bugzilla 1283 - Spec typo.
1203 Bugzilla 1290 - Spec grammar fixes.
1205 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
1207 TL/06 Add Experimental DMARC support using libopendmarc libraries.
1209 TL/07 Fix an out of order global option causing a segfault. Reported to dev
1210 mailing list by by Dmitry Isaikin.
1212 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
1214 JH/02 Support "G" suffix to numbers in ${if comparisons.
1216 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
1218 NM/01 Bugzilla 1197 - Spec typo
1219 Bugzilla 1196 - Spec examples corrections
1221 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
1223 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
1224 gnutls_enable_pkcs11, but renamed to more accurately indicate its
1227 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
1228 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
1230 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
1231 "acl {{name}{arg}...}", and optional args on acl condition
1234 JH/05 Permit multiple router/transport headers_add/remove lines.
1236 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
1238 JH/07 Avoid using a waiting database for a single-message-only transport.
1239 Performance patch from Paul Fisher. Bugzilla 1262.
1241 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
1244 JH/09 Add $headers_added variable, with content from use of ACL modifier
1245 add_header (but not yet added to the message). Bugzilla 199.
1247 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
1248 Pulled from Bugzilla 817 by Wolfgang Breyha.
1250 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
1252 (nb: this is the same fix as in Exim 4.80.1)
1254 JH/11 Add A= logging on delivery lines, and a client_set_id option on
1257 JH/12 Add optional authenticated_sender logging to A= and a log_selector
1260 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
1262 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
1263 advertise SMTP AUTH mechanism to us, instead of a generic
1264 protocol violation error. Also, make Exim more robust to bad
1265 data from the Dovecot auth socket.
1267 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
1269 When a queue runner is handling a message, Exim first routes the
1270 recipient addresses, during which it prunes them based on the retry
1271 hints database. After that it attempts to deliver the message to
1272 any remaining recipients. It then updates the hints database using
1275 So if a recipient address works intermittently, it can get repeatedly
1276 deferred at routing time. The retry hints record remains fresh so the
1277 address never reaches the final cutoff time.
1279 This is a fairly common occurrence when a user is bumping up against
1280 their storage quota. Exim had some logic in its local delivery code
1281 to deal with this. However it did not apply to per-recipient defers
1282 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
1284 This change adds a proper retry rule check during routing so that the
1285 final cutoff time is checked against the message's age. We only do
1286 this check if there is an address retry record and there is not a
1287 domain retry record; this implies that previous attempts to handle
1288 the address had the retry_use_local_parts option turned on. We use
1289 this as an approximation for the destination being like a local
1290 delivery, as in LMTP.
1292 I suspect this new check makes the old local delivery cutoff check
1293 redundant, but I have not verified this so I left the code in place.
1295 TF/02 Correct gecos expansion when From: is a prefix of the username.
1297 Test 0254 submits a message to Exim with the header
1301 When I ran the test suite under the user fanf2, Exim expanded
1302 the header to contain my full name, whereas it should have added
1303 a Resent-Sender: header. It erroneously treats any prefix of the
1304 username as equal to the username.
1306 This change corrects that bug.
1308 GF/01 DCC debug and logging tidyup
1309 Error conditions log to paniclog rather than rejectlog.
1310 Debug lines prefixed by "DCC: " to remove any ambiguity.
1312 TF/03 Avoid unnecessary rebuilds of lookup-related code.
1314 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
1315 Bug spotted by Jeremy Harris; was flawed since initial commit.
1316 Would have resulted in OCSP responses post-SNI triggering an Exim
1317 NULL dereference and crash.
1319 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
1321 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
1322 Bug detection, analysis and fix by Samuel Thibault.
1323 Bugzilla 1331, Debian bug #698092.
1325 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
1327 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
1328 Server implementation by Todd Lyons, client by JH.
1329 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
1330 config variable "prdr_enable" controls whether the server
1331 advertises the facility. If the client requests PRDR a new
1332 acl_data_smtp_prdr ACL is called once for each recipient, after
1333 the body content is received and before the acl_smtp_data ACL.
1334 The client is controlled by both of: a hosts_try_prdr option
1335 on the smtp transport, and the server advertisement.
1336 Default client logging of deliveries and rejections involving
1337 PRDR are flagged with the string "PRDR".
1339 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
1340 fclose(). Diagnosis by Todd Lyons.
1342 PP/17 Update configure.default to handle IPv6 localhost better.
1343 Patch by Alain Williams (plus minor tweaks).
1346 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
1347 This is now consistent with GnuTLS, and is now documented: the
1348 previous undocumented portable approach to treating the option as
1349 unset was to force an expansion failure. That still works, and
1350 an empty string is now equivalent.
1352 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
1353 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
1354 not performing validation itself.
1356 PP/20 Added force_command boolean option to pipe transport.
1357 Patch from Nick Koston, of cPanel Inc.
1359 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
1362 TF/04 Added udpsend ACL modifier and hexquote expansion operator
1364 PP/21 Fix eximon continuous updating with timestamped log-files.
1365 Broken in a format-string cleanup in 4.80, missed when I repaired the
1366 other false fix of the same issue.
1367 Report and fix from Heiko Schlichting.
1370 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
1371 Report from Prashanth Katuri.
1373 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
1374 It's SecureTransport, so affects any MacOS clients which use the
1375 system-integrated TLS libraries, including email clients.
1377 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
1378 using a MIME ACL for non-SMTP local injection.
1379 Report and assistance in diagnosis by Warren Baker.
1381 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
1383 JH/16 Fix comparisons for 64b. Bugzilla 1385.
1385 TL/09 Add expansion variable $authenticated_fail_id to keep track of
1386 last id that failed so it may be referenced in subsequent ACL's.
1388 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
1391 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
1392 ldap library initialization, allowing self-signed CA's to be
1393 used. Also properly sets require_cert option later in code by
1394 using NULL (global ldap config) instead of ldap handle (per
1395 session). Bug diagnosis and testing by alxgomz.
1397 TL/12 Enhanced documentation in the ratelimit.pl script provided in
1398 the src/util/ subdirectory.
1400 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
1401 renamed to Transport Post Delivery Action by Jeremy Harris, as
1404 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
1405 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
1406 redis_servers = needs to be configured which will be used by the redis
1407 lookup. Patch from Warren Baker, of The Packet Hub.
1409 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
1411 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
1412 hostname or reverse DNS when processing a host list. Used suggestions
1413 from multiple comments on this bug.
1415 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
1417 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
1418 Missed a few lines, added it to make the runtest require no keyboard
1421 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
1422 contains upper case chars. Make router use caseful_local_part.
1424 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
1425 support when GnuTLS has been built with p11-kit.
1431 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
1433 This, or similar/improved, will also be change PP/11 of 4.82.
1439 PP/01 Handle short writes when writing local log-files.
1440 In practice, only affects FreeBSD (8 onwards).
1441 Bugzilla 1053, with thanks to Dmitry Isaikin.
1443 NM/01 Bugzilla 949 - Documentation tweak
1445 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
1448 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
1450 PP/02 Implemented gsasl authenticator.
1452 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
1454 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
1455 `pkg-config foo` for cflags/libs.
1457 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
1458 with rest of GSASL and with heimdal_gssapi.
1460 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
1461 `pkg-config foo` for cflags/libs for the TLS implementation.
1463 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
1464 properties get this fed in as external SSF. A number of robustness
1465 and debugging improvements to the cyrus_sasl authenticator.
1467 PP/08 cyrus_sasl server now expands the server_realm option.
1469 PP/09 Bugzilla 1214 - Log authentication information in reject log.
1470 Patch by Jeremy Harris.
1472 PP/10 Added dbmjz lookup type.
1474 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
1476 PP/12 MAIL args handles TAB as well as SP, for better interop with
1477 non-compliant senders.
1478 Analysis and variant patch by Todd Lyons.
1480 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
1481 Bug report from Lars Müller <lars@samba.org> (via SUSE),
1482 Patch from Dirk Mueller <dmueller@suse.com>
1484 PP/13 tls_peerdn now print-escaped for spool files.
1485 Observed some $tls_peerdn in wild which contained \n, which resulted
1486 in spool file corruption.
1488 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
1489 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
1490 or write after TLS renegotiation, which otherwise led to messages
1493 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
1494 as a tracking header (ie: a signed header comes before the signature).
1495 Patch from Wolfgang Breyha.
1497 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
1498 comma-sep list; embedded commas doubled.
1500 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
1502 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
1504 Report and patch from Dmitry Banschikov.
1506 PP/16 Removed "dont_insert_empty_fragments" from "openssl_options".
1507 Removed SSL_clear() after SSL_new() which led to protocol negotiation
1508 failures. We appear to now support TLS1.1+ with Exim.
1510 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
1511 lets Exim select keys and certificates based upon TLS SNI from client.
1512 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
1513 before an outbound SMTP session. New log_selector, +tls_sni.
1515 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
1516 NULL dereference. Report and patch from Alun Jones.
1518 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
1519 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
1520 Not seeing resolver debug output on NetBSD, but suspect this is a
1521 resolver implementation change.
1523 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
1524 Left warnings. Added "eximon gdb" invocation mode.
1526 PP/21 Defaulting "accept_8bitmime" to true, not false.
1528 PP/22 Added -bw for inetd wait mode support.
1530 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
1531 locate the relevant includes and libraries. Made this the default.
1533 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
1534 Bugzilla 1246, report and most of solution from Tomasz Kusy.
1536 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
1537 This may cause build issues on older platforms.
1539 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
1540 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
1541 gnutls_require_mac & gnutls_require_protocols (no longer supported).
1542 Added SNI support via GnuTLS too.
1543 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
1545 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
1547 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
1548 Applied second patch from Janne, implementing suggestion to default
1549 multiple-strings-in-record handling to match SPF spec.
1551 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
1553 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
1554 read-only, out of scope).
1555 Patch from Wolfgang Breyha, report from Stuart Northfield.
1557 PP/29 Fix three issues highlighted by clang analyser static analysis.
1558 Only crash-plausible issue would require the Cambridge-specific
1559 iplookup router and a misconfiguration.
1560 Report from Marcin Mirosław.
1562 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
1564 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
1565 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
1566 As part of this, removing so much warning spew let me fix some minor
1567 real issues in debug logging.
1569 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
1570 assignment on my part. Fixed.
1572 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
1573 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
1574 Janne Snabb (who went above and beyond: thank you).
1576 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
1577 string otherwise requires a connection and a bunch more work and it's
1578 relatively easy to get wrong. Should also expose TLS library linkage
1581 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
1582 64-bit ${eval} (JH/03).
1584 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
1585 GNU libc to support some of the 64-bit stuff, should not lead to
1586 conflicts. Defined before os.h is pulled in, so if a given platform
1587 needs to override this, it can.
1589 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
1590 protection layer was required, which is not implemented.
1591 Bugzilla 1254, patch from Wolfgang Breyha.
1593 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
1594 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
1595 tls_dhparam take prime identifiers. Also unbreak combination of
1596 OpenSSL+DH_params+TLSSNI.
1598 PP/39 Disable SSLv2 by default in OpenSSL support.
1604 PP/01 Solaris build fix for Oracle's LDAP libraries.
1605 Bugzilla 1109, patch from Stephen Usher.
1607 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
1609 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
1612 TF/02 Fix a couple more cases where we did not log the error message
1613 when unlink() failed. See also change 4.74-TF/03.
1615 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
1616 lock up or crash if it happened to be inside a call to libc when it
1617 got a SIGUSR1 from exiwhat.
1619 The SIGUSR1 handler appends the current process status to the process
1620 log which is later printed by exiwhat. It used to use the general
1621 purpose logging code to do this, but several functions it calls are
1622 not safe for signals.
1624 The new output code in the SIGUSR1 handler is specific to the process
1625 log, and simple enough that it's easy to inspect for signal safety.
1626 Removing some special cases also simplifies the general logging code.
1627 Removing the spurious timestamps from the process log simplifies
1630 TF/04 Improved ratelimit ACL condition.
1632 The /noupdate option has been deprecated in favour of /readonly which
1633 has clearer semantics. The /leaky, /strict, and /readonly update modes
1634 are mutually exclusive. The update mode is no longer included in the
1635 database key; it just determines when the database is updated. (This
1636 means that when you upgrade Exim will forget old rate measurements.)
1638 Exim now checks that the per_* options are used with an update mode that
1639 makes sense for the current ACL. For example, when Exim is processing a
1640 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
1641 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
1642 must specify per_mail/readonly. If you omit the update mode it defaults to
1643 /leaky where that makes sense (as before) or /readonly where required.
1645 The /noupdate option is now undocumented but still supported for
1646 backwards compatibility. It is equivalent to /readonly except that in
1647 ACLs where /readonly is required you may specify /leaky/noupdate or
1648 /strict/noupdate which are treated the same as /readonly.
1650 A useful new feature is the /count= option. This is a generalization
1651 of the per_byte option, so that you can measure the throughput of other
1652 aggregate values. For example, the per_byte option is now equivalent
1653 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
1655 The per_rcpt option has been generalized using the /count= mechanism
1656 (though it's more complicated than the per_byte equivalence). When it is
1657 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
1658 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
1659 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
1660 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
1661 Note that using per_rcpt with a non-readonly update mode in more than
1662 one ACL will cause the recipients to be double-counted. (The per_mail
1663 and per_byte options don't have this problem.)
1665 The handling of very low rates has changed slightly. If the computed rate
1666 is less than the event's count (usually one) then this event is the first
1667 after a long gap. In this case the rate is set to the same as this event's
1668 count, so that the first message of a spam run is counted properly.
1670 The major new feature is a mechanism for counting the rate of unique
1671 events. The new per_addr option counts the number of different
1672 recipients that someone has sent messages to in the last time period. It
1673 behaves like per_rcpt if all the recipient addresses are different, but
1674 duplicate recipient addresses do not increase the measured rate. Like
1675 the /count= option this is a general mechanism, so the per_addr option
1676 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
1677 example, measure the rate that a client uses different sender addresses
1678 with the options per_mail/unique=$sender_address. There are further
1679 details in the main documentation.
1681 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
1683 TF/06 Removed a few PCRE remnants.
1685 TF/07 Automatically extract Exim's version number from tags in the git
1686 repository when doing development or release builds.
1688 PP/02 Raise smtp_cmd_buffer_size to 16kB.
1689 Bugzilla 879. Patch from Paul Fisher.
1691 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
1692 Heavily based on revision 40f9a89a from Simon Arlott's tree.
1695 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
1697 PP/05 Variable $av_failed, true if the AV scanner deferred.
1698 Bugzilla 1078. Patch from John Horne.
1700 PP/06 Stop make process more reliably on build failure.
1701 Bugzilla 1087. Patch from Heiko Schlittermann.
1703 PP/07 Make maildir_use_size_file an _expandable_ boolean.
1704 Bugzilla 1089. Patch from Heiko Schlittermann.
1706 PP/08 Handle ${run} returning more data than OS pipe buffer size.
1707 Bugzilla 1131. Patch from Holger Weiß.
1709 PP/09 Handle IPv6 addresses with SPF.
1710 Bugzilla 860. Patch from Wolfgang Breyha.
1712 PP/10 GnuTLS: support TLS 1.2 & 1.1.
1714 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
1717 PP/11 match_* no longer expand right-hand-side by default.
1718 New compile-time build option, EXPAND_LISTMATCH_RHS.
1719 New expansion conditions, "inlist", "inlisti".
1721 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
1723 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
1725 PP/14 fix log_write() format string regression from TF/03.
1726 Bugzilla 1152. Patch from Dmitry Isaikin.
1732 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
1734 PP/02 Harmonised TLS library version reporting; only show if debugging.
1735 Layout now matches that introduced for other libraries in 4.74 PP/03.
1737 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
1739 PP/04 New "dns_use_edns0" global option.
1741 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
1744 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
1745 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
1747 TK/01 Updated PolarSSL code to 0.14.2.
1748 Bugzilla 1097. Patch from Andreas Metzler.
1750 PP/07 Catch divide-by-zero in ${eval:...}.
1751 Fixes bugzilla 1102.
1753 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
1756 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
1757 format-string attack -- SECURITY: remote arbitrary code execution.
1759 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
1760 time unintentionally subject to list matching rules, letting the header
1761 cause arbitrary Exim lookups (of items which can occur in lists, *not*
1762 arbitrary string expansion). This allowed for information disclosure.
1764 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
1765 INT_MIN/-1 -- value coerced to INT_MAX.
1771 NM/01 Workaround for PCRE version dependency in version reporting
1774 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
1775 This fixes portability to compilers other than gcc, notably
1776 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
1778 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
1779 makefiles for portability to HP-UX and POSIX correctness.
1781 PP/01 Permit LOOKUP_foo enabling on the make command-line.
1782 Also via indented variable definition in the Makefile.
1783 (Debugging by Oliver Heesakkers).
1785 PP/02 Restore caching of spamd results with expanded spamd_address.
1786 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
1788 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
1789 Improves build reliability. Fix from: Frank Elsner
1791 NM/02 Fix wide character breakage in the rfc2047 coding
1792 Fixes bug 1064. Patch from Andrey N. Oktyabrski
1794 NM/03 Allow underscore in dnslist lookups
1795 Fixes bug 1026. Patch from Graeme Fowler
1797 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
1798 Code patches from Adam Ciarcinski of NetBSD.
1800 NM/04 Fixed exiqgrep to cope with mailq missing size issue
1803 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
1804 is logged, to avoid truncation. Patch from John Horne.
1806 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
1807 Patch from Jakob Hirsch.
1809 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
1810 SQL string expansion failure details.
1811 Patch from Andrey Oktyabrski.
1813 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
1814 Patch from Simon Arlott.
1816 PP/09 New lookups functionality failed to compile on old gcc which rejects
1817 extern declarations in function scope.
1818 Patch from Oliver Fleischmann
1820 PP/10 Use sig_atomic_t for flags set from signal handlers.
1821 Check getgroups() return and improve debugging.
1822 Fixed developed for diagnosis in bug 927 (which turned out to be
1825 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
1826 Patch from Mark Zealey.
1828 PP/12 Bugzilla 1056: Improved spamd server selection.
1829 Patch from Mark Zealey.
1831 PP/13 Bugzilla 1086: Deal with maildir quota file races.
1832 Based on patch from Heiko Schlittermann.
1834 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
1835 Patch from Uwe Doering, sign-off by Michael Haardt.
1837 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
1838 variable declaration deep within a block. Bug and patch from
1841 PP/15 lookups-Makefile IRIX compatibility coercion.
1843 PP/16 Make DISABLE_DKIM build knob functional.
1845 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
1846 Patch by Simon Arlott
1848 TF/03 Fix valgrind.h portability to C89 compilers that do not support
1849 variable argument macros. Our copy now differs from upstream.
1855 TF/01 Failure to get a lock on a hints database can have serious
1856 consequences so log it to the panic log.
1858 TF/02 Log LMTP confirmation messages in the same way as SMTP,
1859 controlled using the smtp_confirmation log selector.
1861 TF/03 Include the error message when we fail to unlink a spool file.
1863 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
1864 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
1865 for maintaining out-of-tree patches for some time.
1867 PP/01 Bugzilla 139: Documentation and portability issues.
1868 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
1869 Handle per-OS dynamic-module compilation flags.
1871 PP/02 Let /dev/null have normal permissions.
1872 The 4.73 fixes were a little too stringent and complained about the
1873 permissions on /dev/null. Exempt it from some checks.
1874 Reported by Andreas M. Kirchwitz.
1876 PP/03 Report version information for many libraries, including
1877 Exim version information for dynamically loaded libraries. Created
1878 version.h, now support a version extension string for distributors
1879 who patch heavily. Dynamic module ABI change.
1881 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
1882 privilege escalation vulnerability whereby the Exim run-time user
1883 can cause root to append content of the attacker's choosing to
1886 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
1889 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
1890 If dropping privileges for untrusted macros, we disabled normal logging
1891 on the basis that it would fail; for the Exim run-time user, this is not
1892 the case, and it resulted in successful deliveries going unlogged.
1893 Fixed. Reported by Andreas Metzler.
1899 PP/01 Date: & Message-Id: revert to normally being appended to a message,
1900 only prepend for the Resent-* case. Fixes regression introduced in
1901 Exim 4.70 by NM/22 for Bugzilla 607.
1903 PP/02 Include check_rfc2047_length in configure.default because we're seeing
1904 increasing numbers of administrators be bitten by this.
1906 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
1908 PP/03 Bugzilla 994: added openssl_options main configuration option.
1910 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
1912 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
1914 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
1916 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
1917 without a peer certificate, leading to a segfault because of an
1918 assumption that peers always have certificates. Be a little more
1919 paranoid. Problem reported by Martin Tscholak.
1921 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
1922 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
1923 NB: ClamAV planning to remove STREAM in "middle of 2010".
1924 CL also introduces -bmalware, various -d+acl logging additions and
1925 more caution in buffer sizes.
1927 PP/09 Implemented reverse_ip expansion operator.
1929 PP/10 Bugzilla 937: provide a "debug" ACL control.
1931 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
1933 PP/12 Bugzilla 973: Implement --version.
1935 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
1937 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
1939 PP/15 Bugzilla 816: support multiple condition rules on Routers.
1941 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
1942 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
1943 ignore trailing whitespace.
1945 JJ/02 prevent non-panic DKIM error from being sent to paniclog
1947 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
1950 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
1951 Notification from Dr Andrew Aitchison.
1953 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
1954 ExtendedDetectionInfo response format.
1955 Notification from John Horne.
1957 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
1960 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
1961 XSL and documented dependency on system catalogs, with examples of how
1964 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
1967 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
1968 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
1969 configuration file which is writeable by the Exim user or group.
1971 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
1972 of configuration files to cover files specified with the -C option if
1973 they are going to be used with root privileges, not just the default
1976 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
1977 option (effectively making it always true).
1979 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
1980 files to be used while preserving root privileges.
1982 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
1983 that rogue child processes cannot use them.
1985 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
1986 run-time user, instead of root.
1988 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
1989 Exim run-time user without dropping privileges.
1991 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
1992 result string, instead of calling string_vformat() twice with the same
1995 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
1996 for other users. Others should always drop root privileges if they use
1997 -C on the command line, even for a whitelisted configure file.
1999 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
2001 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
2007 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
2008 $data_path, and $header_path variables; fixed documentation bugs and
2011 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
2012 exipick to access non-standard spools, including the "frozen" queue
2015 NM/01 Bugzilla 965: Support mysql stored procedures.
2016 Patch from Alain Williams
2018 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
2020 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
2021 Patch from Andreas Metzler
2023 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
2024 Patch from Kirill Miazine
2026 NM/05 Bugzilla 671: Added umask to procmail example.
2028 JJ/03 installed exipick 20100323.0, fixing doc bug
2030 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
2031 directory. Notification and patch from Dan Rosenberg.
2033 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
2035 TK/02 Improve log output when DKIM signing operation fails.
2037 MH/01 Treat the transport option dkim_domain as a colon separated
2038 list, not as a single string, and sign the message with each element,
2039 omitting multiple occurences of the same signer.
2041 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
2042 Bugzilla 985, 986. Patch by Simon Arlott
2044 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
2045 Patch by Simon Arlott
2047 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
2048 MBX locking. Notification from Dan Rosenberg.
2054 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
2056 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
2058 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
2060 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
2062 NM/04 Bugzilla 915: Flag broken perl installation during build.
2068 TK/01 Added patch by Johannes Berg that expands the main option
2069 "spamd_address" if it starts with a dollar sign.
2071 TK/02 Write list of recipients to X-Envelope-Sender header when building
2072 the mbox-format spool file for content scanning (suggested by Jakob
2075 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
2076 (http://www.dcc-servers.net/) support via dccifd. Activated by
2077 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
2079 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
2080 by Mark Daniel Reidel <mr@df.eu>.
2082 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
2083 When building exim an external PCRE library is now needed -
2084 PCRE is a system library on the majority of modern systems.
2085 See entry on PCRE_LIBS in EDITME file.
2087 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
2088 conversation. Added nologin parameter to request.
2089 Patch contributed by Kirill Miazine.
2091 TF/01 Do not log submission mode rewrites if they do not change the address.
2093 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
2095 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
2096 log files in place. Contributed by Roberto Lima.
2098 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
2100 TF/03 Bugzilla 615: When checking the local_parts router precondition
2101 after a local_part_suffix or local_part_prefix option, Exim now
2102 does not use the address's named list lookup cache, since this
2103 contains cached lookups for the whole local part.
2105 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
2106 Robert Millan. Documentation is in experimental-spec.txt.
2108 TF/04 Bugzilla 668: Fix parallel build (make -j).
2110 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
2112 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
2113 Patch provided by Jan Srzednicki.
2115 TF/05 Leading white space used to be stripped from $spam_report which
2116 wrecked the formatting. Now it is preserved.
2118 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
2119 that they are available at delivery time.
2121 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
2123 TF/08 TLS error reporting now respects the incoming_interface and
2124 incoming_port log selectors.
2126 TF/09 Produce a more useful error message if an SMTP transport's hosts
2127 setting expands to an empty string.
2129 NM/06 Bugzilla 744: EXPN did not work under TLS.
2130 Patch provided by Phil Pennock.
2132 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
2133 Patch provided by Richard Godbee.
2135 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
2136 acl_smtp_notquit, added index entry.
2138 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
2139 Patch provided by Eugene Bujak.
2141 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
2142 accept(). Patch provided by Maxim Dounin.
2144 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
2145 Patch provided by Phil Pennock.
2147 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
2149 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
2150 Patch provided by Brad "anomie" Jorsch.
2152 NM/14 Bugzilla 622: Added timeout setting to transport filter.
2153 Patch provided by Dean Brooks.
2155 TK/05 Add native DKIM support (does not depend on external libraries).
2157 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
2158 Patch provided by Graeme Fowler.
2160 NM/16 Bugzilla 851: Documentation example syntax fix.
2162 NM/17 Changed NOTICE file to remove references to embedded PCRE.
2164 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
2167 NM/19 Bugzilla 745: TLS version reporting.
2168 Patch provided by Phil Pennock.
2170 NM/20 Bugzilla 167: bool: condition support.
2171 Patch provided by Phil Pennock.
2173 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
2174 clients. Patch provided by Phil Pennock.
2176 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
2177 Patch provided by Brad "anomie" Jorsch.
2179 NM/23 Bugzilla 687: Fix misparses in eximstats.
2180 Patch provided by Heiko Schlittermann.
2182 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
2183 Patch provided by Heiko Schlittermann.
2185 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
2186 plus update to original patch.
2188 NM/26 Bugzilla 799: Documentation correction for ratelimit.
2190 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
2191 Patch provided by David Brownlee.
2193 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
2195 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
2197 NM/30 Bugzilla 888: TLS documentation bugfixes.
2199 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
2201 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
2202 Unlike the original bugzilla I have changed all shell scripts in src tree.
2204 NM/33 Bugzilla 898: Transport filter timeout fix.
2205 Patch by Todd Rinaldo.
2207 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
2208 Patch by Serge Demonchaux.
2210 NM/35 Bugzilla 39: Base64 decode bug fixes.
2211 Patch by Jakob Hirsch.
2213 NM/36 Bugzilla 909: Correct connect() call in dcc code.
2215 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
2217 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
2219 NM/39 Bugzilla 911: Fixed MakeLinks build script.
2225 TK/01 Add preliminary DKIM support. Currently requires a forked version of
2226 ALT-N's libdkim that I have put here:
2227 http://duncanthrax.net/exim-experimental/
2229 Note to Michael Haardt: I had to rename some vars in sieve.c. They
2230 were called 'true' and it seems that C99 defines that as a reserved
2231 keyword to be used with 'bool' variable types. That means you could
2232 not include C99-style headers which use bools without triggering
2233 build errors in sieve.c.
2235 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
2236 as mailq or other aliases. Changed the --help handling significantly
2237 to do whats expected. exim_usage() emits usage/help information.
2239 SC/01 Added the -bylocaldomain option to eximstats.
2241 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
2243 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
2245 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
2251 PH/01 Another patch from the Sieve maintainer.
2253 PH/02 When an IPv6 address is converted to a string for single-key lookup
2254 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
2255 dots are used instead of colons so that keys in lsearch files need not
2256 contain colons. This was done some time before quoting was made available
2257 in lsearch files. However, iplsearch files do require colons in IPv6 keys
2258 (notated using the quote facility) so as to distinguish them from IPv4
2259 keys. This meant that lookups for IP addresses in host lists did not work
2260 for iplsearch lookups.
2262 This has been fixed by arranging for IPv6 addresses to be expressed with
2263 colons if the lookup type is iplsearch. This is not incompatible, because
2264 previously such lookups could never work.
2266 The situation is now rather anomalous, since one *can* have colons in
2267 ordinary lsearch keys. However, making the change in all cases is
2268 incompatible and would probably break a number of configurations.
2270 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
2273 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
2274 conversion specification without a maximum field width, thereby enabling
2275 a rogue spamd server to cause a buffer overflow. While nobody in their
2276 right mind would setup Exim to query an untrusted spamd server, an
2277 attacker that gains access to a server running spamd could potentially
2278 exploit this vulnerability to run arbitrary code as the Exim user.
2280 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
2281 $primary_hostname instead of what libspf2 thinks the hosts name is.
2283 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
2284 a directory entry by the name of the lookup key. Previously, if a
2285 symlink pointed to a non-existing file or a file in a directory that
2286 Exim lacked permissions to read, a lookup for a key matching that
2287 symlink would fail. Now it is enough that a matching directory entry
2288 exists, symlink or not. (Bugzilla 503.)
2290 PH/03 The body_linecount and body_zerocount variables are now exported in the
2293 PH/04 Added the $dnslist_matched variable.
2295 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
2296 This means they are set thereafter only if the connection becomes
2299 PH/06 Added the client_condition to authenticators so that some can be skipped
2300 by clients under certain conditions.
2302 PH/07 The error message for a badly-placed control=no_multiline_responses left
2303 "_responses" off the end of the name.
2305 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
2307 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
2308 (without spaces) instead of just copying the configuration text.
2310 PH/10 Added the /noupdate option to the ratelimit ACL condition.
2312 PH/11 Added $max_received_linelength.
2314 PH/12 Added +ignore_defer and +include_defer to host lists.
2316 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
2317 way in which PCRE > 7.0 is built.
2319 PH/14 Implemented queue_only_load_latch.
2321 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
2322 MAIL command. The effect was to mangle the value on 64-bit systems.
2324 PH/16 Another patch from the Sieve maintainer.
2326 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
2328 PH/18 If a system quota error occurred while trying to create the file for
2329 a maildir delivery, the message "Mailbox is full" was not appended to the
2330 bounce if the delivery eventually timed out. Change 4.67/27 below applied
2331 only to a quota excession during the actual writing of the file.
2333 PH/19 It seems that peer DN values may contain newlines (and other non-printing
2334 characters?) which causes problems in log lines. The DN values are now
2335 passed through string_printing() before being added to log lines.
2337 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
2338 and InterBase are left for another time.)
2340 PH/21 Added message_body_newlines option.
2342 PH/22 Guard against possible overflow in moan_check_errorcopy().
2344 PH/23 POSIX allows open() to be a macro; guard against that.
2346 PH/24 If the recipient of an error message contained an @ in the local part
2347 (suitably quoted, of course), incorrect values were put in $domain and
2348 $local_part during the evaluation of errors_copy.
2354 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
2355 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
2358 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
2359 issue a MAIL command.
2361 PH/02 In an ACL statement such as
2363 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
2365 if a client was not listed at all, or was listed with a value other than
2366 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
2367 the condition was not true (as it should be), so access was not denied.
2368 The bug was that the ! inversion was incorrectly passed on to the second
2369 item. This has been fixed.
2371 PH/03 Added additional dnslists conditions == and =& which are different from
2372 = and & when the dns lookup returns more than one IP address.
2374 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
2375 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
2377 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
2378 FSYNC, which compiles an option called disable_fsync that allows for
2379 bypassing fsync(). The documentation is heavily laced with warnings.
2381 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
2383 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
2384 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
2385 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
2386 including adding "make clean"; (3) Added -fPIC when compiling the test
2387 dynamically loaded module, to get rid of a warning.
2389 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
2390 message fails, move_frozen_messages = true and ignore_bounce_errors_after
2391 = 0s. The bug is otherwise harmless.
2393 PH/07 There was a bug in the dovecot authenticator such that the value of
2394 $auth1 could be overwritten, and so not correctly preserved, after a
2395 successful authentication. This usually meant that the value preserved by
2396 the server_setid option was incorrect.
2398 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
2400 PH/09 Installed PCRE release 7.0.
2402 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
2403 run for batched SMTP input. It is now run at the start of every message
2404 in the batch. While fixing this I discovered that the process information
2405 (output by running exiwhat) was not always getting set for -bs and -bS
2406 input. This is fixed, and it now also says "batched" for BSMTP.
2408 PH/11 Added control=no_pipelining.
2410 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
2411 patch, slightly modified), and move the expansion of helo_data till after
2412 the connection is made in the smtp transport (so it can use these
2415 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
2417 PH/14 Added log_selector = +pid.
2419 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
2421 PH/16 Add ${if forany and ${if forall.
2423 PH/17 Added dsn_from option to vary the From: line in DSNs.
2425 PH/18 Flush SMTP output before performing a callout, unless control =
2426 no_callout_flush is set.
2428 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
2429 was true (the default) a successful delivery failed to delete the retry
2430 item, thus causing premature timeout of the address. The bug is now
2433 PH/20 Added hosts_avoid_pipelining to the smtp transport.
2435 PH/21 Long custom messages for fakedefer and fakereject are now split up
2436 into multiline responses in the same way that messages for "deny" and
2437 other ACL rejections are.
2439 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
2440 with slight modification.
2442 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
2443 draft, changing the syntax and factoring some duplicate code".
2445 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
2446 for deliveries of the second and subsequent messages over the same SMTP
2449 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
2450 ${reduce, with only minor "tidies".
2452 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
2454 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
2455 expansion side effects.
2457 PH/27 When a message times out after an over-quota error from an Exim-imposed
2458 quota, the bounce message says "mailbox is full". This message was not
2459 being given when it was a system quota that was exceeded. It now should
2462 MH/03 Made $recipients available in local_scan(). local_scan() already has
2463 better access to the recipient list through recipients_list[], but
2464 $recipients can be useful in postmaster-provided expansion strings.
2466 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
2467 in the case of a MAIL command with additional options following the
2468 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
2469 were accidentally chopped off.
2471 PH/29 SMTP synchronization checks are implemented when a command is read -
2472 there is a check that no more input is waiting when there shouldn't be
2473 any. However, for some commands, a delay in an ACL can mean that it is
2474 some time before the response is written. In this time, more input might
2475 arrive, invalidly. So now there are extra checks after an ACL has run for
2476 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
2477 pipelining has not been advertised.
2479 PH/30 MH's patch to allow iscntrl() characters to be list separators.
2481 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
2482 returned in the SMTP response when smtp_return_error_details was false.
2483 This has been fixed.
2485 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
2486 instead of the C I/O that was originally supplied, because problems were
2487 reported on Solaris.
2489 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
2490 Exim which did not show up earlier: it was assuming that a call to
2491 SSL_CTX_set_info_callback() might give an error value. In fact, there is
2492 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
2493 was a macro that became an assignment, so it seemed to work. This has
2494 changed to a proper function call with a void return, hence the compile
2495 error. Exim's code has been fixed.
2497 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
2500 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
2502 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
2504 PH/37 If a message is not accepted after it has had an id assigned (e.g.
2505 because it turns out to be too big or there is a timeout) there is no
2506 "Completed" line in the log. When some messages of this type were
2507 selected by exigrep, they were listed as "not completed". Others were
2508 picked up by some special patterns. I have improved the selection
2509 criteria to be more general.
2511 PH/38 The host_find_failed option in the manualroute router can now be set
2512 to "ignore", to completely ignore a host whose IP address cannot be
2513 found. If all hosts are ignored, the behaviour is controlled by the new
2514 host_all_ignored option.
2516 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
2517 homing or because of multiple MX records with /mx) generated more than
2518 one IP address, and the following item turned out to be the local host,
2519 all the secondary addresses of the first item were incorrectly removed
2520 from the list, along with the local host and any following hosts (which
2521 is what is supposed to happen).
2523 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
2524 whoever called Exim into the -H file. In the case of the daemon it was
2525 behaving confusingly. When first started, it used values for whoever
2526 started the daemon, but after a SIGHUP it used the Exim user (because it
2527 calls itself on a restart). I have changed the code so that it now always
2530 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
2531 message are rejected with the same error (e.g. no authentication or bad
2532 sender address), and a DATA command is nevertheless sent (as can happen
2533 with PIPELINING or a stupid MUA), the error message that was given to the
2534 RCPT commands is included in the rejection of the DATA command. This is
2535 intended to be helpful for MUAs that show only the final error to their
2538 PH/42 Another patch from the Sieve maintainer.
2540 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
2541 Eximstats - Fixed some broken HTML links and added missing column headers
2543 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
2544 columns for Rejects, Temp Rejects, Ham, and Spam rows.
2546 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
2548 PH/43 Yet another patch from the Sieve maintainer.
2550 PH/44 I found a way to check for a TCP/IP connection going away before sending
2551 the response to the final '.' that terminates a message, but only in the
2552 case where the client has not sent further data following the '.'
2553 (unfortunately, this is allowed). However, in many cases there won't be
2554 any further data because there won't be any more messages to send. A call
2555 to select() can be used: if it shows that the input is "ready", there is
2556 either input waiting, or the socket has been closed. An attempt to read
2557 the next input character can distinguish the two cases. Previously, Exim
2558 would have sent an OK response which the client would never have see.
2559 This could lead to message repetition. This fix should cure that, at
2560 least in a lot of common cases.
2562 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
2563 advertised in response to EHLO.
2569 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
2570 fixed by 4.65/MH/01 (is this a record?) are fixed:
2572 (i) An empty string was always treated as zero by the numeric comparison
2573 operators. This behaviour has been restored.
2575 (ii) It is documented that the numeric comparison operators always treat
2576 their arguments as decimal numbers. This was broken in that numbers
2577 starting with 0 were being interpreted as octal.
2579 While fixing these problems I realized that there was another issue that
2580 hadn't been noticed. Values of message_size_limit (both the global option
2581 and the transport option) were treated as octal if they started with 0.
2582 The documentation was vague. These values are now always treated as
2583 decimal, and I will make that clear in the documentation.
2589 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
2590 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
2593 MH/01 Don't check that the operands of numeric comparison operators are
2594 integers when their expansion is in "skipping" mode (fixes bug
2595 introduced by 4.64-PH/07).
2597 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
2598 child addresses, Exim now panics and dies. Previously, because the count
2599 is held in a short int, deliveries were likely to be lost. As such a
2600 large number of recipients for a single message is ridiculous
2601 (performance will be very, very poor), I have chosen to impose a limit
2602 rather than extend the field.
2608 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
2609 leftover -K file (the existence of which was triggered by #402).
2610 While we were at it, introduced process PID as part of the -K
2611 filename. This should rule out race conditions when creating
2614 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
2615 processing considerably. Previous code took too long for large mails,
2616 triggering a timeout which in turn triggers #401.
2618 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
2619 in the DK code in transports.c. sendfile() is not really portable,
2620 hence the _LINUX specificness.
2622 TF/01 In the add_headers option to the mail command in an Exim filter,
2623 there was a bug that Exim would claim a syntax error in any
2624 header after the first one which had an odd number of characters
2627 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
2628 callout verification, Exim cached a "reject" for the entire domain. This
2629 is correct for most verifications, but it is not correct for a recipient
2630 verification with use_sender or use_postmaster set, because in that case
2631 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
2632 case of MAIL FROM:<> rejection from other early rejections (e.g.
2633 rejection of HELO). When verifying a recipient using a non-null MAIL
2634 address, the cache is ignored if it shows MAIL FROM:<> rejection.
2635 Whatever the result of the callout, the value of the domain cache is
2636 left unchanged (for any other kind of callout, getting as far as trying
2637 RCPT means that the domain itself is ok).
2639 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
2642 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
2643 manifest itself as EPIPE rather than ECONNECT. When tidying away a
2644 session, the daemon ignores ECONNECT errors and logs others; it now
2645 ignores EPIPE as well.
2647 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
2648 (quoted-printable decoding).
2650 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
2651 later the small subsequent patch to fix an introduced bug.
2653 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
2655 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
2657 PH/08 An error is now given if message_size_limit is specified negative.
2659 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
2660 to be given (somewhat) arbitrary names.
2662 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
2665 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
2666 miscellaneous code fixes
2668 PH/10 Added the log_reject_target ACL modifier to specify where to log
2671 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
2672 hostname. This is wrong, because it relates to the incoming message (and
2673 probably the interface on which it is arriving) and not to the outgoing
2674 callout (which could be using a different interface). This has been
2675 changed to use the value of the helo_data option from the smtp transport
2676 instead - this is what is used when a message is actually being sent. If
2677 there is no remote transport (possible with a router that sets up host
2678 addresses), $smtp_active_hostname is used.
2680 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
2681 tweaks were necessary in order to get it to work (see also 21 below):
2682 (a) The code assumed that strncpy() returns a negative number on buffer
2683 overflow, which isn't the case. Replaced with Exim's string_format()
2685 (b) There were several signed/unsigned issues. I just did the minimum
2686 hacking in of casts. There is scope for a larger refactoring.
2687 (c) The code used strcasecmp() which is not a standard C function.
2688 Replaced with Exim's strcmpic() function.
2689 (d) The code set only $1; it now sets $auth1 as well.
2690 (e) A simple test gave the error "authentication client didn't specify
2691 service in request". It would seem that Dovecot has changed its
2692 interface. Fortunately there's a specification; I followed it and
2693 changed what the client sends and it appears to be working now.
2695 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
2698 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
2699 address A is aliased to B and C, where B exists and C does not. Without
2700 -v the output is "A verified" because verification stops after a
2701 successful redirection if more than one address is generated. However,
2702 with -v the child addresses are also verified. Exim was outputting "A
2703 failed to verify" and then showing the successful verification for C,
2704 with its parentage. It now outputs "B failed to verify", showing B's
2705 parentage before showing the successful verification of C.
2707 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
2708 look up a TXT record in a specific list after matching in a combined
2711 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
2712 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
2713 they consult the DNS. I had assumed they would set it the way they
2714 wanted; and indeed my experiments on Linux seem to show that in some
2715 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
2716 To be on the safe side, however, I have now made the interface to
2717 host_find_byname() similar to host_find_bydns(), with an argument
2718 containing the DNS resolver options. The host_find_byname() function now
2719 sets these options at its start, just as host_find_bydns() does. The smtp
2720 transport options dns_qualify_single and dns_search_parents are passed to
2721 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
2722 of host_find_byname() use the default settings of RES_DEFNAMES
2723 (qualify_single) but not RES_DNSRCH (search_parents).
2725 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
2726 spool_read_header() do less string testing, by means of a preliminary
2727 switch on the second character of optional "-foo" lines. (This is
2728 overdue, caused by the large number of possibilities that now exist.
2729 Originally there were few.) While I was there, I also converted the
2730 str(n)cmp tests so they don't re-test the leading "-" and the first
2731 character, in the hope this might squeeze out yet more improvement.
2733 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
2734 flag allowing group syntax was set by the header_syntax check but not
2735 turned off, possible causing trouble later; (2) The flag was not being
2736 set at all for the header_verify test, causing "group"-style headers to
2737 be rejected. I have now set it in this case, and also caused header_
2738 verify to ignore an empty address taken from a group. While doing this, I
2739 came across some other cases where the code for allowing group syntax
2740 while scanning a header line wasn't quite right (mostly, not resetting
2741 the flag correctly in the right place). These bugs could have caused
2742 trouble for malformed header lines. I hope it is now all correct.
2744 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
2745 with the "reply" argument non-NULL. The code, however (which originally
2746 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
2747 but it didn't always do it. This confused somebody who was copying the
2748 code for some other use. I have removed all the tests.
2750 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
2751 feature that was used to support insecure browsers during the U.S. crypto
2752 embargo. It requires special client support, and Exim is probably the
2753 only MTA that supported it -- and would never use it because real RSA is
2754 always available. This code has been removed, because it had the bad
2755 effect of slowing Exim down by computing (never used) parameters for the
2756 RSA_EXPORT functionality.
2758 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
2759 authenticator to fail if there's a tab character in the incoming data
2760 (there should never be unless someone is messing about, as it's supposed
2761 to be base64-encoded). Also added, on Timo's advice, the "secured" option
2762 if the connection is using TLS or if the remote IP is the same as the
2763 local IP, and the "valid-client-cert option" if a client certificate has
2766 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
2767 authenticators. This can be used for authorization after authentication
2768 succeeds. (In the case of plaintext, it servers for both authentication
2771 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
2772 if any retry times were supplied.
2774 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
2775 connection, where there is no client IP address to check. In this
2776 situation, the verify now always succeeds.
2778 PH/25 Applied John Jetmore's -Mset patch.
2780 PH/26 Added -bem to be like -Mset, but loading a message from a file.
2782 PH/27 In a string expansion for a processed (not raw) header when multiple
2783 headers of the same name were present, leading whitespace was being
2784 removed from all of them, but trailing whitespace was being removed only
2785 from the last one. Now trailing whitespace is removed from each header
2786 before concatenation. Completely empty headers in a concatenation (as
2787 before) are ignored.
2789 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
2790 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
2792 PH/29 [Removed. This was a change that I later backed out, and forgot to
2793 correct the ChangeLog entry (that I had efficiently created) before
2794 committing the later change.]
2796 PH/30 Exim was sometimes attempting to deliver messages that had suffered
2797 address errors (4xx response to RCPT) over the same connection as other
2798 messages routed to the same hosts. Such deliveries are always "forced",
2799 so retry times are not inspected. This resulted in far too many retries
2800 for the affected addresses. The effect occurred only when there were more
2801 hosts than the hosts_max_try setting in the smtp transport when it had
2802 the 4xx errors. Those hosts that it had tried were not added to the list
2803 of hosts for which the message was waiting, so if all were tried, there
2804 was no problem. Two fixes have been applied:
2806 (i) If there are any address or message errors in an SMTP delivery, none
2807 of the hosts (tried or untried) are now added to the list of hosts
2808 for which the message is waiting, so the message should not be a
2809 candidate for sending over the same connection that was used for a
2810 successful delivery of some other message. This seems entirely
2811 reasonable: after all the message is NOT "waiting for some host".
2812 This is so "obvious" that I'm not sure why it wasn't done
2813 previously. Hope I haven't missed anything, but it can't do any
2814 harm, as the worst effect is to miss an optimization.
2816 (ii) If, despite (i), such a delivery is accidentally attempted, the
2817 routing retry time is respected, so at least it doesn't keep
2818 hammering the server.
2820 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
2821 in ${readsocket because some servers need this prod.
2823 PH/32 Added some extra debug output when updating a wait-xxx database.
2825 PH/33 The hint "could be header name not terminated by colon", which has been
2826 given for certain expansion errors for a long time, was not being given
2827 for the ${if def:h_colon_omitted{... case.
2829 PH/34 The spec says: "With one important exception, whenever a domain list is
2830 being scanned, $domain contains the subject domain." There was at least
2831 one case where this was not true.
2833 PH/35 The error "getsockname() failed: connection reset by peer" was being
2834 written to the panic log as well as the main log, but it isn't really
2835 panic-worthy as it just means the connection died rather early on. I have
2836 removed the panic log writing for the ECONNRESET error when getsockname()
2839 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
2840 runs only) independently of the message's sender address. This meant
2841 that, if the 4xx error was in fact related to the sender, a different
2842 message to the same recipient with a different sender could confuse
2843 things. In particular, this can happen when sending to a greylisting
2844 server, but other circumstances could also provoke similar problems.
2845 I have changed the default so that the retry time for these errors is now
2846 based a combination of the sender and recipient addresses. This change
2847 can be overridden by setting address_retry_include_sender=false in the
2850 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
2851 remote server are returned as part of bounce messages. This was not
2852 happening for LMTP over a pipe (the lmtp transport), but now it is the
2853 same for both kinds of LMTP.
2855 PH/38 Despite being documented as not happening, Exim was rewriting addresses
2856 in header lines that were in fact CNAMEs. This is no longer the case.
2858 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
2859 and queue runs started by the daemon processed all messages. This has
2860 been fixed so that -R and -S can now usefully be given with -q<time>.
2862 PH/40 Import PCRE release 6.7 (fixes some bugs).
2864 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
2866 PH/42 Give an error if -q is specified more than once.
2868 PH/43 Renamed the variables $interface_address and $interface_port as
2869 $received_ip_address and $received_port, to make it clear that these
2870 values apply to message reception, and not to the outgoing interface when
2871 a message is delivered. (The old names remain recognized, of course.)
2873 PH/44 There was no timeout on the connect() call when using a Unix domain
2874 socket in the ${readsocket expansion. There now is.
2876 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
2877 be meaningful with "accept".
2879 SC/01 Eximstats V1.43
2880 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
2882 SC/02 Eximstats V1.44
2883 Use a glob alias rather than an array ref in the generated
2884 parser. This improves both readability and performance.
2886 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
2887 Collect SpamAssassin and rejection statistics.
2888 Don't display local sender or destination tables unless
2889 there is data to show.
2890 Added average volumes into the top table text output.
2892 SC/04 Eximstats V1.46
2893 Collect data on the number of addresses (recipients)
2894 as well as the number of messages.
2896 SC/05 Eximstats V1.47
2897 Added 'Message too big' to the list of mail rejection
2898 reasons (thanks to Marco Gaiarin).
2900 SC/06 Eximstats V1.48
2901 Mainlog lines which have GMT offsets and are too short to
2902 have a flag are now skipped.
2904 SC/07 Eximstats V1.49 (Alain Williams)
2905 Added the -emptyok flag.
2907 SC/08 Eximstats V1.50
2908 Fixes for obtaining the IP address from reject messages.
2910 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
2911 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
2912 whitespace changes from 4.64-PH/27
2914 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
2917 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
2918 are found, allow negative numbers in numeric criteria)
2920 JJ/06 exipick.20061117.2, added new $message_body_missing variable
2922 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
2923 to match changes made in 4.64-PH/43
2925 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
2927 PH/47 Put in an explicit test for a DNS lookup of an address record where the
2928 "domain" is actually an IP address, and force a failure. This locks out
2929 those revolvers/nameservers that support "A-for-A" lookups, in
2930 contravention of the specifications.
2932 PH/48 When a host name was looked up from an IP address, and the subsequent
2933 forward lookup of the name timed out, the host name was left in
2934 $sender_host_name, contrary to the specification.
2936 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
2937 restricted to single-key lookups, Exim was not diagnosing an error if
2938 * or *@ was used with a query-style lookup.
2940 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
2942 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
2943 long ago, but noone interested enough thought of it. Let's just say that
2944 the "1.1" means that there are some new functions that weren't there at
2945 some point in the past.
2947 PH/51 Error processing for expansion failure of helo_data from an smtp
2948 transport during callout processing was broken.
2950 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
2951 tested/used via the -bh/-bhc/-bs options.
2953 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
2954 bug, fixed in subsequent PCRE releases).
2956 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
2957 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
2959 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
2965 SC/01 Use a glob alias rather than an array ref in eximstats generated
2966 parser. This improves both readability and performance.
2968 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
2969 Don't display local sender or destination tables in eximstats unless
2970 there is data to show.
2971 Added average volumes into the eximstats top table text output.
2973 SC/03 Collect data on the number of addresses (recipients) as well
2974 as the number of messages in eximstats.
2976 TF/01 Correct an error in the documentation for the redirect router. Exim
2977 does (usually) call initgroups() when daemonizing.
2979 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
2980 with consistent privilege compared to when running as a daemon.
2982 TF/03 Note in the spec that $authenticated_id is not set for local
2983 submissions from trusted users.
2985 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
2986 Thanks to Dean Brooks <dean@iglou.com> for the patch.
2988 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
2989 by adding some example configuration directives to the default
2990 configuration file. A little bit of work is required to uncomment the
2991 directives and define how usernames and passwords are checked, but
2992 there is now a framework to start from.
2994 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
2995 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
2996 without this. I don't know how relevant this is to other LDAP libraries.
2998 PH/02 Add the verb name to the "unknown ACL verb" error.
3000 PH/03 Magnus Holmgren's patch for filter_prepend_home.
3002 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
3004 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
3005 directory not expanded when it should be if an expanded home directory
3006 was set for the address (which is overridden by the transport).
3008 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
3011 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
3012 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
3013 because it is too late at that time, and has no effect.
3015 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
3016 security issue with \' (bugzilla #107). I could not use the
3017 PQescapeStringConn() function, because it needs a PGconn value as one of
3020 PH/08 When testing addresses using -bt, indicate those final addresses that
3021 are duplicates that would not cause an additional delivery. At least one
3022 person was confused, thinking that -bt output corresponded to deliveries.
3023 (Suppressing duplicates isn't a good idea as you lose the information
3024 about possibly different redirections that led to the duplicates.)
3026 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
3027 systems where poll() doesn't work, in particular OS X.
3029 PH/10 Added more information to debugging output for retry time not reached.
3031 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
3032 operations in malware.c.
3034 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
3037 PH/13 If write_rejectlog was set false when logging was sent to syslog with
3038 syslog_duplication set false, log lines that would normally be written
3039 both the the main log and to the reject log were not written to syslog at
3042 PH/14 In the default configuration, change the use of "message" in ACL warn
3043 statements to "add_header".
3045 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
3046 not followed by a command (e.g. "seen endif").
3048 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
3049 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
3052 PH/17 Added extra conditions to the default value of delay_warning_condition
3056 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
3057 { match{$h_precedence:}{(?i)bulk|list|junk} } \
3058 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
3061 The Auto-Submitted: and various List- headers are standardised, whereas I
3062 don't think Precedence: ever was.
3064 PH/18 Refactored debugging code in route_finduser() to show more information,
3065 in particular, the error code if getpwnam() issues one.
3067 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
3068 This is apparently needed in addition to the PH/07 change above to avoid
3069 any possible encoding problems.
3071 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
3072 but not after initializing Perl.
3074 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
3075 output them only if debugging. By default they are written stderr,
3076 apparently, which is not desirable.
3078 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
3081 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
3084 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
3086 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
3087 authenticated or an ident call has been made. Suppress the default
3088 values for $authenticated_id and $authenticated_sender (but permit -oMai
3089 and -oMas) when testing with -bh.
3091 PH/24 Re-jigged the order of the tests in the default configuration so that the
3092 tests for valid domains and recipients precede the DNS black list and CSA
3093 tests, on the grounds that those ones are more expensive.
3095 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
3096 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
3097 This bug exists in every version of Exim that I still have, right back to
3100 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
3101 However, an attempt to turn on case-sensitivity in a regex key by
3102 including (?-i) didn't work because the subject string was already
3103 lowercased, and the effects were non-intuitive. It turns out that a
3104 one-line patch can be used to allow (?-i) to work as expected.
3110 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
3111 other effects) broke the use of negated acl sub-conditions.
3113 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
3116 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
3117 "Deny" causes Exim to reject the incoming connection with a 554 error.
3118 Unfortunately, if there is a major crisis, such as a disk failure,
3119 tcp-wrappers gives "deny", whereas what one would like would be some
3120 kind of temporary error. A kludge has been added to help with this.
3121 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
3122 554 error is used if errno is still zero or contains ENOENT (which occurs
3123 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
3126 PH/03 Add -lutil to the default FreeBSD LIBS setting.
3128 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
3129 errors. Otherwise a message that provokes a temporary error (when other
3130 messages do not) can cause a whole host to time out.
3132 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
3133 addresses were routed directly to files or pipes from a redirect router.
3134 File deliveries just didn't batch; pipe deliveries might have suffered
3137 PH/06 A failure to get a lock for a hints database would erroneously always say
3138 "Failed to get write lock", even when it was really a read lock.
3140 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
3141 of 0600. This has been changed to use the value of the lockfile_mode
3142 option (which defaults to 0600).
3144 PH/08 Applied small patch from the Sieve maintainer.
3146 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
3147 folder from quota calculations, a direct delivery into this folder messed
3148 up the contents of the maildirsize file. This was because the regex was
3149 used only to exclude .Trash (or whatever) when the size of the mailbox
3150 was calculated. There was no check that a delivery was happening into an
3151 excluded directory. This bug has been fixed by ignoring all quota
3152 processing for deliveries into excluded directories.
3154 PH/10 Added the maildirfolder_create_regex option to appendfile.
3160 PH/01 The code for finding all the local interface addresses on a FreeBSD
3161 system running IPv6 was broken. This may well have applied to all BSD
3162 systems, as well as to others that have similar system calls. The broken
3163 code found IPv4 interfaces correctly, but gave incorrect values for the
3164 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
3165 that it would not match correctly against @[] and not recognize the IPv6
3168 PH/02 The ipliteral router was not recognizing addresses of the form user@
3169 [ipv6:....] because it didn't know about the "ipv6:" prefix.
3171 PH/03 Added disable_ipv6.
3173 PH/04 Changed $reply_address to use the raw form of the headers instead of the
3174 decoded form, because it is most often used to construct To: headers
3175 lines in autoreplies, and the decoded form may well be syntactically
3176 invalid. However, $reply_address has leading white space removed, and all
3177 newlines turned into spaces so that the autoreply transport does not
3180 PH/05 If group was specified without a user on a router, and no group or user
3181 was specified on a transport, the group from the router was ignored.
3183 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
3184 for visible compile-time settings that can be used to change these
3185 numbers, for those that want even more. Backwards compatibility with old
3186 spool files has been maintained. However, going back to a previous Exim
3187 release will lost any variables that are in spool files.
3189 PH/07 Two small changes when running in the test harness: increase delay when
3190 passing a TCP/IP connection to a new process, in case the original
3191 process has to generate a bounce, and remove special handling of
3192 127.0.0.2 (sic), which is no longer necessary.
3194 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
3195 be the same on different OS.
3197 PH/09 Moved a debug statement in filter processing to avoid a race problem when
3200 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
3201 whether --show-vars was specified or not
3203 JJ/02 exipick: Added support for new ACL variable spool format introduced
3206 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
3207 syntactically invalid From: or Reply-to: line, and a filter used this to
3208 generate an autoreply, and therefore failed to obtain an address for the
3209 autoreply, Exim could try to deliver to a non-existent relative file
3210 name, causing unrelated and misleading errors. What now happens is that
3211 it logs this as a hard delivery error, but does not attempt to create a
3214 PH/11 The exinext utility has a -C option for testing purposes, but although
3215 the given file was scanned by exinext itself; it wasn't being passed on
3216 when Exim was called.
3218 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
3219 an end-of-file indication when reading a command response.
3221 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
3222 compiled. In many other places in Exim, IPv6 addresses are always
3223 recognized, so I have changed this. It also means that IPv4 domain
3224 literals of the form [IPV4:n.n.n.n] are now always recognized.
3226 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
3227 used if the router is not running as root, for example, when verifying at
3228 ACL time, or when using -bh. The debugging output from this situation was
3229 non-existent - all you got was a failure to exec. I have made two
3232 (a) Failures to set uid/gid, the current directory, or a process leader
3233 in a subprocess such as that created by queryprogram now generate
3234 suitable debugging output when -d is set.
3236 (b) The queryprogram router detects when it is not running as root,
3237 outputs suitable debugging information if -d is set, and then runs
3238 the subprocess without attempting to change uid/gid.
3240 PH/15 Minor change to Makefile for building test_host (undocumented testing
3243 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
3244 additional section of a DNS packet that returns MX or SRV records.
3245 Instead, it always explicitly searches for A/AAAA records. This avoids
3246 major problems that occur when a DNS server includes only records of one
3247 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
3248 fixed another bug: if SRV records were looked up and the corresponding
3249 address records were *not* found in the additional section, the port
3250 values from the SRV records were lost.
3252 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
3253 using the correct key (the original address) when searching the retry
3254 rules in order to find which one to use for generating the retry hint.
3256 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
3257 adding the default one. Similarly, if it contains a Reply-To: header, the
3258 errors_reply_to option, if set, is not used.
3260 PH/19 When calculating a retry time, Exim used to measure the "time since
3261 failure" by looking at the "first failed" field in the retry record. Now
3262 it does not use this if it is later than than the arrival time of the
3263 message. Instead it uses the arrival time. This makes for better
3264 behaviour in cases where some deliveries succeed, thus re-setting the
3265 "first failed" field. An example is a quota failure for a huge message
3266 when small messages continue to be delivered. Without this change, the
3267 "time since failure" will always be short, possible causing more frequent
3268 delivery attempts for the huge message than are intended.
3269 [Note: This change was subsequently modified - see PH/04 for 4.62.]
3271 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
3272 $1, $2, $3) because the numerical variables can be reset during some
3273 expansion items (e.g. "match"), thereby losing the authentication data.
3275 PH/21 Make -bV show the size of off_t variables so that the test suite can
3276 decide whether to run tests for quotas > 2G.
3278 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
3279 mailbox_size, and mailbox_filecount in the appendfile transport. If a
3280 filecount value is greater than 2G or if a quota value is greater than 2G
3281 on a system where the size of off_t is not greater than 4, a panic error
3284 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
3285 never match. The debug and -bh output now contains an explicit error
3286 message indicating a malformed IPv4 address or mask.
3288 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
3289 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
3290 PH/23 above applies.
3292 PH/25 Do not write to syslog when running in the test harness. The only
3293 occasion when this arises is a failure to open the main or panic logs
3294 (for which there is an explicit test).
3296 PH/26 Added the /no_tell option to "control=freeze".
3298 PH/27 If a host name lookup failed very early in a connection, for example, if
3299 the IP address matched host_lookup and the reverse lookup yielded a name
3300 that did not have a forward lookup, an error message of the form "no IP
3301 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
3302 could be logged. Now it outputs the IP address instead of "NULL".
3304 PH/28 An enabling patch from MH: add new function child_open_exim2() which
3305 allows the sender and the authenticated sender to be set when
3306 submitting a message from within Exim. Since child_open_exim() is
3307 documented for local_scan(), the new function should be too.
3309 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
3310 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
3311 results in an empty string is now treated as unset.
3313 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
3315 PH/31 Added sender_verify_fail logging option.
3317 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
3318 needed by RFC 822 but not by RFC 2822 was commented out. I have now
3319 tidied the source and removed it altogether.
3321 PH/33 When a queue run was abandoned because the load average was too high, a
3322 log line was always written; now it is written only if the queue_run log
3323 selector is set. In addition, the log line for abandonment now contains
3324 information about the queue run such as the pid. This is always present
3325 in "start" and "stop" lines but was omitted from the "abandon" line.
3327 PH/34 Omit spaces between a header name and the colon in the error message that
3328 is given when verify = headers_syntax fails (if there are lots of them,
3329 the message gets confusing).
3331 PH/35 Change the default for dns_check_names_pattern to allow slashes within
3332 names, as there are now some PTR records that contain slashes. This check
3333 is only to protect against broken name servers that fall over on strange
3334 characters, so the fact that it applies to all lookups doesn't matter.
3336 PH/36 Now that the new test suite is complete, we can remove some of the
3337 special code in Exim that was needed for the old test suite. For example,
3338 sorting DNS records because real resolvers return them in an arbitrary
3339 order. The new test suite's fake resolver always returns records in the
3342 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
3343 bounces) except when queue_only is set, to avoid logging races between
3344 the different processes.
3346 PH/38 Panic-die if .include specifies a non-absolute path.
3348 PH/39 A tweak to the "H" retry rule from its user.
3350 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
3351 a label. They prevented compilation on older perls.
3353 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
3354 a warning to be raised on newish perls.
3356 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
3357 on queue. Changes to match documented behaviour of showing count of
3358 messages matching specified criteria.
3360 PH/40 Changed the default ident timeout from 30s to 5s.
3362 PH/41 Added support for the use of login_cap features, on those BSD systems
3363 that have them, for controlling the resources used by pipe deliveries.
3365 PH/42 The content-scanning code uses fopen() to create files in which to put
3366 message data. Previously it was not paying any attention to the mode of
3367 the files. Exim runs with umask(0) because the rest of the code creates
3368 files with open(), and sets the required mode explicitly. Thus, these
3369 files were ending up world-writeable. This was not a big issue, because,
3370 being within the spool directory, they were not world-accessible. I have
3371 created a function called modefopen, which takes an additional mode
3372 argument. It sets umask(777), creates the file, chmods it to the required
3373 mode, then resets the umask. All the relevant calls to fopen() in the
3374 content scanning code have been changed to use this function.
3376 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
3377 to 24 hours. This avoids potential overflow problems when processing G
3378 and H retry rules. I suspect nobody ever tinkers with this value.
3380 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
3382 PH/45 When the plaintext authenticator is running as a client, the server's
3383 challenges are checked to ensure they are valid base64 strings. By
3384 default, the authentication attempt is cancelled if an invalid string is
3385 received. Setting client_ignore_invalid_base64 true ignores these errors.
3386 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
3387 they are received. Thus, the responses can be made to depend on the
3388 challenges. If an invalid string is ignored, an empty string is placed in
3391 PH/46 Messages that are created by the autoreply transport now contains a
3392 References: header, in accordance with RFCs 2822 and 3834.
3394 PH/47 Added authenticated_sender_force to the smtp transport.
3396 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
3398 PH/49 Installed latest patch from the Sieve maintainer.
3400 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
3401 was also set, the appendfile transport was unnecessarily scanning a
3402 directory of message files (e.g. for maildir delivery) to find the count
3403 of files (along with the size), even though it did not need this
3404 information. It now does the scan only if it needs to find either the
3405 size of the count of files.
3407 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
3409 PH/52 Two bugs concerned with error handling when the smtp transport is
3412 (i) Exim was not creating retry information for temporary errors given
3413 for individual recipients after the DATA command when the smtp transport
3414 was used in LMTP mode. This meant that they could be retried too
3415 frequently, and not timed out correctly.
3417 (ii) Exim was setting the flag that allows error details to be returned
3418 for LMTP errors on RCPT commands, but not for LMTP errors for individual
3419 recipients that were returned after the DATA command.
3421 PH/53 This is related to PH/52, but is more general: for any failing address,
3422 when detailed error information was permitted to be returned to the
3423 sender, but the error was temporary, then after the final timeout, only
3424 "retry timeout exceeded" was returned. Now it returns the full error as
3425 well as "retry timeout exceeded".
3427 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
3428 do this, and (what is worse) MTAs that accept it.
3430 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
3431 will now be deprecated.
3433 PH/56 New os.c-cygwin from the Cygwin maintainer.
3435 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
3436 formats (previously only available in exim formats via -bpr, -bpru,
3437 and -bpra. Now also available in native and exiqgrep formats)
3439 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
3440 with very large, slow to parse queues
3442 JJ/08 exipick: added ! as generic prefix to negate any criteria format
3444 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
3446 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
3447 responses to authentication challenges, though it was showing the
3448 challenges; (ii) I've removed the CR characters from the debug output for
3451 PH/58 Allow for the insertion of a newline as well as a space when a string
3452 is turned into more than one encoded-word during RFC 2047 encoding. The
3453 Sieve code now uses this.
3455 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
3456 data_4xx, lost_connection, tls_required.
3458 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
3459 message was being sent as an SMTP response.
3461 PH/61 Add -l and -k options to exicyclog.
3463 PH/62 When verifying, if an address was redirected to one new address, so that
3464 verification continued, and the new address failed or deferred after
3465 having set something in $address_data, the value of $address_data was not
3466 passed back to the ACL. This was different to the case when no
3467 redirection occurred. The value is now passed back in both cases.
3469 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
3470 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
3471 use login_cap.h, so on its own it isn't the distinguishing feature. The
3472 new name refers directly to the setclassresources() function.
3474 PH/65 Added configuration files for NetBSD3.
3476 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
3478 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
3479 is preferred over IPv4.
3481 PH/68 The bounce_return_message and bounce_return_body options were not being
3482 honoured for bounces generated during the reception of non-SMTP messages.
3483 In particular, this applied to messages rejected by the ACL. This bug has
3484 been fixed. However, if bounce_return_message is true and bounce_return_
3485 body is false, the headers that are returned for a non-SMTP message
3486 include only those that have been read before the error was detected.
3487 (In the case of an ACL rejection, they have all been read.)
3489 PH/69 The HTML version of the specification is now built in a directory called
3490 spec_html instead of spec.html, because the latter looks like a path with
3491 a MIME-type, and this confuses some software.
3493 PH/70 Catch two compiler warnings in sieve.c.
3495 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
3496 function verify_get_ident() calls ip_connect() to connect a socket, but
3497 if the "connect()" function timed out, ip_connect() used to close the
3498 socket. However, verify_get_ident() also closes the socket later, and in
3499 between Exim writes to the log, which may get opened at this point. When
3500 the socket was closed in ip_connect(), the log could get the same file
3501 descriptor number as the socket. This naturally causes chaos. The fix is
3502 not to close the socket in ip_connect(); the socket should be closed by
3503 the function that creates it. There was only one place in the code where
3504 this was missing, in the iplookup router, which I don't think anybody now
3505 uses, but I've fixed it anyway.
3507 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
3508 well as to direct DNS lookups. Otherwise the handling of names in host
3509 lists is inconsistent and therefore confusing.
3515 PH/01 Two changes to the default runtime configuration:
3517 (1) Move the checks for relay_from_hosts and authenticated clients from
3518 after to before the (commented out) DNS black list checks.
3520 (2) Add control=submission to the relay_from_hosts and authenticated
3521 clients checks, on the grounds that messages accepted by these
3522 statements are most likely to be submissions.
3524 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
3526 (1) Generate an error if the third argument for the ${prvs expansion is
3529 (2) Treat a missing third argument of ${prvscheck as if it were an empty
3532 (3) Reset the variables that are obtained from the first argument of
3533 ${prvscheck and used in the second argument before leaving the code,
3534 because their memory is reclaimed, so using them afterwards may do
3537 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
3538 one (it's much easier than Tom thought :-).
3540 (5) Because of (4), we can now allow for the use of $prvscheck_result
3541 inside the third argument.
3543 PH/03 For some reason, the default setting of PATH when running a command from
3544 a pipe transport was just "/usr/bin". I have changed it to
3547 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
3548 anything to be listed in the output from -bV.
3550 PH/05 When a filter generated an autoreply, the entire To: header line was
3551 quoted in the delivery log line, like this:
3553 => >A.N.Other <ano@some.domain> <original@ddress> ...
3555 This has been changed so that it extracts the operative address. There
3556 may be more than one such address. If so, they are comma-separated, like
3559 => >ano@some.domain,ona@other.domain <original@ddress> ...
3561 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
3562 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
3563 not being looked up in the rDNS to get a host name, Exim was showing the
3564 IP address twice in Received: lines, even though the IP addresses were
3565 identical. For example:
3567 Received: from [1.2.3.4] (helo=[1.2.3.4])
3569 However, if the real host name was known, it was omitting the HELO data
3570 if it matched the actual IP address. This has been tidied up so that it
3571 doesn't show the same IP address twice.
3573 PH/07 When both +timestamp and +memory debugging was on, the value given by
3574 $tod_xxx expansions could be wrong, because the tod_stamp() function was
3575 called by the debug printing, thereby overwriting the timestamp buffer.
3576 Debugging no longer uses the tod_stamp() function when +timestamp is set.
3578 PH/08 When the original message was included in an autoreply transport, it
3579 always said "this is a copy of the message, including all the headers",
3580 even if body_only or headers_only was set. It now gives an appropriate
3583 PH/09 Applied a patch from the Sieve maintainer which:
3585 o fixes some comments
3586 o adds the (disabled) notify extension core
3587 o adds some debug output for the result of if/elsif tests
3588 o points to the current vacation draft in the documentation
3589 and documents the missing references header update
3593 o fixes a bug in processing the envelope test (when testing
3594 multiple envelope elements, the last element determined the
3597 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
3598 Electronic Mail") by including:
3600 Auto-submitted: auto-generated
3602 in the messages that it generates (bounce messages and others, such as
3603 warnings). In the case of bounce messages for non-SMTP messages, there was
3604 also a typo: it was using "Auto_submitted" (underscore instead of
3605 hyphen). Since every message generated by Exim is necessarily in response
3606 to another message, thes have all been changed to:
3608 Auto-Submitted: auto-replied
3610 in accordance with these statements in the RFC:
3612 The auto-replied keyword:
3614 - SHOULD be used on messages sent in direct response to another
3615 message by an automatic process,
3617 - MUST NOT be used on manually-generated messages,
3619 - MAY be used on Delivery Status Notifications (DSNs) and Message
3620 Disposition Notifications (MDNs),
3622 - MUST NOT be used on messages generated by automatic or periodic
3623 processes, except for messages which are automatic responses to
3626 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
3627 to the default Received: header definition.
3629 PH/12 Added log selector acl_warn_skipped (default on).
3631 PH/13 After a successful wildlsearch lookup, discard the values of numeric
3632 variables because (a) they are in the wrong storage pool and (b) even if
3633 they were copied, it wouldn't work properly because of the caching.
3635 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
3636 checking when decoding. Apparently there are clients that generate
3637 overlong encoded strings. Why am I not surprised?
3639 PH/15 If the first argument of "${if match_address" was not empty, but did not
3640 contain an "@" character, Exim crashed. Now it writes a panic log message
3641 and treats the condition as false.
3643 PH/16 In autoreply, treat an empty string for "once" the same as unset.
3645 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
3646 extension "envelope-auth". The code is finished and in agreement with
3647 other implementations, but there is no documentation so far and in fact,
3648 nobody wrote the draft yet. This extension is currently #undef'ed, thus
3649 not changing the active code.
3651 Print executed "if" and "elsif" statements when debugging is used. This
3652 helps a great deal to understand what a filter does.
3654 Document more things not specified clearly in RFC3028. I had all this
3655 sorted out, when out of a sudden new issues came to my mind. Oops."
3657 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
3660 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
3661 canonical form (as documented). However, after a host name lookup from
3662 the IP address, check_host() was doing a simple string comparison with
3663 addresses acquired from the DNS when checking that the found name did
3664 have the original IP as one of its addresses. Since any found IPv6
3665 addresses are likely to be in abbreviated form, the comparison could
3666 fail. Luckily, there already exists a function for doing the comparison
3667 by converting both addresses to binary, so now that is used instead of
3668 the text comparison.
3670 PH/20 There was another similar case to PH/19, when a complete host name was
3671 given in a host list; looking up its IP address could give an abbreviated
3672 form, whereas the current host's name might or might not be abbreviated.
3673 The same fix has been applied.
3679 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
3680 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
3683 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
3684 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
3686 PH/03 Typo: missing ".o" in src/pcre/Makefile.
3688 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
3689 header line, restrict the check to what is listed in RFCs 2369 and 2929.
3690 Also, for "Auto-Submitted", treat anything other than "no" as
3691 non-personal, in accordance with RFC 3834. (Previously it treated
3692 anything starting "auto-" as non-personal.)
3694 TF/01 The control=submission/name=... option had a problem with syntax
3695 errors if the name included a slash character. The /name= option
3696 now slurps the rest of the string, so it can include any characters
3697 but it must come last in the list of options (after /sender_retain
3700 PH/05 Some modifications to the interface to the fake nameserver for the new
3708 TK/01 Added the "success_on_redirect" address verification option. See
3709 NewStuff for rationale and an example.
3711 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
3713 PH/02 Patch to exigrep to allow it to work on syslog lines.
3715 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
3716 fread() to skip over the body file's header line, because in Cygwin the
3717 header line is locked and is inaccessible.
3719 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
3720 co-exist for some time) to make it clear that it is the Exim ID that is
3721 referenced, not the Message-ID: header line.
3723 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
3724 string_format() function, because snprintf() does not exist on all
3727 PH/06 The use of forbid_filter_existstest now also locks out the use of the
3728 ${stat: expansion item.
3730 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
3731 protocol synchronization error", to keep the pedants happy.
3733 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
3734 well as for IRIX systems, when gcc is being used. See the host.c source
3737 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
3739 PH/10 Named domain lists were not working if used in a queue_smtp_domains
3742 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
3743 transport and to the smtp transport in LMTP mode.
3745 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
3747 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
3748 run a filter in a subprocess. This could lead to confusion in subsequent
3749 lookups in the parent process. There should also be a search_tidyup() at
3750 the end of the subprocess.
3752 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
3753 only if the host matched helo_try_verify_hosts, which caused the
3754 verification to occur when the EHLO/HELO command was issued. The ACL just
3755 tested the remembered result. Now, if a previous verification attempt has
3756 not happened, "verify = helo" does it there and then.
3758 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
3760 TK/03 Fix log output including CR from clamd.
3762 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
3763 exist provoked a memory error which could cause a segfault.
3765 PH/15 Installed PCRE 6.2
3767 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
3769 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
3770 of the problem. Specifically, suggested +O2 rather than +O1 for the
3773 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
3775 PH/20 If a delivery was routed to a non-standard port by means of an SRV
3776 record, the port was not correctly logged when the outgoing_port log
3777 selector was set (it logged the transort's default port).
3779 PH/21 Added support for host-specific ports to manualroute, queryprogram,
3780 fallback_hosts, and "hosts" in the smtp transport.
3782 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
3783 host errors such as "Connection refused".
3785 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
3786 authentication with radiusclient 0.4.9:
3788 - Error returned from rc_read_config was caught wrongly
3789 - Username/password not passed on to radius server due to wrong length.
3791 The presumption is that some radiusclient API changes for 4.51/PH/17
3792 were not taken care of correctly. The code is still untested by me (my
3793 Linux distribution still has 0.3.2 of radiusclient), but it was
3794 contributed by a Radius user.
3796 PH/24 When doing a callout, the value of $domain wasn't set correctly when
3797 expanding the "port" option of the smtp transport.
3799 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
3800 while reading a MIME header. Thanks to Tom Hughes for a patch.
3802 PH/24 Include config.h inside local_scan.h so that configuration settings are
3805 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
3806 that in an ACL for RCPT (for example), you can examine exactly what was
3809 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
3810 commands, but it was not correctly comparing the address with the actual
3811 client host address. Thus, it would show the EHLO address in Received:
3812 header lines when this was not necessary.
3814 PH/27 Added the % operator to ${eval:}.
3816 PH/28 Exim tries to create and chdir to its spool directory when it starts;
3817 it should be ignoring failures (because with -C, for example, it has lost
3818 privilege). It wasn't ignoring creation failures other than "already
3821 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
3824 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
3825 because an input line was too long, either on its own, or by virtue of
3826 too many continuations, the temporary file was not being removed, and the
3827 return code was incorrect.
3829 PH/31 Missing "BOOL" in function definition in filtertest.c.
3831 PH/32 Applied Sieve patches from the maintainer.
3833 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
3835 PH/33 Added "verify = not_blind".
3837 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
3838 Local/Makefile (with some defaults set). These are used in built scripts
3839 such as exicyclog, but they have never been used in the exim_install
3840 script (though there are many overriding facilities there). I have
3841 arranged that the exim_install script now takes note of these two
3844 PH/35 Installed configuration files for Dragonfly.
3846 PH/36 When a locally submitted message by a trusted user did not contain a
3847 From: header, and the sender address was obtained from -f or from an SMTP
3848 MAIL command, and the trusted user did not use -F to supply a sender
3849 name, $originator_name was incorrectly used when constructing a From:
3850 header. Furthermore, $originator_name was used for submission mode
3851 messages from external hosts without From: headers in a similar way,
3852 which is clearly wrong.
3854 PH/37 Added control=suppress_local_fixups.
3856 PH/38 When log_selector = +received_sender was set, and the addition of the
3857 sender made the log line's construction buffer exactly full, or one byte
3858 less than full, an overflow happened when the terminating "\n" was
3861 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
3862 when the result of a list match is failure because a DNS lookup failed.
3864 PH/40 RM_COMMAND is now used in the building process.
3866 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
3867 the "build-* directories that it finds.
3869 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
3870 address in a domain literal was a prefix of an interface address.
3872 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
3873 when verifying a sender address, unless rewrite_headers is false.
3875 PH/44 Wrote a long comment about why errors_to addresses are verified as
3876 recipients, not senders.
3878 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
3879 the ratelimit ACL was added.
3881 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
3883 PH/46 Added extra information about PostgreSQL errors to the error string.
3885 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
3886 avoiding the need to install special zones in a real server. This is
3887 backwards compatible; if it can't find the fake resolver, it drops back.
3888 Thus, both old and new test suites can be run.
3890 TF/02 Added util/ratelimit.pl
3892 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
3893 clock is set back in time.
3895 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
3896 Candler <B.Candler@pobox.com>.
3898 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
3899 OK for addresses that are the result of redirections.
3901 PH/48 A number of further additions for the benefit of the new test suite,
3902 including a fake gethostbyname() that interfaces to the fake DNS resolver
3905 TF/06 The fix for widen_domains has also been applied to qualify_single and
3906 search_parents which are the other dnslookup options that can cause
3909 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
3912 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
3914 TF/07 Exim produced the error message "an SRV record indicated no SMTP
3915 service" if it encountered an MX record with an empty target hostname.
3916 The message is now "an MX or SRV record indicated no SMTP service".
3918 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
3919 if the DNS of the sending site is misconfigured. This is quite a
3920 common situation. This change restores the behaviour of treating a
3921 helo verification defer as a failure.
3923 PH/51 If self=fail was set on a router, the bounce message did not include the
3924 actual error message.
3930 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
3932 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
3933 command itself ended in error, the underlying message about the transport
3934 filter timeout was being overwritten with the pipe command error. Now the
3935 underlying error message should be appended to the second error message.
3937 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
3939 PH/02 Reduce the amount of output that "make" produces by default. Full output
3940 can still be requested.
3942 PH/03 The warning log line about a condition test deferring for a "warn" verb
3943 was being output only once per connection, rather than after each
3944 occurrence (because it was using the same function as for successful
3945 "warn" verbs). This seems wrong, so I have changed it.
3947 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
3948 it should not have, which might have caused a crash in the right
3949 circumstances, but probably never did.
3951 PH/04 Installed a modified version of Tony Finch's patch to make submission
3952 mode fix the return path as well as the Sender: header line, and to
3953 add a /name= option so that you can make the user's friendly name appear
3956 TF/03 Added the control = fakedefer ACL modifier.
3958 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
3959 Mark Lowes for thorough testing.
3961 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
3963 TK/03 Merged latest SRS patch from Miles Wilton.
3965 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
3966 with the definition in sysexits.h (which is #included earlier).
3967 Fortunately, Exim does not actually use EX_OK. The code used to try to
3968 preserve the sysexits.h value, by assuming that macro definitions were
3969 scanned for macro replacements. I have been disabused of this notion,
3970 so now the code just undefines EX_OK before #including unistd.h.
3972 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
3973 in the smtp transport. When a block could not be written in a single
3974 write() function, the timeout was being re-applied to each part-write.
3975 This seems wrong - if the receiver was accepting one byte at a time it
3976 would take for ever. The timeout is now adjusted when this happens. It
3977 doesn't have to be particularly precise.
3979 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
3980 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
3982 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
3983 without a domain if the check to <postmaster@domain> fails.
3985 SC/01 Eximstats: added -xls and the ability to specify output files
3986 (patch written by Frank Heydlauf).
3988 SC/02 Eximstats: use FileHandles for outputting results.
3990 SC/03 Eximstats: allow any combination of xls, txt, and html output.
3992 SC/04 Eximstats: fixed display of large numbers with -nvr option
3994 SC/05 Eximstats: fixed merging of reports with empty tables.
3996 SC/06 Eximstats: added the -include_original_destination flag
3998 SC/07 Eximstats: removed tabs and trailing whitespace.
4000 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
4002 TK/06 MBOX spool code: Add real "From " MBOX separator line
4003 so the .eml file is really in mbox format (even though
4004 most programs do not really care). Patch from Alex Miller.
4006 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
4007 The latter is generated from $received_to and is only set if the
4008 message has one envelope recipient. SA can use these headers,
4009 obviously out-of-the-box. Patch from Alex Miller.
4011 PH/08 The ${def test on a variable was returning false if the variable's
4012 value was "0", contrary to what the specification has always said!
4013 The result should be true unless the variable is empty.
4015 PH/09 The syntax error of a character other than { following "${if
4016 def:variable_name" (after optional whitespace) was not being diagnosed.
4017 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
4018 accidental colon was present, for example, could give incorrect results.
4020 PH/10 Tidied the code in a number of places where the st_size field of a stat()
4021 result is used (not including appendfile, where other changes are about
4024 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
4025 This involved changing a lot of size variables from int to off_t. It
4026 should work with maildirs and everything.
4028 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
4029 spamd dying while we are connected to it.
4031 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
4034 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
4035 function for BDB 4.3.
4037 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
4039 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
4040 the sources for PCRE have been re-arranged and more files are now
4043 PH/15 The code I had for printing potentially long long variables in PH/11
4044 above was not the best (it lost precision). The length of off_t variables
4045 is now inspected at build time, and an appropriate printing format (%ld
4046 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
4047 to be "long long int" or "long int". This is needed for the internal
4048 formatting function string_vformat().
4050 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
4051 the configuration file to be ":syslog", then the script "guesses" where
4052 the logs files are, rather than using the compiled in default. In our
4053 case the guess is not the same as the compiled default, so the script
4054 suddenly stopped working when I started to use syslog. The patch checks
4055 to see if log_file_path is "". If so, it attempts to read it from exim
4056 with no configuration file to get the compiled in version, before it
4057 falls back to the previous guessing code."
4059 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
4060 implementing BATV in an Exim configuration. See NewStuff for the gory
4063 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
4064 Makefile that are specific to HP-UX.
4066 PH/18 If the "use_postmaster" option was set for a recipient callout together
4067 with the "random" option, the postmaster address was used as the MAIL
4068 FROM address for the random test, but not for the subsequent recipient
4069 test. It is now used for both.
4071 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
4072 patch removes a few documentation additions to RFC 3028, because the
4073 latest draft now contains them. It adds the new en;ascii-case comparator
4074 and a new error check for 8bit text in MIME parts. Comparator and
4075 require names are now matched exactly. I enabled the subaddress
4076 extension, but it is not well tested yet (read: it works for me)."
4078 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
4079 rework some of the code of TK/09 above to avoid the hardwired use of
4080 "%lld" and "long long". Replaced the call to snprintf() with a call to
4083 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
4084 records point to non-existent hosts", "retry timeout exceeded", and
4085 "retry time not reached for any host after a long failure period".
4087 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
4088 experimental DomainKeys support:
4090 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
4091 (2) On an error such as an illegally used "control", the wrong name for
4092 the control was given.
4094 These problems did NOT occur unless DomainKeys support was compiled.
4096 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
4098 PH/24 Added ${if match_ip condition.
4100 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
4101 databases so that it will be absolutely obvious if a crash occurs in the
4102 DB library. This is a regular occurrence (often caused by mis-matched
4105 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
4106 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
4107 picked up on a user's system that detects such things. There doesn't seem
4108 to be a gcc warning option for this - only an attribute that has to be
4109 put on the function's prototype. It seems that in Fedora Core 4 they have
4110 set this on a number of new functions. No doubt there will be more in due
4113 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
4114 specify a transport. However, if an address that was verified by such a
4115 router was the subject of a callout, Exim crashed because it tried to
4116 read the rcpt_include_affixes from the non-existent transport. Now it
4117 just assumes that the setting of that option is false. This bug was
4118 introduced by 4.51/PH/31.
4120 PH/28 Changed -d+all to exclude +memory, because that information is very
4121 rarely of interest, but it makes the output a lot bigger. People tend to
4122 do -d+all out of habit.
4124 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
4125 code in os-type was giving problems when libc.so lives in lib64, like on
4128 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
4129 aren't the modern standard, and it seems that some systems' include files
4130 don't always have them. Exim was already checking for some of the newer
4131 ones like T_AAAA, and defining it itself. I've added checks for all the
4132 record types that Exim uses.
4134 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
4135 not automatically generating a new one, as it is supposed to. This
4136 prevented TLS from working. If the file did exist, but contained invalid
4137 data, a new version was generated, as expected. It was only the case of a
4138 non-existent file that was broken.
4140 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
4141 with a change in libdomainkeys > 0.64.
4143 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
4144 from DNS. If the selector record carries the flag, it now has
4145 precedence over the domain-wide flag.
4147 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
4149 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
4150 the use of a port name that isn't defined in /etc/services) occurred, the
4151 message was deferred as in a normal delivery, and thus remained on the
4152 spool, instead of being failed because of the mua_wrapper setting. This
4153 is now fixed, and I tidied up some of the mua_wrapper messages at the
4156 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
4157 the messages in a hash of arrays rather than using individual hashes.
4158 This is a bit cleaner and results in dramatic memory savings, albeit
4159 at a slight CPU cost.
4161 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
4162 as requested by Marc Sherman.
4164 SC/10 Eximstats: added histograms for user specified patterns as requested
4167 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
4169 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
4170 fopen() in the content-scanning modules that did not already have it.
4176 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
4177 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
4179 TK/02 Fix ACL "control" statement not being available in MIME ACL.
4181 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
4183 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
4184 to test Sieve filters that use "vacation".
4186 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
4187 that changes the way the GnuTLS parameters are stored in the cache file.
4188 The new format can be generated externally. For backward compatibility,
4189 if the data in the cache doesn't make sense, Exim assumes it has read an
4190 old-format file, and it generates new data and writes a new file. This
4191 means that you can't go back to an older release without removing the
4194 PH/03 A redirect router that has both "unseen" and "one_time" set does not
4195 work if there are any delivery delays because "one_time" forces the
4196 parent to be marked "delivered", so its unseen clone is never tried
4197 again. For this reason, Exim now forbids the simultaneous setting of
4200 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
4201 redirected to themselves ("homonym" addresses). Read the long ChangeLog
4202 entry if you want to know the details. The fix, however, neglected to
4203 consider the case when local delivery batching is involved. The test for
4204 "previously delivered" was not happening when checking to see if an
4205 address could be batched with a previous (undelivered) one; under
4206 certain circumstances this could lead to multiple deliveries to the same
4209 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
4210 in its include files, and this causes problems building Exim.
4212 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
4213 header_syntax) but Exim was just ignoring anything given after a slash.
4214 In particular, this caused confusion with an attempt to use "verify =
4215 reverse_host_lookup/defer_ok". An error is now given when options are
4216 supplied for verify items that do not have them. (Maybe reverse_host_
4217 lookup should have a defer_ok option, but that's a different point.)
4219 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
4220 defined by RFC 821) to 2048, because there were problems with some AUTH
4221 commands, and RFC 1869 says the size should be increased for extended
4222 SMTP commands that take arguments.
4224 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
4227 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
4228 "unknown" error; now it says that the functionality isn't in the binary.
4230 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
4231 an address' error message when a string expansion fails (syntax or
4232 whatever). Otherwise the password may appear in the log. Following change
4233 PH/42 below, there is no longer a chance of it appearing in a bounce
4236 PH/11 Installed exipick version 20050225.0 from John Jetmore.
4238 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
4239 of its addresses was ever tried. (Bugzilla bug #2.)
4241 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
4242 the result incorrectly in the debug output. (It correctly added a newline
4243 to what was transported.)
4245 TF/01 Added $received_time.
4247 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
4248 commented out examples of how to interface to a virus scanner and to
4249 SpamAssassin. Also added commented examples of av_scanner and
4250 spamd_address settings.
4252 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
4253 and controls are allowed in which ACLs. There were a couple of minor
4254 errors. Some of the entries in the conditions table (which is a table of
4255 where they are NOT allowed) were getting very unwieldy; rewrote them as a
4256 negation of where the condition IS allowed.
4258 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
4260 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
4261 header file does not have a version number, so I've had to invent a new
4262 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
4263 API. The code is untested by me (my Linux distribution still has 0.3.2 of
4264 radiusclient), but it was contributed by a Radius user.
4266 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
4267 files or directories, for OpenSSL.
4269 PH/19 When an Exim process that is running as root has to create an Exim log
4270 file, it does so in a subprocess that runs as exim:exim so as to get the
4271 ownership right at creation (otherwise, other Exim processes might see
4272 the file with the wrong ownership). There was no test for failure of this
4273 fork() call, which would lead to the process getting stuck as it waited
4274 for a non-existent subprocess. Forks do occasionally fail when resources
4275 run out. I reviewed all the other calls to fork(); they all seem to check
4278 PH/20 When checking for unexpected SMTP input at connect time (before writing
4279 the banner), Exim was not dealing correctly with a non-positive return
4280 from the read() function. If the client had disconnected by this time,
4281 the result was a log entry for a synchronization error with an empty
4282 string after "input=" when read() returned zero. If read() returned -1
4283 (an event I could not check), uninitialized data bytes were printed.
4284 There were reports of junk text (parts of files, etc) appearing after
4287 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
4289 PH/22 Added support for macro redefinition, and (re)definition in between
4290 driver and ACL definitions.
4292 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
4293 forgetting to use the resulting value; it was using the unexpanded value.
4295 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
4296 hadn't been configured. The fix is from Juergen Kreileder, who
4297 understands it better than I do:
4299 "Here's what I see happening with three configured cyrus_sasl
4300 authenticators configured (plain, login, cram-md5):
4302 On startup auth_cyrus_sasl_init() gets called for each of these.
4303 This means three calls to sasl_listmech() without a specified mech_list.
4304 => SASL tests which mechs of all available mechs actually work
4305 => three warnings about OTP not working
4306 => the returned list contains: plain, login, cram-md5, digest-md5, ...
4308 With the patch, sasl_listmech() also gets called three times. But now
4309 SASL's mech_list option is set to the server_mech specified in the the
4310 authenticator. Or in other words, the answer from sasl_listmech()
4311 gets limited to just the mech you're testing for (which is different
4313 => the return list contains just 'plain' or 'login', 'cram-md5' or
4314 nothing depending on the value of ob->server_mech.
4316 I've just tested the patch: Authentication still works fine,
4317 unavailable mechs specified in the exim configuration are still
4318 caught, and the auth.log warnings about OTP are gone."
4320 PH/25 When debugging is enabled, the contents of the command line are added
4321 to the debugging output, even when log_selector=+arguments is not
4324 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
4325 answer is "GNU", and only if the return is "GNU/something" is the answer
4328 PH/27 $acl_verify_message is now set immediately after the failure of a
4329 verification in an ACL, and so is available in subsequent modifiers. In
4330 particular, the message can be preserved by coding like this:
4332 warn !verify = sender
4333 set acl_m0 = $acl_verify_message
4335 Previously, $acl_verify_message was set only while expanding "message"
4336 and "log_message" when a very denied access.
4338 PH/28 Modified OS/os.c-Linux with
4340 -#ifndef OS_LOAD_AVERAGE
4341 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
4343 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
4344 nomenclature these days.)
4346 PH/29 Installed patch from the Sieve maintainer that adds the options
4347 sieve_useraddress and sieve_subaddress to the redirect router.
4349 PH/30 In these circumstances:
4350 . Two addresses routed to the same list of hosts;
4351 . First host does not offer TLS;
4352 . First host accepts first address;
4353 . First host gives temporary error to second address;
4354 . Second host offers TLS and a TLS session is established;
4355 . Second host accepts second address.
4356 Exim incorrectly logged both deliveries with the TLS parameters (cipher
4357 and peerdn, if requested) that were in fact used only for the second
4360 PH/31 When doing a callout as part of verifying an address, Exim was not paying
4361 attention to any local part prefix or suffix that was matched by the
4362 router that accepted the address. It now behaves in the same way as it
4363 does for delivery: the affixes are removed from the local part unless
4364 rcpt_include_affixes is set on the transport.
4366 PH/32 Add the sender address, as F=<...>, to the log line when logging a
4367 timeout during the DATA phase of an incoming message.
4369 PH/33 Sieve envelope tests were broken for match types other than :is. I have
4370 applied a patch sanctioned by the Sieve maintainer.
4372 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
4373 the uid or gid is negative. A case of a negative gid caused this to be
4374 noticed. The fix allows for either to be negative.
4376 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
4377 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
4380 PH/36 The change PH/12 above was broken. Fixed it.
4382 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
4383 the grounds that routing the same address twice would always produce the
4384 same answer. This might have been true once, but it is certainly no
4385 longer true now. Routing a child address may depend on the previous
4386 routing that produced that child. Some complicated redirection strategies
4387 went wrong when messages had multiple recipients, and made Exim's
4388 behaviour dependent on the order in which the addresses were given.
4390 I have moved the duplicate checking until after the routing is complete.
4391 Exim scans the addresses that are assigned to local and remote
4392 transports, and removes any duplicates. This means that more work will be
4393 done, as duplicates will always all be routed, but duplicates are
4394 presumably rare, so I don't expect this is of any significance.
4396 For deliveries to pipes, files, and autoreplies, the duplicate checking
4397 still happens during the routing process, since they are not going to be
4400 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
4401 It corrects a timeout issue with spamd. This is Ian's comment: "The
4402 background is that sometimes spamd either never reads data from a
4403 connection it has accepted, or it never writes response data. The exiscan
4404 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
4405 blindly assumes that writes won't block so it may never time out."
4407 PH/39 Allow G after quota size as well as K and M.
4409 PH/40 The value set for $authenticated_id in an authenticator may not contain
4410 binary zeroes or newlines because the value is written to log lines and
4411 to spool files. There was no check on this. Now the value is run through
4412 the string_printing() function so that such characters are converted to
4413 printable escape sequences.
4415 PH/41 $message_linecount is a new variable that contains the total number of
4416 lines in the message. Compare $body_linecount, which is the count for the
4419 PH/42 Exim no longer gives details of delivery errors for specific addresses in
4420 bounce and delay warning messages, except in certain special cases, which
4423 (a) An SMTP error message from a remote host;
4424 (b) A message specified in a :fail: redirection;
4425 (c) A message specified in a "fail" command in a system filter;
4426 (d) A message specified in a FAIL return from the queryprogram router;
4427 (e) A message specified by the cannot_route_message router option.
4429 In these cases only, Exim does include the error details in bounce and
4430 warning messages. There are also a few cases where bland messages such
4431 as "unrouteable address" or "local delivery error" are given.
4433 PH/43 $value is now also set for the "else" part of a ${run expansion.
4435 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
4436 being worked on, but at least Exim now implements the latest version to
4439 PH/45 In a pipe transport, although a timeout while waiting for the pipe
4440 process to complete was treated as a delivery failure, a timeout while
4441 writing the message to the pipe was logged, but erroneously treated as a
4442 successful delivery. Such timeouts include transport filter timeouts. For
4443 consistency with the overall process timeout, these timeouts are now
4444 treated as errors, giving rise to delivery failures by default. However,
4445 there is now a new Boolean option for the pipe transport called
4446 timeout_defer, which, if set TRUE, converts the failures into defers for
4447 both kinds of timeout. A transport filter timeout is now identified in
4450 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
4451 systems where "make" and "gmake" are different, calling "gmake" at top
4452 level broke things. I've arranged for the value of $(MAKE) to be passed
4453 from the Makefile to this script so that it can call the same version of
4457 A note about Exim versions 4.44 and 4.50
4458 ----------------------------------------
4460 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
4461 changes of various kinds. As a consequence, a big documentation update was
4462 needed. This delayed the release for rather longer than seemed good, especially
4463 in the light of a couple of (minor) security issues. Therefore, the changes
4464 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
4465 release. So 4.44 and 4.50 are in effect two different branches that both start
4468 I have left the 4.50 change log unchanged; it contains all the changes since
4469 4.43. The change log for 4.44 is below; many of its items are identical to
4470 those for 4.50. This seems to be the most sensible way to preserve the
4471 historical information.
4477 1. Minor wording change to the doc/README.SIEVE file.
4479 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
4480 computation of the current number of files was incorrect.
4482 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4483 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4486 4. Give more explanation in the error message when the command for a transport
4487 filter fails to execute.
4489 5. There are several places where Exim runs a non-Exim command in a
4490 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4491 was being done only for the command run by the queryprogram router. It is
4492 now done for all such subprocesses. The other cases are: ${run, transport
4493 filters, and the commands run by the lmtp and pipe transports.
4495 6. Added CONFIGURE_GROUP build-time option.
4497 7. Some older OS have a limit of 256 on the maximum number of file
4498 descriptors. Exim was using setrlimit() to set 1000 as a large value
4499 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4500 systems. I've change it so that if it can't get 1000, it tries for 256.
4502 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4503 was an oversight, and furthermore, ever since the addition of extra
4504 controls (e.g. 4.43/32), the checks on when to allow different forms of
4505 "control" were broken. There should now be diagnostics for all cases when a
4506 control that does not make sense is encountered.
4508 9. Added the /retain_sender option to "control=submission".
4510 10. $recipients is now available in the predata ACL (oversight).
4512 11. Tidy the search cache before the fork to do a delivery from a message
4513 received from the command line. Otherwise the child will trigger a lookup
4514 failure and thereby defer the delivery if it tries to use (for example) a
4515 cached ldap connection that the parent has called unbind on.
4517 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4518 of $address_data from the recipient verification was clobbered by the
4519 sender verification.
4521 13. The value of address_data from a sender verification is now available in
4522 $sender_address_data in subsequent conditions in the ACL statement.
4524 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
4526 15. Added a new option "connect=<time>" to callout options, to set a different
4529 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4530 was its contents. (It was OK if the option was not defined at all.)
4532 17. A "Completed" log line is now written for messages that are removed from
4533 the spool by the -Mrm option.
4535 18. New variables $sender_verify_failure and $recipient_verify_failure contain
4536 information about exactly what failed.
4538 19. Added -dd to debug only the daemon process.
4540 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
4541 handles timeouts, both on the server side and network timeouts. Renamed the
4542 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
4544 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
4545 It is now set to "smtps".
4547 22. $host_address is now set to the target address during the checking of
4548 ignore_target_hosts.
4550 23. When checking ignore_target_hosts for an ipliteral router, no host name was
4551 being passed; this would have caused $sender_host_name to have been used if
4552 matching the list had actually called for a host name (not very likely,
4553 since this list is usually IP addresses). A host name is now passed as
4556 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4557 code that specifies a non-restarting handler (typically sigaction() in
4558 modern systems) in an attempt to fix a rare and obscure crash bug.
4560 25. Narrowed the window for a race in the daemon that could cause it to ignore
4561 SIGCHLD signals. This is not a major problem, because they are used only to
4562 wake it up if nothing else does.
4564 26. A malformed maildirsize file could cause Exim to calculate negative values
4565 for the mailbox size or file count. Odd effects could occur as a result.
4566 The maildirsize information is now recalculated if the size or filecount
4569 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4570 support for a long time. Removed HAVE_SYS_VFS_H.
4572 28. Installed the latest version of exipick from John Jetmore.
4574 29. In an address list, if the pattern was not a regular expression, an empty
4575 subject address (from a bounce message) matched only if the pattern was an
4576 empty string. Non-empty patterns were not even tested. This was the wrong
4577 because it is perfectly reasonable to use an empty address as part of a
4578 database query. An empty address is now tested by patterns that are
4579 lookups. However, all the other forms of pattern expect the subject to
4580 contain a local part and a domain, and therefore, for them, an empty
4581 address still always fails if the pattern is not itself empty.
4583 30. Exim went into a mad DNS loop when attempting to do a callout where the
4584 host was specified on an smtp transport, and looking it up yielded more
4585 than one IP address.
4587 31. Re-factored the code for checking spool and log partition space into a
4588 function that finds that data and another that does the check. The former
4589 is then used to implement four new variables: $spool_space, $log_space,
4590 $spool_inodes, and $log_inodes.
4592 32. The RFC2047 encoding function was originally intended for short strings
4593 such as real names; it was not keeping to the 75-character limit for
4594 encoded words that the RFC imposes. It now respects the limit, and
4595 generates multiple encoded words if necessary. To be on the safe side, I
4596 have increased the buffer size for the ${rfc2047: expansion operator from
4599 33. It is now permitted to omit both strings after an "if" condition; if the
4600 condition is true, the result is "true". As before, when the second string
4601 is omitted, a false condition yields an empty string. This makes it less
4602 cumbersome to write custom ACL and router conditions.
4604 34. Failure to deliver a bounce message always caused it to be frozen, even if
4605 there was an errors_to setting on the router. The errors_to setting is now
4608 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4609 canonical form (fully expanded) before being placed in
4610 $sender_host_address.
4612 36. The table in the code that translates DNS record types into text (T_A to
4613 "A" for instance) was missing entries for NS and CNAME. It is just possible
4614 that this could have caused confusion if both these types were looked up
4615 for the same domain, because the text type is used as part of Exim's
4616 per-process caching. But the chance of anyone hitting this buglet seems
4619 37. The dnsdb lookup has been extended in a number of ways.
4621 (1) There is a new type, "zns", which walks up the domain tree until it
4622 finds some nameserver records. It should be used with care.
4624 (2) There is a new type, "mxh", which is like "mx" except that it returns
4625 just the host names, not the priorities.
4627 (3) It is now possible to give a list of domains (or IP addresses) to be
4628 looked up. The behaviour when one of the lookups defers can be
4629 controlled by a keyword.
4631 (4) It is now possible to specify the separator character for use when
4632 multiple records are returned.
4634 38. The dnslists ACL condition has been extended: it is now possible to supply
4635 a list of IP addresses and/or domains to be looked up in a particular DNS
4638 39. Added log_selector=+queue_time_overall.
4640 40. When running the queue in the test harness, wait just a tad after forking a
4641 delivery process, to get repeatability of debugging output.
4643 41. Include certificate and key file names in error message when GnuTLS fails
4644 to set them up, because the GnuTLS error message doesn't include the name
4645 of the failing file when there is a problem reading it.
4647 42. Allow both -bf and -bF in the same test run.
4649 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
4651 44. The "Exiscan patch" is now merged into the mainline Exim source.
4653 45. Sometimes the final signoff response after QUIT could fail to get
4654 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4655 < 0 before doing a fflush(). This bug looks as though it goes back to the
4656 introduction of TLS in release 3.20, but "sometimes" must have been rare
4657 because the tests only now provoked it.
4659 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
4660 (this can affect the format of dates).
4662 47. exim_tidydb, when checking for the continued existence of a message for
4663 which it has found a message-specific retry record, was not finding
4664 messages that were in split spool directories. Consequently, it was
4665 deleting retry records that should have stayed in existence.
4667 48. Steve fixed some bugs in eximstats.
4669 49. The SPA authentication driver was not abandoning authentication and moving
4670 on to the next authenticator when an expansion was forced to fail,
4671 contradicting the general specification for all authenticators. Instead it
4672 was generating a temporary error. It now behaves as specified.
4674 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
4675 (the order specifies the preference for clients). The order is now AES256,
4676 AES128, 3DES, ARCFOUR128.
4678 51. Small patch to Sieve code - explicitly set From: when generating an
4681 52. Exim crashed if a remote delivery caused a very long error message to be
4682 recorded - for instance if somebody sent an entire SpamAssassin report back
4683 as a large number of 550 error lines. This bug was coincidentally fixed by
4684 increasing the size of one of Exim's internal buffers (big_buffer) that
4685 happened as part of the Exiscan merge. However, to be on the safe side, I
4686 have made the code more robust (and fixed the comments that describe what
4689 53. Now that there can be additional text after "Completed" in log lines (if
4690 the queue_time_overall log selector is set), a one-byte patch to exigrep
4691 was needed to allow it to recognize "Completed" as not the last thing in
4694 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
4695 patch that reportedly fixes this has been added. I am not expert enough to
4696 create a test for it. This is what the patch creator wrote:
4698 "I found a little strange behaviour of ldap code when working with
4699 Windows 2003 AD Domain, where users was placed in more than one
4700 Organization Units. When I tried to give exim partial DN, the exit code
4701 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
4702 But simultaneously result of request was absolutely normal ldap result,
4703 so I produce this patch..."
4705 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
4706 REFERENCE, so I have modified the code to exclude the patch when that macro
4709 55. Some experimental protocols are using DNS PTR records for new purposes. The
4710 keys for these records are domain names, not reversed IP addresses. The
4711 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4712 leaves it alone. Component reversal etc. now happens only for IP addresses.
4715 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
4717 57. Double the size of the debug message buffer (to 2048) so that more of very
4718 long debug lines gets shown.
4720 58. The exicyclog utility now does better if the number of log files to keep
4721 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
4723 59. Two changes related to the smtp_active_hostname option:
4725 (1) $smtp_active_hostname is now available as a variable.
4726 (2) The default for smtp_banner uses $smtp_active_hostname instead
4727 of $primary_hostname.
4729 60. The host_aton() function is supposed to be passed a string that is known
4730 to be a valid IP address. However, in the case of IPv6 addresses, it was
4731 not checking this. This is a hostage to fortune. Exim now panics and dies
4732 if the condition is not met. A case was found where this could be provoked
4733 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4734 components; fortuitously, this particular loophole had already been fixed
4735 by change 4.50/55 above.
4737 If there are any other similar loopholes, the new check in host_aton()
4738 itself should stop them being exploited. The report I received stated that
4739 data on the command line could provoke the exploit when Exim was running as
4740 exim, but did not say which command line option was involved. All I could
4741 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4742 running as the user.
4745 61. There was a buffer overflow vulnerability in the SPA authentication code
4746 (which came originally from the Samba project). I have added a test to the
4747 spa_base64_to_bits() function which I hope fixes it.
4750 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
4751 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
4753 63. The daemon start-up calls getloadavg() while still root for those OS that
4754 need the first call to be done as root, but it missed one case: when
4755 deliver_queue_load_max is set with deliver_drop_privilege. This is
4756 necessary for the benefit of the queue runner, because there is no re-exec
4757 when deliver_drop_privilege is set.
4759 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
4760 This has been fixed.
4762 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4763 was in use, was not putting the data itself into the right store pool;
4764 consequently, it could be overwritten for a subsequent message in the same
4765 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4768 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
4770 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
4771 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
4772 the function were treating the return as a boolean value, which happened to
4773 work because 0=false and not-0=true, but is not correct code.
4775 68. The host_aton() function was not handling scoped IPv6 addresses (those
4776 with, for example, "%eth0" on the end) correctly.
4778 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
4779 negated items (that is, ~something) in unsigned ints. Some compilers
4780 apparently mutter when there is no cast.
4782 70. If an address verification called from an ACL failed, and did not produce a
4783 user-specific message (i.e. there was only a "system" message), nothing was
4784 put in $acl_verify_message. In this situation, it now puts the system
4787 71. Change 4.23/11 added synchronization checking at the start of an SMTP
4788 session; change 4.31/43 added the unwanted input to the log line - except
4789 that it did not do this in the start of session case. It now does.
4791 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
4792 This is wrong and can cause the other end to generate a synchronization
4793 error if it is another Exim or anything else that does the synchronization
4794 check. A QUIT command is no longer sent after a timeout.
4796 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
4797 during host lookups.
4799 74. The defer_ok option of callout verification was not working if it was used
4800 when verifying addresses in header lines, that is, for this case:
4802 verify = header_sender/callout=defer_ok
4804 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
4805 those file descriptors could be used for SMTP connections. If anything
4806 wrote to stderr (the example that came up was "warn" in embedded Perl), it
4807 could be sent to the SMTP client, causing chaos. The daemon now opens
4808 stdin, stdout, and stderr to /dev/null when it puts itself into the
4811 76. Arrange for output from Perl's "warn" command to be written to Exim's main
4812 log by default. The user can override this with suitable Perl magic.
4814 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
4815 the log message when discard triggers, was not working for the DATA ACL or
4816 for the non-SMTP ACL.
4818 78. Error message wording change in sieve.c.
4820 79. If smtp_accept_max_per_host was set, the number of connections could be
4821 restricted to fewer than expected, because the daemon was trying to set up
4822 a new connection before checking whether the processes handling previous
4823 connections had finished. The check for completed processes is now done
4824 earlier. On busy systems, this bug wouldn't be noticed because something
4825 else would have woken the daemon, and it would have reaped the completed
4828 80. If a message was submitted locally by a user whose login name contained one
4829 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
4830 It caused a spool format error. I have fixed the spool reading code. A
4831 related problem was that the "from" clause in the Received: line became
4832 illegal because of the space(s). It is now covered by ${quote_local_part.
4834 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
4837 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
4839 83. Patch from Sieve maintainer to fix unterminated string problem in
4840 "vacation" handling.
4842 84. Some minor changes to the Linux configuration files to help with other
4843 OS variants using glibc.
4845 85. One more patch for Sieve to update vacation handling to latest spec.
4848 ----------------------------------------------------
4849 See the note above about the 4.44 and 4.50 releases.
4850 ----------------------------------------------------
4856 1. Change 4.43/35 introduced a bug that caused file counts to be
4857 incorrectly computed when quota_filecount was set in an appendfile
4860 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4861 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4864 3. Give more explanation in the error message when the command for a transport
4865 filter fails to execute.
4867 4. There are several places where Exim runs a non-Exim command in a
4868 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4869 was being done only for the command run by the queryprogram router. It is
4870 now done for all such subprocesses. The other cases are: ${run, transport
4871 filters, and the commands run by the lmtp and pipe transports.
4873 5. Some older OS have a limit of 256 on the maximum number of file
4874 descriptors. Exim was using setrlimit() to set 1000 as a large value
4875 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4876 systems. I've change it so that if it can't get 1000, it tries for 256.
4878 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4879 was an oversight, and furthermore, ever since the addition of extra
4880 controls (e.g. 4.43/32), the checks on when to allow different forms of
4881 "control" were broken. There should now be diagnostics for all cases when a
4882 control that does not make sense is encountered.
4884 7. $recipients is now available in the predata ACL (oversight).
4886 8. Tidy the search cache before the fork to do a delivery from a message
4887 received from the command line. Otherwise the child will trigger a lookup
4888 failure and thereby defer the delivery if it tries to use (for example) a
4889 cached ldap connection that the parent has called unbind on.
4891 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4892 of $address_data from the recipient verification was clobbered by the
4893 sender verification.
4895 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4896 was its contents. (It was OK if the option was not defined at all.)
4898 11. A "Completed" log line is now written for messages that are removed from
4899 the spool by the -Mrm option.
4901 12. $host_address is now set to the target address during the checking of
4902 ignore_target_hosts.
4904 13. When checking ignore_target_hosts for an ipliteral router, no host name was
4905 being passed; this would have caused $sender_host_name to have been used if
4906 matching the list had actually called for a host name (not very likely,
4907 since this list is usually IP addresses). A host name is now passed as
4910 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4911 code that specifies a non-restarting handler (typically sigaction() in
4912 modern systems) in an attempt to fix a rare and obscure crash bug.
4914 15. Narrowed the window for a race in the daemon that could cause it to ignore
4915 SIGCHLD signals. This is not a major problem, because they are used only to
4916 wake it up if nothing else does.
4918 16. A malformed maildirsize file could cause Exim to calculate negative values
4919 for the mailbox size or file count. Odd effects could occur as a result.
4920 The maildirsize information is now recalculated if the size or filecount
4923 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4924 support for a long time. Removed HAVE_SYS_VFS_H.
4926 18. Updated exipick to current release from John Jetmore.
4928 19. Allow an empty sender to be matched against a lookup in an address list.
4929 Previously the only cases considered were a regular expression, or an
4932 20. Exim went into a mad DNS lookup loop when doing a callout where the
4933 host was specified on the transport, if the DNS lookup yielded more than
4936 21. The RFC2047 encoding function was originally intended for short strings
4937 such as real names; it was not keeping to the 75-character limit for
4938 encoded words that the RFC imposes. It now respects the limit, and
4939 generates multiple encoded words if necessary. To be on the safe side, I
4940 have increased the buffer size for the ${rfc2047: expansion operator from
4943 22. Failure to deliver a bounce message always caused it to be frozen, even if
4944 there was an errors_to setting on the router. The errors_to setting is now
4947 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4948 canonical form (fully expanded) before being placed in
4949 $sender_host_address.
4951 24. Updated eximstats to version 1.33
4953 25. Include certificate and key file names in error message when GnuTLS fails
4954 to set them up, because the GnuTLS error message doesn't include the name
4955 of the failing file when there is a problem reading it.
4957 26. Expand error message when OpenSSL has problems setting up cert/key files.
4960 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
4961 (this can affect the format of dates).
4963 28. exim_tidydb, when checking for the continued existence of a message for
4964 which it has found a message-specific retry record, was not finding
4965 messages that were in split spool directories. Consequently, it was
4966 deleting retry records that should have stayed in existence.
4968 29. eximstats updated to version 1.35
4969 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
4970 1.35 - bugfix such that pie charts by volume are generated correctly
4972 30. The SPA authentication driver was not abandoning authentication and moving
4973 on to the next authenticator when an expansion was forced to fail,
4974 contradicting the general specification for all authenticators. Instead it
4975 was generating a temporary error. It now behaves as specified.
4977 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
4978 (the order specifies the preference for clients). The order is now AES256,
4979 AES128, 3DES, ARCFOUR128.
4981 31. Small patch to Sieve code - explicitly set From: when generating an
4984 32. Exim crashed if a remote delivery caused a very long error message to be
4985 recorded - for instance if somebody sent an entire SpamAssassin report back
4986 as a large number of 550 error lines. This bug was coincidentally fixed by
4987 increasing the size of one of Exim's internal buffers (big_buffer) that
4988 happened as part of the Exiscan merge. However, to be on the safe side, I
4989 have made the code more robust (and fixed the comments that describe what
4992 33. Some experimental protocols are using DNS PTR records for new purposes. The
4993 keys for these records are domain names, not reversed IP addresses. The
4994 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4995 leaves it alone. Component reversal etc. now happens only for IP addresses.
4998 34. The host_aton() function is supposed to be passed a string that is known
4999 to be a valid IP address. However, in the case of IPv6 addresses, it was
5000 not checking this. This is a hostage to fortune. Exim now panics and dies
5001 if the condition is not met. A case was found where this could be provoked
5002 from a dnsdb PTR lookup with an IPv6 address that had more than 8
5003 components; fortuitously, this particular loophole had already been fixed
5004 by change 4.50/55 or 4.44/33 above.
5006 If there are any other similar loopholes, the new check in host_aton()
5007 itself should stop them being exploited. The report I received stated that
5008 data on the command line could provoke the exploit when Exim was running as
5009 exim, but did not say which command line option was involved. All I could
5010 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
5011 running as the user.
5014 35. There was a buffer overflow vulnerability in the SPA authentication code
5015 (which came originally from the Samba project). I have added a test to the
5016 spa_base64_to_bits() function which I hope fixes it.
5019 36. The daemon start-up calls getloadavg() while still root for those OS that
5020 need the first call to be done as root, but it missed one case: when
5021 deliver_queue_load_max is set with deliver_drop_privilege. This is
5022 necessary for the benefit of the queue runner, because there is no re-exec
5023 when deliver_drop_privilege is set.
5025 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
5026 was in use, was not putting the data itself into the right store pool;
5027 consequently, it could be overwritten for a subsequent message in the same
5028 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
5031 38. Sometimes the final signoff response after QUIT could fail to get
5032 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
5033 < 0 before doing a fflush(). This bug looks as though it goes back to the
5034 introduction of TLS in release 3.20, but "sometimes" must have been rare
5035 because the tests only now provoked it.
5041 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
5042 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
5043 Now it returns an integer. A number of calls were still expecting a T/F
5044 return. Fortuitously, in all cases, the tests worked in OK situations,
5045 which is the norm. However, things would have gone wrong on any write
5046 failures on the smtp file descriptor. This function is used when sending
5047 messages over SMTP and also when doing verify callouts.
5049 2. When Exim is called to do synchronous delivery of a locally submitted
5050 message (the -odf or -odi options), it no longer closes stderr before doing
5053 3. Implemented the mua_wrapper option.
5055 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
5057 5. Implemented the functions header_remove(), header_testname(),
5058 header_add_at_position(), and receive_remove_recipient(), and exported them
5061 6. If an ACL "warn" statement specified the addition of headers, Exim already
5062 inserted X-ACL-Warn: at the start if there was no header name. However, it
5063 was not making this test for the second and subsequent header lines if
5064 there were newlines in the string. This meant that an invalid header could
5065 be inserted if Exim was badly configured.
5067 7. Allow an ACL "warn" statement to add header lines at the start or after all
5068 the Received: headers, as well as at the end.
5070 8. Added the rcpt_4xx retry error code.
5072 9. Added postmaster_mailfrom=xxx to callout verification option.
5074 10. Added mailfrom=xxxx to the callout verification option, for verify=
5077 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
5078 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
5080 12. Inserted some casts to stop certain compilers warning when using pointer
5081 differences as field lengths or precisions in printf-type calls (mostly
5082 affecting debugging statements).
5084 13. Added optional readline() support for -be (dynamically loaded).
5086 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
5087 same clock tick as a message's arrival, so that its received time was the
5088 same as the "first fail" time on the retry record, and that message
5089 remained on the queue past the ultimate address timeout, every queue runner
5090 would try a delivery (because it was past the ultimate address timeout) but
5091 after another failure, the ultimate address timeout, which should have then
5092 bounced the address, did not kick in. This was a "< instead of <=" error;
5093 in most cases the first failure would have been in the next clock tick
5094 after the received time, and all would be well.
5096 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
5097 being recognized when the domain list was tested by the match_domain
5098 condition in an expansion string.
5100 16. Added the ${str2b64: operator.
5102 17. Exim was always calling setrlimit() to set a large limit for the number of
5103 processes, without checking whether the existing limit was already
5104 adequate. (It did check for the limit on file descriptors.) Furthermore,
5105 errors from getrlimit() and setrlimit() were being ignored. Now they are
5106 logged to the main and panic logs, but Exim does carry on, to try to do its
5107 job under whatever limits there are.
5109 18. Imported PCRE 5.0.
5111 19. Trivial typo in log message " temporarily refused connection" (the leading
5114 20. If the log selector return_path_on_delivery was set and an address was
5115 redirected to /dev/null, the delivery process crashed because it assumed
5116 that a return path would always be set for a "successful" delivery. In this
5117 case, the whole delivery is bypassed as an optimization, and therefore no
5120 21. Internal re-arrangement: the function for sending a challenge and reading
5121 a response while authentication was assuming a zero-terminated challenge
5122 string. It's now changed to take a pointer and a length, to allow for
5123 binary data in such strings.
5125 22. Added the cyrus_sasl authenticator (code supplied by MBM).
5127 23. Exim was not respecting finduser_retries when seeking the login of the
5128 uid under which it was called; it was always trying 10 times. (The default
5129 setting of finduser_retries is zero.) Also, it was sleeping after the final
5130 failure, which is pointless.
5132 24. Implemented tls_on_connect_ports.
5134 25. Implemented acl_smtp_predata.
5136 26. If the domain in control=submission is set empty, Exim assumes that the
5137 authenticated id is a complete email address when it generates From: or
5138 Sender: header lines.
5140 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
5141 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
5142 chown and chgrp in /bin and hostname in /usr/bin.
5144 28. Exim was keeping the "process log" file open after each use, just as it
5145 does for the main log. This opens the possibility of it remaining open for
5146 long periods when the USR1 signal hits a daemon. Occasional processlog
5147 errors were reported, that could have been caused by this. Anyway, it seems
5148 much more sensible not to leave this file open at all, so that is what now
5151 29. The long-running daemon process does not normally write to the log once it
5152 has entered its main loop, and it closes the log before doing so. This is
5153 so that log files can straightforwardly be renamed and moved. However,
5154 there are a couple of unusual error situations where the daemon does write
5155 log entries, and I had neglected to close the log afterwards.
5157 30. The text of an SMTP error response that was received during a remote
5158 delivery was being truncated at 512 bytes. This is too short for some of
5159 the long messages that one sometimes sees. I've increased the limit to
5162 31. It is now possible to make retry rules that apply only when a message has a
5163 specific sender, in particular, an empty sender.
5165 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
5166 it possible to be selective about when SMTP synchronization is enforced.
5168 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
5170 32. Implemented hosts_connection_nolog.
5172 33. Added an ACL for QUIT.
5174 34. Setting "delay_warning=" to disable warnings was not working; it gave a
5177 35. Added mailbox_size and mailbox_filecount to appendfile.
5179 36. Added control = no_multiline_responses to ACLs.
5181 37. There was a bug in the logic of the code that waits for the clock to tick
5182 in the case where the clock went backwards by a substantial amount such
5183 that the microsecond fraction of "now" was more than the microsecond
5184 fraction of "then" (but the whole seconds number was less).
5186 38. Added support for the libradius Radius client library this is found on
5187 FreeBSD (previously only the radiusclient library was supported).
5193 1. When certain lookups returned multiple values in the form name=value, the
5194 quoting of the values was not always being done properly. Specifically:
5195 (a) If the value started with a double quote, but contained no whitespace,
5197 (b) If the value contained whitespace other than a space character (i.e.
5198 tabs or newlines or carriage returns) it was not quoted.
5199 This fix has been applied to the mysql and pgsql lookups by writing a
5200 separate quoting function and calling it from the lookup code. The fix
5201 should probably also be applied to nisplus, ibase and oracle lookups, but
5202 since I cannot test any of those, I have not disturbed their existing code.
5204 2. A hit in the callout cache for a specific address caused a log line with no
5205 reason for rejecting RCPT. Now it says "Previous (cached) callout
5206 verification failure".
5208 3. There was an off-by-one bug in the queryprogram router. An over-long
5209 return line was truncated at 256 instead of 255 characters, thereby
5210 overflowing its buffer with the terminating zero. As well as fixing this, I
5211 have increased the buffer size to 1024 (and made a note to document this).
5213 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
5214 when Exim is waiting for an SMTP response from a remote server, Exim
5215 restarts its select() call on the socket, thereby resetting its timeout.
5216 This is not a problem when such interrupts are rare. Somebody set up a cron
5217 job to run exiwhat every 2 minutes, which is less than the normal select()
5218 timeout (5 or 10 minutes). This meant that the select() timeout never
5219 kicked in because it was always reset. I have fixed this by comparing the
5220 time when an interrupt arrives with the time at the start of the first call
5221 to select(). If more time than the timeout has elapsed, the interrupt is
5222 treated as a timeout.
5224 5. Some internal re-factoring in preparation for the addition of Sieve
5225 extensions (by MH). In particular, the "personal" test is moved to a
5226 separate function, and given an option for scanning Cc: and Bcc: (which is
5227 not set for Exim filters).
5229 6. When Exim created an email address using the login of the caller as the
5230 local part (e.g. when creating a From: or Sender: header line), it was not
5231 quoting the local part when it contained special characters such as @.
5233 7. Installed new OpenBSD configuration files.
5235 8. Reworded some messages for syntax errors in "and" and "or" conditions to
5236 try to make them clearer.
5238 9. Callout options, other than the timeout value, were being ignored when
5239 verifying sender addresses in header lines. For example, when using
5241 verify = header_sender/callout=no_cache
5243 the cache was (incorrectly) being used.
5245 10. Added a missing instance of ${EXE} to the exim_install script; this affects
5246 only the Cygwin environment.
5248 11. When return_path_on_delivery was set as a log selector, if different remote
5249 addresses in the same message used different return paths and parallel
5250 remote delivery occurred, the wrong values would sometimes be logged.
5251 (Whenever a remote delivery process finished, the return path value from
5252 the most recently started remote delivery process was logged.)
5254 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
5255 lines when AUTH and/or TLS are in use. This is the "received protocol"
5256 field. Exim used to use "asmtp" for authenticated SMTP, without any
5257 indication (in the protocol name) for TLS use. Now it follows the RFC and
5258 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
5259 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
5260 names appear in log lines as well as in Received: header lines.
5262 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
5263 extensions, and comparison tests, and to fix some bugs.
5265 14. Changes to the "personal" filter test:
5267 (1) The test was buggy in that it was just doing the equivalent of
5268 "contains" tests on header lines. For example, if a user's address was
5269 anne@some.where, the "personal" test would incorrectly be true for
5271 To: susanne@some.where
5273 This test is now done by extracting each address from the header in turn,
5274 and checking the entire address. Other tests that are part of "personal"
5275 are now done using regular expressions (for example, to check local parts
5276 of addresses in From: header lines).
5278 (2) The list of non-personal local parts in From: addresses has been
5279 extended to include "listserv", "majordomo", "*-request", and "owner-*",
5280 taken from the Sieve specification recommendations.
5282 (3) If the message contains any header line starting with "List-" it is
5283 treated as non-personal.
5285 (4) The test for "circular" in the Subject: header line has been removed
5286 because it now seems ill-conceived.
5288 15. Minor typos in src/EDITME comments corrected.
5290 16. Installed latest exipick from John Jetmore.
5292 17. If headers_add on a router specified a text string that was too long for
5293 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
5294 of string_sprintf() is now avoided.
5296 18. $message_body_size was not set (it was always zero) when running the DATA
5297 ACL and the local_scan() function.
5299 19. For the "mail" command in an Exim filter, no default was being set for
5300 the once_repeat time, causing a random time value to be used if "once" was
5301 specified. (If the value happened to be <= 0, no repeat happened.) The
5302 default is now 0s, meaning "never repeat". The "vacation" command was OK
5303 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
5304 (I found it when inspecting the code).
5306 20. There is now an overall timeout for performing a callout verification. It
5307 defaults to 4 times the callout timeout, which applies to individual SMTP
5308 commands during the callout. The overall timeout applies when there is more
5309 than one host that can be tried. The timeout is checked before trying the
5310 next host. This prevents very long delays if there are a large number of
5311 hosts and all are timing out (e.g. when the network connections are timing
5312 out). The value of the overall timeout can be changed by specifying an
5313 additional sub-option for "callout", called "maxwait". For example:
5315 verify = sender/callout=5s,maxwait=20s
5317 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
5318 to the end before writing, but this should make it even safer).
5320 22. Exim was forgetting that it had advertised PIPELINING for the second and
5321 subsequent messages on an SMTP connection. It was also not resetting its
5322 memory on STARTTLS and an internal HELO.
5324 23. When Exim logs an SMTP synchronization error within a session, it now
5325 records whether PIPELINING has been advertised or not.
5327 24. Added 3 instances of "(long int)" casts to time_t variables that were being
5328 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
5329 rather than long int.
5331 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
5333 26. Added the never_mail option to autoreply.
5339 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
5340 crash if the getsockname() call failed; this can happen if a connection is
5341 closed very soon after it is established. The problem was simply in the
5342 order in which certain operations were done, causing Exim to try to write
5343 to the SMTP stream before it had set up the file descriptor. The bug has
5344 been fixed by making things happen in the correct order.
5350 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
5351 before the connection was closed, thus losing the rejection response.
5353 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
5354 some early Solaris releases, but causes trouble in current releases where
5355 socklen_t is defined.
5357 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
5360 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
5363 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
5364 that releasing the top of it at the end releases what was used for sub-
5365 expansions (unless the block got too big). However, discard this block if
5366 the first thing is a variable or header, so that we can use its block when
5367 it is dynamic (useful for very large $message_headers, for example).
5369 6. Lookups now cache *every* query, not just the most recent. A new, separate
5370 store pool is used for this. It can be recovered when all lookup caches are
5371 flushed. Lookups now release memory at the end of their result strings.
5372 This has involved some general refactoring of the lookup sources.
5374 7. Some code has been added to the store_xxx() functions to reduce the amount
5375 of flapping under certain conditions.
5377 8. log_incoming_interface used to affect only the <= reception log lines. Now
5378 it causes the local interface and port to be added to several more SMTP log
5379 lines, for example "SMTP connection from", and rejection lines.
5381 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
5383 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
5385 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
5386 could be overwritten at the end of the current message (or the start of a
5387 new message if it was set in a HELO ACL). The value is now preserved for
5388 the duration of the SMTP connection.
5390 12. If a transport had a headers_rewrite setting, and a matching header line
5391 contained an unqualified address, that address was qualified, even if it
5392 did not match any rewriting rules. The underlying bug was that the values
5393 of the flags that permit the existence of unqualified sender and recipient
5394 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
5395 non-local messages, and by -bnq for local messages) were not being
5396 preserved with the message after it was received.
5398 13. When Exim was logging an SMTP synchronization error, it could sometimes log
5399 "next input=" as part of the text comprising the host identity instead of
5400 the correct text. The code was using the same buffer for two different
5401 strings. However, depending on which order the printing function evaluated
5402 its arguments, the bug did not always show up. Under Linux, for example, my
5403 test suite worked just fine.
5405 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
5406 doesn't work with some older versions of Perl. It has been changed to "my",
5407 which in any case is probably the better facility to use.
5409 15. A really picky compiler found some instances of statements for creating
5410 error messages that either had too many or two few arguments for the format
5413 16. The size of the buffer for calls to the DNS resolver has been increased
5414 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
5415 for addresses that have a lot of PTR records. This alleviates a problem; it
5416 does not fully solve it.
5418 17. A dnsdb lookup for PTR records that receives more data than will fit in the
5419 buffer now truncates the list and logs the incident, which is the same
5420 action as happens when Exim is looking up a host name and its aliases.
5421 Previously in this situation something unpredictable would happen;
5422 sometimes it was "internal error: store_reset failed".
5424 18. If a server dropped the connection unexpectedly when an Exim client was
5425 using GnuTLS and trying to read a response, the client delivery process
5426 crashed while trying to generate an error log message.
5428 19. If a "warn" verb in an ACL added multiple headers to a message in a single
5429 string, for example:
5431 warn message = H1: something\nH2: something
5433 the text was added as a single header line from Exim's point of view
5434 though it ended up OK in the delivered message. However, searching for the
5435 second and subsequent header lines using $h_h2: did not work. This has been
5436 fixed. Similarly, if a system filter added multiple headers in this way,
5437 the routers could not see them.
5439 20. Expanded the error message when iplsearch is called with an invalid key to
5440 suggest using net-iplsearch in a host list.
5442 21. When running tests using -bh, any delays imposed by "delay" modifiers in
5443 ACLs are no longer actually imposed (and a message to that effect is
5446 22. If a "gecos" field in a passwd entry contained escaped characters, in
5447 particular, if it contained a \" sequence, Exim got it wrong when building
5448 a From: or a Sender: header from that name. A second bug also caused
5449 incorrect handling when an unquoted " was present following a character
5450 that needed quoting.
5452 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
5453 was not being matched caselessly.
5455 24. Arranged for all hyphens in the exim.8 source to be escaped with
5458 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
5459 properly. Recipient callout cache records were still being keyed to include
5460 the sender, even when use_sender was set false. This led to far more
5461 callouts that were necessary. The sender is no longer included in the key
5462 when use_sender is false.
5464 26. Added "control = submission" modifier to ACLs.
5466 27. Added the ${base62d: operator to decode base 62 numbers.
5468 28. dnsdb lookups can now access SRV records.
5470 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
5471 the configuration file.
5473 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
5474 (-v) mode. This makes the output for a verbose queue run more intelligible.
5476 31. Added a use_postmaster feature to recipient callouts.
5478 32. Added the $body_zerocount variable, containing the number of binary zero
5479 bytes in the message body.
5481 33. The time of last modification of the "new" subdirectory is now used as the
5482 "mailbox time last read" when there is a quota error for a maildir
5485 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
5487 35. Added +ignore_unknown as a special item in host lists.
5489 36. Code for decoding IPv6 addresses in host lists is now included, even if
5490 IPv6 support is not being compiled. This fixes a bug in which an IPv6
5491 address was recognized as an IP address, but was then not correctly decoded
5492 into binary, causing unexpected and incorrect effects when compared with
5499 1. Very minor rewording of debugging text in manualroute to say "list of
5500 hosts" instead of "hostlist".
5502 2. If verify=header_syntax was set, and a header line with an unqualified
5503 address (no domain) and a large number of spaces between the end of the
5504 name and the colon was received, the reception process suffered a buffer
5505 overflow, and (when I tested it) crashed. This was caused by some obsolete
5506 code that should have been removed. The fix is to remove it!
5508 3. When running in the test harness, delay a bit after writing a bounce
5509 message to get a bit more predictability in the log output.
5511 4. Added a call to search_tidyup() just before forking a reception process. In
5512 theory, someone could use a lookup in the expansion of smtp_accept_max_
5513 per_host which, without the tidyup, could leave open a database connection.
5515 5. Added the variables $recipient_data and $sender_data which get set from a
5516 lookup success in an ACL "recipients" or "senders" condition, or a router
5517 "senders" option, similar to $domain_data and $local_part_data.
5519 6. Moved the writing of debug_print from before to after the "senders" test
5522 7. Change 4.31/66 (moving the time when the Received: is generated) caused
5523 problems for message scanning, either using a data ACL, or using
5524 local_scan() because the Received: header was not generated till after they
5525 were called (in order to set the time as the time of reception completion).
5526 I have revised the way this works. The header is now generated after the
5527 body is received, but before the ACL or local_scan() are called. After they
5528 are run, the timestamp in the header is updated.
5534 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
5535 before starting a queue runner without re-exec. This happened only when
5536 deliver_drop_privilege was set or when the Exim user was set to root. The
5537 effect of the bug was that timeouts during subsequent deliveries caused
5538 crashes instead of being properly handled. The handler is now left at its
5539 default (and expected) setting.
5541 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
5542 message, again when deliver_drop_privilege is set or Exim is run as root.
5543 The bug described in (1) was not present in this case, but the tidying up
5544 of the other signals was missing. I have made the two cases consistent.
5546 3. The ignore_target_hosts setting on a manualroute router was being ignored
5547 for hosts that were looked up using the /MX notation.
5549 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
5552 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
5553 operated on the sender address. After changing the $sender_address to <>
5554 for the sender address verify, Exim was re-instated it as the original
5555 (before rewriting) address, but remembering that it had rewritten it, so it
5556 wasn't rewriting it again. This bug also had the effect of breaking the
5557 sender address verification caching when the sender address was rewritten.
5559 6. The ignore_target_hosts option was being ignored by the ipliteral router.
5560 This has been changed so that if the ip literal address matches
5561 ignore_target_hosts, the router declines.
5563 7. Added expansion conditions match_domain, match_address, and match_local_
5564 part (NOT match_host).
5566 8. The placeholder for the Received: header didn't have a length field set.
5568 9. Added code to Exim itself and to exim_lock to test for a specific race
5569 condition that could lead to file corruption when using MBX delivery. The
5570 issue is with the lockfile that is created in /tmp. If this file is removed
5571 after a process has opened it but before that process has acquired a lock,
5572 there is the potential for a second process to recreate the file and also
5573 acquire a lock. This could lead to two Exim processes writing to the file
5574 at the same time. The added code performs the same test as UW imapd; it
5575 checks after acquiring the lock that its file descriptor still refers to
5576 the same named file.
5578 10. The buffer for building added header lines was of fixed size, 8192 bytes.
5579 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
5582 11. Added the smtp_active_hostname option. If used, this will typically be made
5583 to depend on the incoming interface address. Because $interface_address is
5584 not set up until the daemon has forked a reception process, error responses
5585 that can happen earlier (such as "too many connections") no longer contain
5588 12. If an expansion in a condition on a "warn" statement fails because a lookup
5589 defers, the "warn" statement is abandoned, and the next ACL statement is
5590 processed. Previously this caused the whole ACL to be aborted.
5592 13. Added the iplsearch lookup type.
5594 14. Added ident_timeout as a log selector.
5596 15. Added tls_certificate_verified as a log selector.
5598 16. Added a global option tls_require_ciphers (compare the smtp transport
5599 option of the same name). This controls incoming TLS connections.
5601 17. I finally figured out how to make tls_require_ciphers do a similar thing
5602 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
5603 before starting the TLS session.
5605 18. Tabs are now shown as \t in -bP output.
5607 19. If the log selector return_path_on_delivery was set, Exim crashed when
5608 bouncing a message because it had too many Received: header lines.
5610 20. If two routers both had headers_remove settings, and the first one included
5611 a superfluous trailing colon, the final name in the first list and the
5612 first name in the second list were incorrectly joined into one item (with a
5613 colon in the middle).
5619 1. Added -C and -D options to the exinext utility, mainly to make it easier
5620 to include in the automated testing, but these could be helpful when
5621 multiple configurations are in use.
5623 2. The exinext utility was not formatting the output nicely when there was
5624 an alternate port involved in the retry record key, nor when there was a
5625 message id as well (for retries that were specific to a specific message
5626 and a specific host). It was also confused by IPv6 addresses, because of
5627 the additional colons they contain. I have fixed the IPv4 problem, and
5628 patched it up to do a reasonable job for IPv6.
5630 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
5631 delivery, the log line now contains "pipelined" if PIPELINING was used.
5633 4. An SMTP transport process used to panic and die if the bind() call to set
5634 an explicit outgoing interface failed. This has been changed; it is now
5635 treated in the same way as a connect() failure.
5637 5. A reference to $sender_host_name in the part of a conditional expansion
5638 that was being skipped was still causing a DNS lookup. This no longer
5641 6. The def: expansion condition was not recognizing references to header lines
5642 that used bh_ and bheader_.
5644 7. Added the _cache feature to named lists.
5646 8. The code for checking quota_filecount in the appendfile transport was
5647 allowing one more file than it should have been.
5649 9. For compatibility with Sendmail, the command line option
5657 and sets the incoming protocol and host name (for trusted callers). The
5658 host name and its colon can be omitted when only the protocol is to be set.
5659 Note the Exim already has two private options, -pd and -ps, that refer to
5660 embedded Perl. It is therefore impossible to set a protocol value of "d" or
5661 "s", but I don't think that's a major issue.
5663 10. A number of refactoring changes to the code, none of which should affect
5666 (a) The number of logging options was getting close to filling up the
5667 32-bit word that was used as a bit map. I have split them into two classes:
5668 those that are passed in the argument to log_write(), and those that are
5669 only ever tested independently outside of that function. These are now in
5670 separate 32-bit words, so there is plenty of room for expansion again.
5671 There is no change in the user interface or the logging behaviour.
5673 (b) When building, for example, log lines, the code previously used a
5674 macro that called string_cat() twice, in order to add two strings. This is
5675 not really sufficiently general. Furthermore, there was one instance where
5676 it was actually wrong because one of the argument was used twice, and in
5677 one call a function was used. (As it happened, calling the function twice
5678 did not affect the overall behaviour.) The macro has been replaced by a
5679 function that can join an arbitrary number of extra strings onto a growing
5682 (c) The code for expansion conditions now uses a table and a binary chop
5683 instead of a serial search (which was left over from when there were very
5684 few conditions). Also, it now recognizes conditions like "pam" even when
5685 the relevant support is not compiled in: a suitably worded error message is
5686 given if an attempt is made to use such a condition.
5688 11. Added ${time_interval:xxxxx}.
5690 12. A bug was causing one of the ddress fields not to be passed back correctly
5691 from remote delivery subprocesses. The field in question was not being
5692 subsequently used, so this caused to problems in practice.
5694 13. Added new log selectors queue_time and deliver_time.
5696 14. Might have fixed a bug in maildirsizefile handling that threw up
5697 "unexpected character" debug warnings, and recalculated the data
5698 unnecessarily. In any case, I expanded the warning message to give more
5701 15. Added the message "Restricted characters in address" to the statements in
5702 the default ACL that block characters like @ and % in local parts.
5704 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
5705 Three changes have been made:
5707 (a) There was a serious bug; a negative response to MAIL caused the whole
5708 recipient domain to be cached as invalid, thereby blocking all messages
5709 to all local parts at the same domain, from all senders. This bug has
5710 been fixed. The domain is no longer cached after a negative response to
5711 MAIL if the sender used is not empty.
5713 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
5716 (c) A new callout option, "use_sender" has been added for people who want
5717 the modified behaviour.
5723 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
5726 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
5727 indeed breaks things for older releases.
5729 3. Added additional logging to the case where there is a problem reading data
5730 from a filter that is running in a subprocess using a pipe, in order to
5731 try to track down a specific problem.
5733 4. Testing facility fudge: when running in the test harness and attempting
5734 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
5735 getting "No route to host". Convert this to a timeout.
5737 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
5740 6. Some OS don't have socklen_t but use size_t instead. This affects the
5741 fifth argument of getsockopt() amongst other things. This is now
5742 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
5743 can be set for individual OS. I have set it for SunOS5, OSF1, and
5744 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
5745 some earlier ones do not.
5747 7. Change 4.30/15 was not doing the test caselessly.
5749 8. The standard form for an IPv6 address literal was being rejected by address
5750 parsing in, for example, MAIL and RCPT commands. An example of this kind of
5751 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5752 this, as well as the form without the "IPv6" on the front (but only when
5753 address literals are enabled, of course).
5755 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
5757 10. Exim crashed if a message with an empty sender address specified by -f
5758 encountered a router with an errors_to setting. This could be provoked only
5759 by a command such as
5763 where an empty string was supplied; "<>" did not hit this bug.
5765 11. Installed PCRE release 4.5.
5767 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
5768 remained set. It is now erased.
5770 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
5771 times from message ids (which are base 36 rather than the normal 62).
5773 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
5774 were being counted as actual protocol errors, and logged if the log
5775 selector +smtp_protocol_error was set. One cannot be perfect in this test,
5776 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
5777 and DATA following a set of rejected RCPTs do not count as protocol errors.
5778 In other words, Exim assumes they were pipelined, though this may not
5779 actually be the case. Of course, in all cases the client gets an
5780 appropriate error code.
5782 15. If a lookup fails in an ACL condition, a message about the failure may
5783 be available; it is used if testing the ACL cannot continue, because most
5784 such messages specify what the cause of the deferral is. However, some
5785 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
5786 that caused an old message to be retained and used if a later statement
5787 caused a defer, replacing the real cause of the deferral.
5789 16. If an IP address had so many PTR records that the DNS lookup buffer
5790 was not large enough to hold them, Exim could crash while trying to process
5791 the truncated data. It now detects and logs this case.
5793 17. Further to 4.21/58, another change has been made: if (and only if) the
5794 first line of a message (the first header line) ends with CRLF, a bare LF
5795 in a subsequent header line has a space inserted after it, so as not to
5796 terminate the header.
5798 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
5799 unnecessarily, used atoi() instead of strtol(), and didn't check the
5800 termination when getting file sizes from file names by regex.
5802 19. Completely re-implemented the support for maildirsize files, in the light
5803 of a number of problems with the previous contributed implementation
5804 (4.30/29). In particular:
5806 . If the quota is zero, the maildirsize file is maintained, but no quota is
5809 . If the maildir directory does not exist, it is created before any attempt
5810 to write a maildirsize file.
5812 . The quota value in the file is just a cache; if the quota is changed in
5813 the transport, the new value overrides.
5815 . A regular expression is available for excluding directories from the
5818 20. The autoreply transport checks the characters in options that define the
5819 message's headers; it allows continued headers, but it was checking with
5820 isspace() after an embedded newline instead of explicitly looking for a
5823 21. If all the "regular" hosts to which an address was routed had passed their
5824 expiry times, and had not reached their retry times, the address was
5825 bounced, even if fallback hosts were defined. Now Exim should go on to try
5828 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
5829 equivalent code in the SMTP transport. Some hosts send humungous responses
5830 to HELO/EHLO, more than 1024 it seems.
5832 23. Refactoring: code in filter.c used (void *) for "any old type" but this
5833 gives compiler warnings in some environments. I've now done it "properly",
5836 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
5837 (because of problems with the built-in one) was declared to return uschar *
5838 instead of char *, causing compiler failure.
5840 25. Fixed a file descriptor leak when processing alias/forward files.
5842 26. Fixed a minor format string issue in dbfn.c.
5844 27. Typo in exim.c: ("dmbnz" for "dbmnz").
5846 28. If a filter file refered to $h_xxx or $message_headers, and the headers
5847 contained RFC 2047 "words", Exim's memory could, under certain conditions,
5850 29. When a sender address is verified, it is cached, to save repeating the test
5851 when there is more than one recipient in a message. However, when the
5852 verification involves a callout, it is possible for different callout
5853 options to be set for different recipients. It is too complicated to keep
5854 track of this in the cache, so now Exim always runs a verification when a
5855 callout is required, relying on the callout cache for the optimization.
5856 The overhead is duplication of the address routing, but this should not be
5859 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
5860 to be verified with callout=postmaster, and the main callout worked but the
5861 postmaster check failed, the verification correctly failed. However, if a
5862 subsequent RCPT command asked for sender verification *without* the
5863 postmaster check, incorrect caching caused this verification also to fail,
5866 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
5867 it was not caching the DNS options (qualify_single, search_parents) that
5868 were used when the lookup failed. A subsequent lookup with different
5869 options therefore always gave the same answer, though there were cases
5870 where it should not have. (Example: a "domains = !$mx_any" option on a
5871 dnslookup router: the "domains" option is always processed without any
5872 widening, but the router might have qualify_single set.) Now Exim uses the
5873 cached value only when the same options are set.
5875 32. Added John Jetmore's "exipick" utility to the distribution.
5877 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
5878 than a timeout (e.g. a certificate is required, and is not provided), an
5879 Exim server now closes the connection immediately. Previously it waited for
5880 the client to close - but if the client is SSL, it seems that they each
5881 wait for each other, leading to a delay before one of them times out.
5883 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
5884 maintained 0.8.x compatibility because I don't think many are using it, and
5885 it is clearly obsolete.
5887 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
5890 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
5891 client certificate was expired. A simple patch fixes this, though I don't
5892 understand the full logic of why the verify callback is called multiple
5895 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
5896 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
5897 which causes problems with some clients (such as the Certicom SSL Plus
5898 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
5899 disables the coutermeasure allowing Eudora to connect."
5901 38. Exim was not checking that a write() to a log file succeeded. This could
5902 lead to Bad Things if a log got too big, in particular if it hit a file
5903 size limit. Exim now panics and dies if it cannot write to a log file, just
5904 as it does if it cannot open a log file.
5906 39. Modified OS/Makefile-Linux so that it now contains
5908 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
5910 The two -D definitions ensure that Exim is compiled with large file
5911 support, which makes it possible to handle log files that are bigger than
5914 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
5915 instance) a domain was checked against a named list that involved a lookup,
5916 causing $domain_data to be set, then another domain was checked against the
5917 same list, then the first domain was re-checked, the value of $domain_data
5918 after the final check could be wrong. In particular, if the second check
5919 failed, it could be set empty. This bug probably also applied to
5922 41. The strip_trailing_dot option was not being applied to the address given
5923 with the -f command-line option.
5925 42. The code for reading a message's header from the spool was incrementing
5926 $received_count, but never initializing it. This meant that the value was
5927 incorrect (doubled) while delivering a message in the same process in which
5928 it was received. In the most common configuration of Exim, this never
5929 happens - a fresh exec is done - but it can happen when
5930 deliver_drop_privilege is set.
5932 43. When Exim logs an SMTP synchronization error - client data sent too soon -
5933 it now includes up to 150 characters of the unexpected data in the log
5936 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
5937 and building data strings. The size of both of these buffers was 10 000
5938 bytes - far larger than anybody would *ever* want, thought I. Needless to
5939 say, somebody hit the limit. I have increased the maximum line length to
5940 20 000 and the maximum data length of concatenated lines to 100 000. I have
5941 also fixed two bugs, because there was no checking on these buffers. Tsk,
5942 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
5943 buffer is too small.
5945 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
5946 lsearch lookups. Now it does.
5948 46. When parsing a route_list item in a manualroute router, a fixed-length
5949 buffer was used for the list of hosts. I made this 1024 bytes long,
5950 thinking that nobody would ever have a list of hosts that long. Wrong.
5951 Somebody had a whole pile of complicated expansion conditions, and the
5952 string was silently truncated, leading to an expansion error. It turns out
5953 that it is easier to change to an unlimited length (owing to other changes
5954 that have happened since this code was originally written) than to build
5955 structure for giving a limitation error. The length of the item that
5956 expands into the list of hosts is now unlimited.
5958 47. The lsearch lookup could not handle data where the length of text line was
5959 more than 4095 characters. Such lines were truncated, leading to shortened
5960 data being returned. It should now handle lines of any length.
5962 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
5963 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
5966 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
5967 replacing strings such as BIN_DIRECTORY in a source file: the replacement
5968 no longer happens in comment lines. A list of replacements is now placed
5969 at the head of all of the source files, except those whose only change is
5970 to replace PERL_COMMAND in the very first #! line.
5972 50. Replaced the slow insertion sort in queue.c, for sorting the list of
5973 messages on the queue, with a bottom-up merge sort, using code contributed
5974 by Michael Haardt. This should make operations like -bp somewhat faster on
5975 large queues. It won't affect queue runners, except when queue_run_in_order
5978 51. Installed eximstats 1.31 in the distribution.
5980 52. Added support for SRV lookups to the dnslookup router.
5982 53. If an ACL referred to $message_body or $message_body_end, the value was not
5983 reset for any messages that followed in the same SMTP session.
5985 54. The store-handling optimization for building very long strings was not
5986 differentiating between the different store pools. I don't think this
5987 actually made any difference in practice, but I've tidied it.
5989 55. While running the routers to verify a sender address, $sender_address
5990 was still set to the sender address. This is wrong, because when routing to
5991 send a bounce to the sender, it would be empty. Therefore, I have changed
5992 it so that, while verifying a sender address, $sender_address is set to <>.
5993 (There is no change to what happens when verifying a recipient address.)
5995 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
5996 target A or AAAA records (if not already returned) without resetting the
5997 qualify_single or search_parents options of the DNS resolver. These are
5998 inappropriate in this case because the targets of MX and SRV records must
5999 be FQDNs. A broken DNS record could cause trouble if it happened to have a
6000 target that, when qualified, matched something in the local domain. These
6001 two options are now turned off when doing these lookups.
6003 57. It seems that at least some releases of Reiserfs (which does not have the
6004 concept of a fixed number of inodes) returns zero and not -1 for the
6005 number of available inodes. This interacted badly with check_spool_inodes,
6006 which assumed that -1 was the "no such thing" setting. What I have done is
6007 to check that the total number of inodes is greater than zero before doing
6008 the test of how many are available.
6010 58. When a "warn" ACL statement has a log_message modifier, the message is
6011 remembered, and not repeated. This is to avoid a lot of repetition when a
6012 message has many recipients that cause the same warning to be written.
6013 However, Exim was preserving the list of already written lines for an
6014 entire SMTP session, which doesn't seem right. The memory is now reset if a
6015 new message is started.
6017 59. The "rewrite" debugging flag was not showing the result of rewriting in the
6018 debugging output unless log_rewrite was also set.
6020 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
6021 of (int)(handle) when we know that handle contains (void *)(-1).
6023 61. The Exim daemon panic-logs an error return when it closes the incoming
6024 connection. However "connection reset by peer" seems to be common, and
6025 isn't really an error worthy of noting specially, so that particular error
6028 62. When Exim is trying to find all the local interfaces, it used to panic and
6029 die if the ioctl to get the interface flags failed. However, it seems that
6030 on at least one OS (Solaris 9) it is possible to have an interface that is
6031 included in the list of interfaces, but for which you get a failure error
6032 for this call. This happens when the interface is not "plumbed" into a
6033 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
6034 failure of the "get flags" call assumes that the interface is down.
6036 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
6037 makes life easier for people who are doing arithmetic on fields extracted
6038 from dates, where you often get leading zeros that should not be
6039 interpreted as octal.
6041 64. Added qualify_domain to the redirect router, to override the global
6044 65. If a pathologically long header line contained very many addresses (the
6045 report of this problem mentioned 10 000) and each of them was rewritten,
6046 Exim could use up a very large amount of memory. (It kept on making new
6047 copies of the header line as it rewrote, and never released the old ones.)
6048 At the expense of a bit more processing, the header rewriting function has
6049 been changed so that it no longer eats memory in this way.
6051 66. The generation of the Received: header has been moved from the time that a
6052 message starts to be received, to the time that it finishes. The timestamp
6053 in the Received: header should now be very close to that of the <= log
6054 line. There are two side-effects of this change:
6056 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
6057 logged header lines no longer include the local Received: line, because
6058 it has not yet been created. The same applies to a copy of the message
6059 that is returned to a non-SMTP sender when a message is rejected.
6061 (b) When a filter file is tested using -bf, no additional Received: header
6062 is added to the test message. After some thought, I decided that this
6065 This change does not affect the value of $received_for. It is still set
6066 after address rewriting, but before local_scan() is called.
6068 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
6070 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
6071 gave an unhelpful panic error message, and a defer error. I have managed to
6072 change this behaviour so that it now rejects any supplied certificate,
6073 which seems right, as the list of acceptable certificates is empty.
6075 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
6076 gave an unhelpful defer error. I have not managed to make this reject any
6077 supplied certificates, but the error message it gives is "no certificate
6078 supplied", which is not helpful.
6080 70. exigrep's output now also includes lines that are not associated with any
6081 message, but which match the given pattern. Implemented by a patch from
6082 Martin Sluka, which also tidied up the Perl a bit.
6084 71. Recipient callout verification, like sender verification, was using <> in
6085 the MAIL FROM command. This isn't really the right thing, since the actual
6086 sender may affect whether the remote host accepts the recipient or not. I
6087 have changed it to use the actual sender in the callout; this means that
6088 the cache record is now keyed on a recipient/sender pair, not just the
6089 recipient address. There doesn't seem to be a real danger of callout loops,
6090 since a callout by the remote host to check the sender would use <>.
6091 [SEE ABOVE: changed after hitting problems.]
6093 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
6094 temporary errors. However, in the case of such a code being given after
6095 the end of a data transmission (i.e. after ".") Exim was failing to write
6096 a retry record for the message. (Yes, there was some broken host that was
6097 actually sending 8xx at this point.)
6099 73. An unknown lookup type in a host list could cause Exim to panic-die when
6100 the list was checked. (An example that provoked this was putting <; in the
6101 middle of a list instead of at the start.) If this happened during a DATA
6102 ACL check, a -D file could be left lying around. This kind of configuration
6103 error no longer causes Exim to die; instead it causes a defer error. The
6104 incident is still logged to the main and panic logs.
6106 74. Buglet left over from Exim 3 conversion. The message "too many messages
6107 in one connection" was written to the rejectlog but not the mainlog, except
6108 when address rewriting (yes!) was being logged.
6110 75. Added write_rejectlog option.
6112 76. When a system filter was run not as root (that is, when system_filter_user
6113 was set), the values of the $n variables were not being returned to the
6114 main process; thus, they were not subsequently available in the $sn
6117 77. Added +return_path_on_delivery log selector.
6119 78. A connection timeout was being treated differently from recipients deferred
6120 when testing hosts_max_try with a message that was older than the host's
6121 retry timeout. (The host should not be counted, thus allowing all hosts to
6122 be tried at least once before bouncing.) This may have been the cause of an
6123 occasionally reported bug whereby a message would remain on the queue
6124 longer than the retry timeout, but would be bounced if a delivery was
6125 forced. I say "may" because I never totally pinned down the problem;
6126 setting up timeout/retry tests is difficult. See also the next item.
6128 79. The ultimate address timeout was not being applied to errors that involved
6129 a combination of host plus message (for example, a timeout on a MAIL
6130 command). When an address resolved to a number of possible hosts, and they
6131 were not all tried for each delivery (e.g. because of hosts_max_try), a
6132 message could remain on the queue longer than the retry timeout.
6134 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
6137 81. Fixed an obscure SMTP outgoing bug which required at least the following
6138 conditions: (a) there was another message waiting for the same server;
6139 (b) the server returned 5xx to all RCPT commands in the first message so
6140 that the message was not completed; (c) the server dropped the connection
6141 or gave a negative response to the RSET that Exim sends to abort the
6142 transaction. The observed case was a dropped connection after DATA that had
6143 been sent in pipelining mode. That is, the server had advertised PIPELINING
6144 but was not implementing it correctly. The effect of the bug was incorrect
6145 behaviour, such as trying another host, and this could lead to a crash.
6151 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
6152 and daemon.c were passed as pointers to ints; they should have been
6153 pointers to socklen_t variables (which are typically unsigned ints).
6155 2. Some signed/unsigned type warnings in the os.c file for Linux have been
6158 3. Fixed a really odd bug that affected only the testing scheme; patching a
6159 certain fixed string in the binary changed the value of another string that
6160 happened to be identical to the end of the original first string.
6162 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
6163 name", it returns that address as the IP address. On some operating
6164 systems (e.g. Solaris), it also passes back the IP address string as the
6165 "host name". However, on others (e.g. Linux), it passes back an empty
6166 string. Exim wasn't checking for this, and was changing the host name to an
6167 empty string, assuming it had been canonicalized.
6169 5. Although rare, it is permitted to have more than one PTR record for a given
6170 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
6171 all the names associated with an address, because they do in Solaris.
6172 However, it seems that they do not in Linux for data that comes from the
6173 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
6174 I found this out when I moved to a new Linux workstation and tried to run
6175 the Exim test suite.
6177 To get round this problem I have changed the code so that it now does its
6178 own call to the DNS to look up PTR records when searching for a host name.
6179 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
6180 addresses that are only in /etc/hosts are still found.
6182 This behaviour is, however, controlled by an option called host_lookup_
6183 order, which defaults to "bydns:byaddr". If people want to use the other
6184 order, or indeed, just use one or the other means of lookup, they can
6185 specify it in this variable.
6187 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
6188 some operating systems, this comes back from gethostbyaddr() as an empty
6189 string, and this is what Exim used to test for. However, it seems that in
6190 other systems, "." is yielded. Exim now tests for this case too.
6192 7. The values of check_spool_space and check_log_space are now held internally
6193 as a number of kilobytes instead of an absolute number of bytes. If a
6194 numbers is specified without 'K' or 'M', it is rounded up to the nearest
6195 kilobyte. This means that much larger values can be stored.
6197 8. Exim monitor: an attempt to get the action menu when not actually pointing
6198 at a message produces an empty menu entitled "No message selected". This
6199 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
6200 no entries in it ("Shell widget menu has zero width and/or height"). So I
6201 have added a single, blank menu entry in this case.
6203 9. Added ${quote_local_part.
6205 10. MIME decoding is now applied to the contents of Subject: header lines when
6208 11. Now that a reference to $sender_host_address automatically causes a reverse
6209 lookup to occur if necessary (4.13/18), there is no need to arrange for a
6210 host lookup before query-style lookups in lists that might use this
6211 variable. This has therefore been abolished, and the "net-" prefix is no
6212 longer necessary for query-style lookups.
6214 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
6215 have been a typo for LFLAGS, so it has been changed.
6217 13. The install script calls Exim with "-C /dev/null" in order to find the
6218 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
6219 to be output. However, since Exim outputs its version number before the
6220 error, it didn't break the script. It just looked ugly. I fixed this by
6221 always allowing "-C /dev/null" if the caller is root.
6223 14. Ignore overlarge ACL variable number when reading spool file - insurance
6224 against a later release with more variables having written the file.
6226 15. The standard form for an IPv6 address literal was being rejected by EHLO.
6227 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
6228 this, as well as the form without the "IPv6" on the front.
6230 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
6231 OS/Makefile-Darwin file.
6233 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
6234 only to LDAP libraries that do not have LDAP_OPT_DEREF.
6236 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
6237 variable. However, some OS use ints for inodes. I've added cast to long int
6238 to get rid of the compiler warning.
6240 19. I had forgotten to lock out "/../" in configuration file names when
6241 ALT_CONFIG_PREFIX was set.
6243 20. Routers used for verification do not need to specify transports. However,
6244 if such a router generated a host list, and callout was configured, Exim
6245 crashed, because it could not find a port number from the (non-existent)
6246 transport. It now assumes port 25 in this circumstance.
6248 21. Added the -t option to exigrep.
6250 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
6251 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
6252 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
6253 them is set, LOOKUP_LSEARCH is forced.
6255 23. "exim -bV" now outputs a list of lookups that are included in the binary.
6257 24. Added sender and host information to the "rejected by local_scan()" log
6258 line; previously there was no indication of these.
6260 25. Added .include_if_exists.
6262 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
6263 Linux. It turns out that not all file systems support this. Apparently some
6264 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
6265 do it.) To cope with this, the error EINVAL, which means that sync-ing is
6266 not supported on the file descriptor, is now ignored when Exim is trying to
6267 sync a directory. This applies only to Linux.
6269 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
6271 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
6272 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
6275 29. Added support for maildirsize files from supplied patch (modified a bit).
6277 30. The use of :fail: followed by an empty string could lead Exim to respond to
6278 sender verification failures with (e.g.):
6280 550 Verification failed for <xxx>
6281 550 Sender verify failed
6283 where the first response line was missing the '-' that indicates it is not
6284 the final line of the response.
6286 31. The loop for finding the name of the user that called Exim had a hardwired
6287 limit of 10; it now uses the value of finduser_retries, which is used for
6288 all other user lookups.
6290 32. Added $received_count variable, available in data and not_smtp ACLs, and at
6293 33. Exim was neglecting to zero errno before one call of strtol() when
6294 expanding a string and expecting an integer value. On some systems this
6295 resulted in spurious "integer overflow" errors. Also, it was casting the
6296 result into an int without checking.
6298 34. Testing for a connection timeout using "timeout_connect" in the retry rules
6299 did not work. The code looks as if it has *never* worked, though it appears
6300 to have been documented since at least release 1.62. I have made it work.
6302 35. The "timeout_DNS" error in retry rules, also documented since at least
6303 1.62, also never worked. As it isn't clear exactly what this means, and
6304 clearly it isn't a major issue, I have abolished the feature by treating it
6305 as "timeout", and writing a warning to the main and panic logs.
6307 36. The display of retry rules for -brt wasn't always showing the error code
6310 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
6311 timeout_connect_A, timeout_connect_MX.
6313 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
6314 to the empty sender.
6316 39. The daemon was not analysing the content of -oX till after it had closed
6317 stderr and disconnected from the controlling terminal. This meant that any
6318 syntax errors were only noted on the panic log, and the return code from
6319 the command was 0. By re-arranging the code a little, I've made the
6320 decoding happen first, so such errors now appear on stderr, and the return
6321 code is 1. However, the actual setting up of the sockets still happens in
6322 the disconnected process, so errors there are still only recorded on the
6325 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
6326 connections (as happens on some IP stacks) was logged at start up time as
6327 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
6328 it from "IPv6 and IPv4", which means that two separate sockets are being
6331 41. The debug output for gethostbyname2() or getipnodebyname() failures now
6332 says whether AF_INET or AF_INET6 was passed as an argument.
6334 42. Exiwhat output was messed up when time zones were included in log
6337 43. Exiwhat now gives more information about the daemon's listening ports,
6338 and whether -tls-on-connect was used.
6340 44. The "port" option of the smtp transport is now expanded.
6342 45. A "message" modifier in a "warn" statement in a non-message ACL was being
6343 silently ignored. Now an error message is written to the main and panic
6346 46. There's a new ACL modifier called "logwrite" which writes to a log file
6347 as soon as it is encountered.
6349 47. Added $local_user_uid and $local_user_gid at routing time.
6351 48. Exim crashed when trying to verify a sender address that was being
6354 49. Exim was recognizing only a space character after ".include". It now also
6355 recognizes a tab character.
6357 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
6358 extracting the relevant information from the specification. The man page no
6359 longer contains scrambled data for the -d option, and I've added a section
6360 at the front about calling Exim under different names.
6362 51. Added "extra_headers" argument to the "mail" command in filter files.
6364 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
6367 53. Installed eximstats 1.29.
6369 54. Added transport_filter_timeout as a generic transport option.
6371 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
6372 Cc: header lines. This was required by RFC 822, but it not required by RFC
6375 56. Exim used to add From:, Date:, and Message-Id: header lines to any
6376 incoming messages that did not have them. Now it does so only if the
6377 message originates locally, that is, if there is no associated remote host
6378 address. When Resent- header lines are present, this applies to the Resent-
6379 lines rather than the non-Resent- lines.
6381 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
6382 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
6384 58. Messages for configuration errors now include the name of the main
6385 configuration file - useful now that there may be more than one file in a
6386 list (.included file names were always shown).
6388 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
6389 for those rare installations that do not start the daemon as root or run it
6390 setuid root. I've cut out the call to initgroups() if the daemon is not
6393 60. The Exim user and group can now be bound into the binary as text strings
6394 that are looked up at the start of Exim's processing.
6396 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
6398 62. Added $mailstore_basename variable.
6400 63. Installed patch to sieve.c from Michael Haardt.
6402 64. When Exim failed to open the panic log after failing to open the main log,
6403 the original message it was trying to log was written to stderr and debug
6404 output, but if they were not available (the usual case in production), it
6405 was lost. Now it is written to syslog before the two lines that record the
6406 failures to open the logs.
6408 65. Users' Exim filters run in subprocesses under the user's uid. It is
6409 possible for a "deliver" command or an alias in a "personal" command to
6410 provoke an address rewrite. If logging of address rewriting is configured,
6411 this fails because the process is not running as root or exim. There may be
6412 a better way of dealing with this, but for the moment (because 4.30 needs
6413 to be released), I have disabled address rewrite logging when running a
6414 filter in a non-root, non-exim process.
6420 1. The buildconfig auxiliary program wasn't quoting the value set for
6421 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
6422 not defined. This bug was masked in 4.22 by the effect that was fixed in
6425 2. Some messages that were rejected after a message id was allocated were
6426 shown as "incomplete" by exigrep. It no longer does this for messages that
6427 are rejected by local_scan() or the DATA or non-SMTP ACLs.
6429 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
6430 have allow_domain_literals set, the ID did not get logged in the <= line.
6431 Domain literals are now always recognized in Message-ID: header lines.
6433 4. The first argument for a ${extract expansion item is the key name or field
6434 number. Leading and trailing spaces in this item were not being ignored,
6435 causing some misleading effects.
6437 5. When deliver_drop_privilege was set, single queue runner processes started
6438 manually (i.e. by the command "exim -q") or by the daemon (which uses the
6439 same command in the process it spins off) were not dropping privilege.
6441 6. When the daemon running as "exim" started a queue runner, it always
6442 re-executed Exim in the spun-off process. This is a waste of effort when
6443 deliver_drop_privilege is set. The new process now just calls the
6444 queue-runner function directly.
6450 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
6453 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
6454 ignored. Though the use of -oP was forcing the writing of a pid file, it
6455 was always written to the default place.
6457 3. If the message "no IP address found for host xxxx" is generated during
6458 incoming verification, it is now followed by identification of the incoming
6459 connection (so you can more easily find what provoked it).
6461 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
6463 5. Added some features to "harden" Exim a bit more against certain attacks:
6465 (a) There is now a build-time option called FIXED_NEVER_USERS that can
6466 be put in Local/Makefile. This is like the never_users runtime option,
6467 but it cannot be overridden. The default setting is "root".
6469 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
6470 prefix string with which any file named in a -C command line option
6473 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
6474 is retained for -C and -D only if the caller of Exim is root. Without
6475 it, the exim user may also use -C and -D and retain privilege.
6477 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
6478 command line option is disabled.
6480 6. Macro names set by the -D option must start with an upper case letter, just
6481 like macro names defined in the configuration file.
6483 7. Added "dereference=" facility to LDAP.
6485 8. Two instances of the typo "uknown" in the source files are fixed.
6487 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
6488 the Configure-Makefile script screwed up while processing it.
6490 10. Incorporated PCRE 4.4.
6492 11. The SMTP synchronization check was not operating right at the start of an
6493 SMTP session. For example, it could not catch a HELO sent before the client
6494 waited for the greeting. There is now a check for outstanding input at the
6495 point when the greeting is written. Because of the duplex, asynchronous
6496 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
6497 way, but not yet received, when the check is performed.
6499 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
6500 on TCP/IP sockets, because this apparently causes some broken clients to
6503 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
6504 unchanged) from the Cygwin maintainer.
6506 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
6507 was supported instead of testing for mbx. Effectively a typo.
6509 15. The spa authenticator server code was not checking that the input it
6510 received was valid base64.
6512 16. The debug output line for the "set" modifier in ACLs was not showing the
6513 name of the variable that was being set.
6515 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
6517 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
6518 lookup on its own. Something else was needed to trigger the lookup. For
6519 example, a match in host_lookup or the need for a host name in a host list.
6520 Now, if $sender_host_name is referenced and the host name has not yet been
6521 looked up, a lookup is performed. If the lookup fails, the variable remains
6522 empty, and $host_lookup_failed is set to "1".
6524 19. Added "eqi" as a case-independent comparison operator.
6526 20. The saslauthd authentication condition could segfault if neither service
6527 nor realm was specified.
6529 21. If an overflowing value such as "2048M" was set for message_size_limit, the
6530 error message that was logged was misleading, and incoming SMTP
6531 connections were dropped. The message is now more accurate, and temporary
6532 errors are given to SMTP connections.
6534 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
6535 (except RSET) with a 421 error, until QUIT is received. However, it was
6536 failing to send a response to QUIT.
6538 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
6539 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
6540 ACL is now run after the helo_try_verify_hosts code.
6542 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
6543 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
6544 case-independent, so other case variants are also recognized). Apparently
6545 some systems use these upper case variants.
6547 25. If more than two messages were waiting for the same host, and a transport
6548 filter was specified for the transport, Exim sent two messages over the
6549 same TCP/IP connection, and then failed with "socket operation on non-
6550 socket" when it tried to send the third.
6552 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
6554 27. The extern definition of crypt16() in expand.c was not being excluded when
6555 the OS had its own crypt16() function.
6557 28. Added bounce_return_body as a new option, and bounce_return_size_limit
6558 as a preferred synonym for return_size_limit, both as an option and as an
6561 29. Added LIBS=-liconv to OS/Makefile-OSF1.
6563 30. Changed the default configuration ACL to relax the local part checking rule
6564 for addresses that are not in any local domains. For these addresses,
6565 slashes and pipe symbols are allowed within local parts, but the sequence
6566 /../ is explicitly forbidden.
6568 31. SPA server authentication was not clearing the challenge buffer before
6571 32. log_message in a "warn" ACL statement was writing to the reject log as
6572 well as to the main log, which contradicts the documentation and doesn't
6573 seem right (because no rejection is happening). So I have stopped it.
6575 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
6576 However, I am unable to do any testing of this.
6578 34. Fixed an infelicity in the appendfile transport. When checking directories
6579 for a mailbox, to see if any needed to be created, it was accidentally
6580 using path names with one or more superfluous leading slashes; tracing
6581 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
6583 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
6584 contents are added to the log line that is written for every discarded
6585 recipient. (Previously a log_message setting was ignored.)
6587 36. The ${quote: operator now quotes the string if it is empty.
6589 37. The install script runs exim in order to find its version number. If for
6590 some reason other than non-existence or emptiness, which it checks, it
6591 could not run './exim', it was installing it with an empty version number,
6592 i.e. as "exim-". This error state is now caught, and the installation is
6595 38. An argument was missing from the function that creates an error message
6596 when Exim fails to connect to the socket for saslauthd authentication.
6597 This could cause Exim to crash, or give a corrupted message.
6599 39. Added isip, isip4, and isip6 to ${if conditions.
6601 40. The ACL variables $acl_xx are now saved with the message, and can be
6602 accessed later in routers, transports, and filters.
6604 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
6605 strings in the file are not string-expanded.
6607 42. If a MAIL command specified a SIZE value that was too large to fit into an
6608 int variable, the check against message_size_limit failed. Such values are
6609 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
6610 day this will have to be increased, but I don't think I want to be around
6611 when emails are that large.
6618 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
6619 iconv() is not standard in FreeBSD.
6621 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
6622 IPv6 enabled. The observed symptom was a segmentation fault on return from
6623 the function os_common_find_running_interfaces() in src/os.c.
6625 3. In the check_special_case() function in daemon.c I had used "errno" as an
6626 argument name, which causes warnings on some systems. This was basically a
6627 typo, since it was named "eno" in the comments!
6629 4. The code that waits for the clock to tick (at a resolution of some fraction
6630 of a second) so as to ensure message-id uniqueness was always waiting for
6631 at least one whole tick, when it could have waited for less. [This is
6632 almost certainly not relevant at current processor speeds, where it is
6633 unlikely to ever wait at all. But we try to future-proof.]
6635 5. The function that sleeps for a time interval that includes fractions of a
6636 second contained a race. It did not block SIGALRM between setting the
6637 timer, and suspending (a couple of lines later). If the interval was short
6638 and the sigsuspend() was delayed until after it had expired, the suspension
6639 never ended. On busy systems this could lead to processes getting stuck for
6642 6. Some uncommon configurations may cause a lookup to happen in a queue runner
6643 process, before it forks any delivery processes. The open lookup caching
6644 mechanism meant that the open file or database connection was passed into
6645 the delivery process. The problem was that delivery processes always tidy
6646 up cached lookup data. This could cause a problem for the next delivery
6647 process started by the queue runner, because the external queue runner
6648 process does not know about the closure. So the next delivery process
6649 still has data in the lookup cache. In the case of a file lookup, there was
6650 no problem because closing a file descriptor in a subprocess doesn't affect
6651 the parent. However, if the lookup was caching a connection to a database,
6652 the connection was closed, and the second delivery process was likely to
6653 see errors such as "PGSQL: query failed: server closed the connection
6654 unexpectedly". The problem has been fixed by closing all cached lookups
6655 in a queue runner before running a delivery process.
6657 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
6658 seem to have the "const" qualifier which it has on other OS. I've
6661 8. Change 4.21/2 was too strict. It is only if there are two authenticators
6662 *of the same type* (client or server) with the same public name that an
6663 error should be diagnosed.
6665 9. When Exim looked up a host name for an IP address, but failed to find the
6666 original IP address when looking up the host name (a safety check), it
6667 output the message "<ip address> does not match any IP for NULL", which was
6668 confusing, to say the least. The bug was that the host name should have
6669 appeared instead of "NULL".
6671 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
6672 user that is built into the binary, and the -C or -D options is used, root
6673 privilege is dropped before the configuration file is read. In addition,
6674 logging is switched to stderr instead of the normal log files. If the
6675 configuration then re-defines the Exim user, the unprivileged environment
6676 is probably not what is expected, so Exim logs a panic warning message (but
6679 However, if deliver_drop_privilege is set, the unprivileged state may well
6680 be exactly what is intended, so the warning has been cut out in that case,
6681 and Exim is allowed to try to write to its normal log files.
6687 1. smtp_return_error_details was not giving details for temporary sender
6688 or receiver verification errors.
6690 2. Diagnose a configuration error if two authenticators have the same public
6693 3. Exim used not to create the message log file for a message until the first
6694 delivery attempt. This could be confusing when incoming messages were held
6695 for policy or load reasons. The message log file is now created at the time
6696 the message is received, and an initial "Received" line is written to it.
6698 4. The automatically generated man page for command line options had a minor
6699 bug that caused no ill effects; however, a more serious problem was that
6700 the procedure for building the man page automatically didn't always
6701 operate. Consequently, release 4.20 contains an out-of-date version. This
6702 shouldn't happen again.
6704 5. When building Exim with embedded Perl support, the script that builds the
6705 Makefile was calling 'perl' to find its compile-time parameters, ignoring
6706 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
6708 6. The freeze_tell option was not being used for messages that were frozen on
6709 arrival, either by an ACL or by local_scan().
6711 7. Added the smtp_incomplete_transaction log selector.
6713 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
6714 was accepting AUTH without a new EHLO.
6716 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
6717 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
6720 10. Logging of TCP/IP connections (when configured) now happens in the main
6721 daemon process instead of the child process, so that the TCP/IP connection
6722 count is more accurate (but it can never be perfect).
6724 11. The use of "drop" in a nested ACL was not being handled correctly in the
6725 outer ACL. Now, if condition failure induced by the nested "drop" causes
6726 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
6727 or "require"), the connection is dropped.
6729 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
6730 that yield "discard" can now be used with an "accept" or a "discard" verb,
6731 but an error is generated for any others (because I can't see a useful way
6732 to define what should happen).
6734 13. When an ACL is read dynamically from a file (or anywhere else), the lines
6735 are now processed in the same way as lines in the Exim configuration file.
6736 In particular, continuation lines are supported.
6738 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
6740 15. Added -ti meaning -t -i.
6742 16. Check for letters, digits, hyphens, and dots in the names of dnslist
6743 domains, and warn by logging if others are found.
6745 17. At least on BSD, alignment is not guaranteed for the array of ifreq's
6746 returned from GIFCONF when Exim is trying to find the list of interfaces on
6747 a host. The code in os.c has been modified to copy each ifreq to an aligned
6748 structure in all cases.
6750 Also, in some cases, the returned ifreq's were being copied to a 'struct
6751 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
6752 means the last couple of bytes of an IPv6 address could be chopped if the
6753 ifreq contained only a normal sockaddr (14 bytes storage).
6755 18. Named domain lists were not supported in the hosts_treat_as_local option.
6756 An entry such as +xxxx was not recognized, and was treated as a literal
6759 19. Ensure that header lines added by a DATA ACL are included in the reject log
6760 if the ACL subsequently rejects the message.
6762 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
6763 MD5 (which is deprecated).
6765 21. When testing a filter file using -bf, Exim was writing a message when it
6766 took the sender from a "From " line in the message, but it was not doing so
6767 when it took $return_path from a Return-Path: header line. It now does.
6769 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
6770 with a valid header line field name (a series of printing characters
6771 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
6773 23. Changed "disc" in the source to "disk" to conform to the documentation and
6774 the book and for uniformity.
6776 24. Ignore Sendmail's -Ooption=value command line item.
6778 25. When execve() failed while trying to run a command in a pipe transport,
6779 Exim was returning EX_UNAVAILABLE (69) from the subprocess. However, this
6780 could be confused with a return value of 69 from the command itself. This
6781 has been changed to 127, the value the shell returns if it is asked to run
6782 a non-existent command. The wording for the related log line suggests a
6783 non-existent command as the problem.
6785 26. If received_header_text expands to an empty string, do not add a Received:
6786 header line to the message. (Well, it adds a token one on the spool, but
6787 marks it "old" so that it doesn't get used or transmitted.)
6789 27. Installed eximstats 1.28 (addition of -nt option).
6791 28. There was no check for failure on the call to getsockname() in the daemon
6792 code. This can fail if there is a shortage of resources on the system, with
6793 ENOMEM, for example. A temporary error is now given on failure.
6795 29. Contrary to the C standard, it seems that in some environments, the
6796 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
6797 program. Exim now does this explicitly; it affects the formatting of
6798 timestamps using strftime().
6800 30. If exiqsumm was given junk data, it threw up some uninitialized variable
6801 complaints. I've now initialized all the variables, to avoid this.
6803 32. Header lines added by a system filter were not being "seen" during
6804 transport-time rewrites.
6806 33. The info_callback() function passed to OpenSSL is set up with type void
6807 (*)(SSL *, int, int), as described somewhere. However, when calling the
6808 function (actually a macro) that sets it up, the type void(*)() is
6809 expected. I've put in a cast to prevent warnings from picky compilers.
6811 34. If a DNS black list lookup found a CNAME record, but there were no A
6812 records associated with the domain it pointed at, Exim crashed.
6814 35. If a DNS black list lookup returned more than one A record, Exim ignored
6815 all but the first. It now scans all returned addresses if a particular IP
6816 value is being sought. In this situation, the contents of the
6817 $dnslist_value variable are a list of all the addresses, separated by a
6820 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
6821 to accept a host name and all its aliases if the forward lookup for any of
6822 them yielded the IP address of the incoming connection. Now it accepts only
6823 those names whose forward lookup yields the correct IP address. Any other
6824 names are discarded. This closes a loophole whereby a rogue DNS
6825 administrator could create reverse DNS records to break through a
6826 wildcarded host restriction in an ACL.
6828 37. If a user filter or a system filter that ran in a subprocess used any of
6829 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
6830 the wrong values were passed to the pipe command ($thisaddress had the
6831 value of $0, $0 had the value of $1, etc). This bug was introduced by
6832 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
6834 38. Improved the line breaking for long SMTP error messages from ACLs.
6835 Previously, if there was no break point between 40 and 75 characters, Exim
6836 left the rest of the message alone. Two changes have been made: (a) I've
6837 reduced the minimum length to 35 characters; (b) if it can't find a break
6838 point between 35 and 75 characters, it looks ahead and uses the first one
6839 that it finds. This may give the occasional overlong line, but at least the
6840 remaining text gets split now.
6842 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
6843 file descriptors might be low, and that setting 1000 would always raise it.
6844 It turns out that in some environments, the limit is already over 1000 and
6845 that lowering it causes trouble. So now Exim takes care not to decrease it.
6847 40. When delivering a message, the value of $return_path is set to $sender_
6848 address at the start of routing (routers may change the value). By an
6849 oversight, this default was not being set up when an address was tested by
6850 -bt or -bv, which affected the outcome if any router or filter referred to
6853 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
6854 log only when "message" or "log_message" are set. However, if one of the
6855 conditions was an address verification, or a call to a nested ACL, the
6856 messages generated by the underlying test were being passed through. This
6857 no longer happens. The underlying message is available in $acl_verify_
6858 message for both "message" and "log_message" expansions, so it can be
6859 passed through if needed.
6861 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
6862 new expansion $bh_ to give the encoded byte string without charset
6863 translation. Translation happens only if iconv() is available; HAVE_ICONV
6864 indicates this at build time. HEADERS_CHARSET gives the charset to
6865 translate to; headers_charset can change it in the configuration, and
6866 "headers charset" can change it in an individual filter file.
6868 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
6869 that creates RFC 2047 encoded "words" labels them as that charset instead
6870 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
6871 expansion operator; (ii) when Exim creates a From: line for a local
6872 message; (iii) when a header line is rewritten to include a "phrase" part.
6874 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
6875 buggy, causing it to skip the first lines of messages whose message ID
6876 ended in 'D'. This would not have bitten before Exim release 4.14, because
6877 message IDs were unlikely to end in 'D' before then. The effect was to have
6878 incorrect size information for certain domains.
6880 45. #include "config.h" was missing at the start of the crypt16.c module. This
6881 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
6884 46. If there was a timeout during a "random" callout check, Exim treated it as
6885 a failure of the random address, and carried on sending RSET and the real
6886 address. If the delay was just some slowness somewhere, the response to the
6887 original RCPT would be taken as a response to RSET and so on, causing
6888 mayhem of various kinds.
6890 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
6891 when I implemented it. It didn't allow for the fact that some option values
6892 may legitimately be negative (e.g. size_addition), and it didn't even do
6893 the right test for positive values.
6895 48. Domain names in DNS records are case-independent. Exim always looks them up
6896 in lower case. Some resolvers return domain names in exactly the case they
6897 appear in the zone file, that is, they may contain uppercase letters. Not
6898 all resolvers do this - some return always lower case. Exim was treating a
6899 change of case by a resolver as a change of domain, similar to a widening
6900 of a domain abbreviation. This triggered its re-routing code and so it was
6901 trying to route what was effectively the same domain again. This normally
6902 caused routing to fail (because the router wouldn't handle the domain
6903 twice). Now Exim checks for this case specially, and just changes the
6904 casing of the domain that it ultimately uses when it transmits the message
6907 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
6910 50. If a filter generated a file delivery with a non-absolute name (possible if
6911 no home directory exists for the router), the forbid_file option was not
6914 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
6915 the existing equality matching.
6917 52. Exim was using ints instead of ino_t variables in some places where it was
6918 dealing with inode numbers.
6920 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
6921 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
6922 called TMPDIR, and if it finds it is different, it changes its value.
6924 54. The smtp_printf() function is now made available to local_scan() so
6925 additional output lines can be written before returning. There is also an
6926 smtp_fflush() function to enable the detection of a dropped connection.
6927 The variables smtp_input and smtp_batched_input are exported to
6930 55. Changed the default runtime configuration: the message "Unknown user"
6931 has been removed from the ACL, and instead placed on the localuser router,
6932 using the cannot_route_message feature. This means that any verification
6933 failures that generate their own messages won't get overridden. Similarly,
6934 the "Unrouteable address" message that was in the ACL for unverifiable
6935 relay addresses has also been removed.
6937 56. Added hosts_avoid_esmtp to the smtp transport.
6939 57. The exicyclog script was not checking for the esoteric option
6940 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
6941 will work only if exicyclog is run under the appropriate euid.
6943 58. Following a discussion on the list, the rules by which Exim recognises line
6944 endings on incoming messages have been changed. The -dropcr and drop_cr
6945 options are now no-ops, retained only for backwards compatibility. The
6946 following line terminators are recognized: LF CRLF CR. However, special
6947 processing applies to CR:
6949 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
6950 nor a local message in the state where . is a terminator.
6952 (ii) If a bare CR is encountered in a header line, an extra space is added
6953 after the line terminator so as not to end the header. The reasoning
6954 behind this is that bare CRs in header lines are most likely either
6955 to be mistakes, or people trying to play silly games.
6957 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
6958 was being incorrectly given as 18 bytes larger than it should have been.
6959 This is a VOB (very old bug).
6961 60. This may never have affected anything current, but just in case it has:
6962 When the local host is found other than at the start of a list of hosts,
6963 the local host, those with the same MX, and any that follow, are discarded.
6964 When the list in question was part of a longer list of hosts, the following
6965 hosts (not currently being processed) were also being discarded. This no
6966 longer happens. I'm not sure if this situation could ever has previously
6969 61. Added the "/MX" feature to lists of hosts in the manualroute and query
6972 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
6973 header. This is something that is recommended in a new Internet Draft, and
6974 is something that is documented as being done by Sendmail. There are two
6975 possible values. For messages generated by the autoreply transport, Exim
6978 Auto-Submitted: auto-replied
6980 whereas for all other generated messages (e.g. bounces) it adds
6982 Auto-Submitted: auto-generated
6984 63. The "personal" condition in filters now includes a test for the
6985 Auto-Submitted: header. If it contains the string "auto-" the message it
6986 not considered personal.
6988 64. Added rcpt_include_affixes as a generic transport option.
6990 65. Added queue_only_override (default true).
6992 66. Added the syslog_duplication option.
6994 67. If what should have been the first header line of a message consisted of
6995 a space followed by a colon, Exim was mis-interpreting it as a header line.
6996 It isn't of course - it is syntactically invalid and should therefore be
6997 treated as the start of the message body. The misbehaviour could have
6998 caused a number of strange effects, including loss of data in subsequent
6999 header lines, and spool format errors.
7001 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
7002 client host had authenticated. This control can now be exercised by an ACL
7003 for more flexibility.
7005 69. By default, callouts do not happen when testing with -bh. There is now a
7006 variant, -bhc, which does actually run the callout code, including
7007 consulting and updating the callout cache.
7009 70. Added support for saslauthd authentication, courtesy of Alexander
7012 71. If statvfs() failed on the spool or log directories while checking their
7013 size for availability, Exim confusingly gave the error "space shortage".
7014 Furthermore, in debugging mode it crashed with a floating point exception.
7015 These checks are done if check_{spool,log}_{space,inodes} are set, and when
7016 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
7017 serious problem, Exim now writes to the main and panic logs when this
7018 happens, with details of the failure. It then refuses to accept the
7019 incoming message, giving the message "spool directory problem" or "log
7020 directory problem" with a 421 code for SMTP messages.
7022 72. When Exim is about to re-exec itself, it ensures that the file descriptors
7023 0, 1, and 2 exist, because some OS complain for execs without them (see
7024 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
7025 descriptors. However, the code omitted to check that the open succeeded,
7026 causing mysterious errors if for some reason the permissions on /dev/null
7027 got screwed. Now Exim writes a message to the main and panic logs, and
7028 bombs out if it can't open /dev/null.
7030 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
7031 interact so that it is all more flexible. It is supposed to remain
7032 backwards compatible. Also added extra_local_interfaces.
7034 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
7035 to bomb out with an assertion failure - to the client this appears as a
7036 connection drop. This problem occurs in the part of the code that was taken
7037 from the Samba project. Fortunately, the assertion is in a very simple
7038 function, so I have fixed this by reproducing the function inline in the
7039 one place where it is called, and arranging for authentication to fail
7040 instead of killing the process with assert().
7042 75. The SPA client code was not working when the server requested OEM rather
7043 than Unicode encoding.
7045 76. Added code to make require_files with a specific uid setting more usable in
7046 the case where statting the file as root fails - usually a non-root-mounted
7047 NFS file system. When this happens and the failure is EACCES, Exim now
7048 forks a subprocess and does the per-uid checking as the relevant uid.
7050 77. Added process_log_path.
7052 78. If log_file_path was not explicitly set, a setting of check_log_space or
7053 check_log_inodes was ignored.
7055 79. If a space check for the spool or log partitions fails, the incident is now
7056 logged. Of course, in the latter case the data may get lost...
7058 80. Added the %p formatting code to string_format() so that it can be used to
7059 print addresses in debug_print(). Adjusted all the address printing in the
7060 debugging in store.c to use %p rather than %d.
7062 81. There was a concern that a line of code in smtp_in.c could overflow a
7063 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
7064 initially expressed, the concern was not well-founded, because trailing
7065 spaces are removed early. However, if the trailing spaces were followed by
7066 a NULL, they did not get removed, so the overflow was possible. Two fixes
7069 (a) I re-wrote the offending code in a cleaner fashion.
7070 (b) If an incoming SMTP command contains a NULL character, it is rejected
7073 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
7074 runs initgroups(), so that if the Exim user is in any additional groups,
7075 they will be used during message reception.
7081 The change log for 4.20 and earlier releases has been archived.
projects / exim.git / blob