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.
11 GF/01 DEFER rather than ERROR on redis cluster MOVED response.
12 When redis_servers is set to a list of > 1 element, and the Redis servers
13 in that list are in cluster configuration, convert the REDIS_REPLY_ERROR
14 case of MOVED into a DEFER case instead, thus moving the query onto the
15 next server in the list. For a cluster of N elements, all N servers must
16 be defined in redis_servers.
18 GF/02 Catch and remove uninitialized value warning in exiqsumm
19 Check for existence of @ARGV before looking at $ARGV[0]
21 JH/01 Replace the store_release() internal interface with store_newblock(),
22 which internalises the check required to safely use the old one, plus
23 the allocate and data copy operations duplicated in both (!) of the
26 JH/02 Disallow '/' characters in queue names specified for the "queue=" ACL
27 modifier. This matches the restriction on the commandline.
29 JH/03 Fix pgsql lookup for multiple result-tuples with a single column.
30 Previously only the last row was returned.
32 JH/04 Bug 2217: Tighten up the parsing of DKIM signature headers. Previously
33 we assumed that tags in the header were well-formed, and parsed the
34 element content after inspecting only the first char of the tag.
35 Assumptions at that stage could crash the receive process on malformed
38 JH/05 Bug 2215: Fix crash associated with dnsdb lookup done from DKIM ACL.
39 While running the DKIM ACL we operate on the Permanent memory pool so that
40 variables created with "set" persist to the DATA ACL. Also (at any time)
41 DNS lookups that fail create cache records using the Permanent pool. But
42 expansions release any allocations made on the current pool - so a dnsdb
43 lookup expansion done in the DKIM ACL releases the memory used for the
44 DNS negative-cache, and bad things result. Solution is to switch to the
45 Main pool for expansions.
46 While we're in that code, add checks on the DNS cache during store_reset,
47 active in the testsuite.
48 Problem spotted, and debugging aided, by Wolfgang Breyha.
50 JH/06 Fix issue with continued-connections when the DNS shifts unreliably.
51 When none of the hosts presented to a transport match an already-open
52 connection, close it and proceed with the list. Previously we would
53 queue the message. Spotted by Lena with Yahoo, probably involving
56 JH/07 Bug 2214: Fix SMTP responses resulting from non-accept result of MIME ACL.
57 Previously a spurious "250 OK id=" response was appended to the proper
60 JH/08 The "support for" informational output now, which built with Content
61 Scanning support, has a line for the malware scanner interfaces compiled
62 in. Interface can be individually included or not at build time.
64 JH/09 The "aveserver", "kavdaemon" and "mksd" interfaces are now not included
65 by the template makefile "src/EDITME". The "STREAM" support for an older
66 ClamAV interface method is removed.
68 JH/10 Bug 2223: Fix mysql lookup returns for the no-data case (when the number of
69 rows affected is given instead).
71 JH/11 The runtime Berkeley DB library version is now additionally output by
72 "exim -d -bV". Previously only the compile-time version was shown.
74 JH/12 Bug 2230: Fix cutthrough routing for nonfirst messages in an initiating
75 SMTP connection. Previously, when one had more receipients than the
76 first, an abortive onward connection was made. Move to full support for
77 multiple onward connections in sequence, handling cutthrough connection
78 for all multi-message initiating connections.
80 JH/13 Bug 2229: Fix cutthrough routing for nonstandard port numbers defined by
81 routers. Previously, a multi-recipient message would fail to match the
82 onward-connection opened for the first recipient, and cause its closure.
84 JH/14 Bug 2174: A timeout on connect for a callout was also erroneously seen as
85 a timeout on read on a GnuTLS initiating connection, resulting in the
86 initiating connection being dropped. This mattered most when the callout
87 was marked defer_ok. Fix to keep the two timeout-detection methods
90 JH/15 Relax results from ACL control request to enable cutthrough, in
91 unsupported situations, from error to silently (except under debug)
92 ignoring. This covers use with PRDR, frozen messages, queue-only and
95 HS/01 Fix Buffer overflow in base64d() (CVE-2018-6789)
97 JH/16 Fix bug in DKIM verify: a buffer overflow could corrupt the malloc
98 metadata, resulting in a crash in free().
100 PP/01 Fix broken Heimdal GSSAPI authenticator integration.
101 Broken in f2ed27cf5, missing an equals sign for specified-initialisers.
102 Broken also in d185889f4, with init system revamp.
104 JH/17 Bug 2113: Fix conversation closedown with the Avast malware scanner.
105 Previously we abruptly closed the connection after reading a malware-
106 found indication; now we go on to read the "scan ok" response line,
109 JH/18 Bug 2239: Enforce non-usability of control=utf8_downconvert in the mail
110 ACL. Previously, a crash would result.
112 JH/19 Speed up macro lookups during configuration file read, by skipping non-
113 macro text after a replacement (previously it was only once per line) and
114 by skipping builtin macros when searching for an uppercase lead character.
116 JH/20 DANE support moved from Experimental to mainline. The Makefile control
117 for the build is renamed.
119 JH/21 Fix memory leak during multi-message connections using STARTTLS. A buffer
120 was allocated for every new TLS startup, meaning one per message. Fix
121 by only allocating once (OpenSSL) or freeing on TLS-close (GnuTLS).
123 JH/22 Bug 2236: When a DKIM verification result is overridden by ACL, DMARC
124 reported the original. Fix to report (as far as possible) the ACL
125 result replacing the original.
127 JH/23 Fix memory leak during multi-message connections using STARTTLS under
128 OpenSSL. Certificate information is loaded for every new TLS startup,
129 and the resources needed to be freed.
131 JH/24 Bug 2242: Fix exim_dbmbuild to permit directoryless filenames.
133 JH/25 Fix utf8_downconvert propagation through a redirect router. Previously it
136 JH/26 Bug 2253: For logging delivery lines under PRDR, append the overall
137 DATA response info to the (existing) per-recipient response info for
138 the "C=" log element. It can have useful tracking info from the
139 destination system. Patch from Simon Arlott.
141 JH/27 Bug 2251: Fix ldap lookups that return a single attribute having zero-
142 length value. Previously this would segfault.
144 HS/02 Support Avast multiline protoocol, this allows passing flags to
145 newer versions of the scanner.
147 JH/28 Ensure that variables possibly set during message acceptance are marked
148 dead before release of memory in the daemon loop. This stops complaints
149 about them when the debug_store option is enabled. Discovered specifically
150 for sender_rate_period, but applies to a whole set of variables.
151 Do the same for the queue-runner and queue-list loops, for variables set
152 from spool message files. Do the same for the SMTP per-message loop, for
153 certain variables indirectly set in ACL operations.
155 JH/29 Bug 2250: Fix a longstanding bug in heavily-pipelined SMTP input (such
156 as a multi-recipient message from a mailinglist manager). The coding had
157 an arbitrary cutoff number of characters while checking for more input;
158 enforced by writing a NUL into the buffer. This corrupted long / fast
159 input. The problem was exposed more widely when more pipelineing of SMTP
160 responses was introduced, and one Exim system was feeding another.
161 The symptom is log complaints of SMTP syntax error (NUL chars) on the
162 receiving system, and refused recipients seen by the sending system
163 (propating to people being dropped from mailing lists).
164 Discovered and pinpointed by David Carter.
166 JH/30 The (EXPERIMENTAL_DMARC) variable $dmarc_ar_header is withdrawn, being
167 replaced by the ${authresults } expansion.
169 JH/31 Bug 2257: Fix pipe transport to not use a socket-only syscall.
171 HS/03 Set a handler for SIGTERM and call exit(3) if running as PID 1. This
172 allows proper process termination in container environments.
174 JH/32 Bug 2258: Fix spool_wireformat in combination with LMTP transport.
175 Previously the "final dot" had a newline after it; ensure it is CR,LF.
177 JH/33 SPF: remove support for the "spf" ACL condition outcome values "err_temp"
178 and "err_perm", deprecated since 4.83 when the RFC-defined words
179 "temperror" and "permerror" were introduced.
181 JH/34 Re-introduce enforcement of no cutthrough delivery on transports having
182 transport-filters or DKIM-signing. The restriction was lost in the
183 consolidation of verify-callout and delivery SMTP handling.
184 Extend the restriction to also cover ARC-signing.
186 JH/35 Cutthrough: for a final-dot response timeout (and nonunderstood responses)
187 in defer=pass mode supply a 450 to the initiator. Previously the message
190 PP/02 DANE: add dane_require_tls_ciphers SMTP Transport option; if unset,
191 tls_require_ciphers is used as before.
193 HS/03 Malware Avast: Better match the Avast multiline protocol. Add
194 "pass_unscanned". Only tmpfails from the scanner are written to
195 the paniclog, as they may require admin intervention (permission
196 denied, license issues). Other scanner errors (like decompression
197 bombs) do not cause a paniclog entry.
199 JH/36 Fix reinitialisation of DKIM logging variable between messages.
200 Previously it was possible to log spurious information in receive log
203 JH/37 Bug 2255: Revert the disable of the OpenSSL session caching. This
204 triggered odd behaviour from Outlook Express clients.
210 JH/01 Rework error string handling in TLS interface so that the caller in
211 more cases is responsible for logging. This permits library-sourced
212 string to be attached to addresses during delivery, and collapses
213 pairs of long lines into single ones.
215 PP/01 Allow PKG_CONFIG_PATH to be set in Local/Makefile and use it correctly
216 during configuration. Wildcards are allowed and expanded.
218 JH/02 Rework error string handling in DKIM to pass more info back to callers.
219 This permits better logging.
221 JH/03 Rework the transport continued-connection mechanism: when TLS is active,
222 do not close it down and have the child transport start it up again on
223 the passed-on TCP connection. Instead, proxy the child (and any
224 subsequent ones) for TLS via a unix-domain socket channel. Logging is
225 affected: the continued delivery log lines do not have any DNSSEC, TLS
226 Certificate or OCSP information. TLS cipher information is still logged.
228 JH/04 Shorten the log line for daemon startup by collapsing adjacent sets of
229 identical IP addresses on different listening ports. Will also affect
232 PP/02 Bug 2070: uClibc defines __GLIBC__ without providing glibc headers;
233 add noisy ifdef guards to special-case this sillyness.
234 Patch from Bernd Kuhls.
236 JH/05 Tighten up the checking in isip4 (et al): dotted-quad components larger
237 than 255 are no longer allowed.
239 JH/06 Default openssl_options to include +no_ticket, to reduce load on peers.
240 Disable the session-cache too, which might reduce our load. Since we
241 currrectly use a new context for every connection, both as server and
242 client, there is no benefit for these.
243 GnuTLS appears to not support tickets server-side by default (we don't
244 call gnutls_session_ticket_enable_server()) but client side is enabled
245 by default on recent versions (3.1.3 +) unless the PFS priority string
248 PP/03 Add $SOURCE_DATE_EPOCH support for reproducible builds, per spec at
249 <https://reproducible-builds.org/specs/source-date-epoch/>.
251 JH/07 Fix smtp transport use of limited max_rcpt under mua_wrapper. Previously
252 the check for any unsuccessful recipients did not notice the limit, and
253 erroneously found still-pending ones.
255 JH/08 Pipeline CHUNKING command and data together, on kernels that support
256 MSG_MORE. Only in-clear (not on TLS connections).
258 JH/09 Avoid using a temporary file during transport using dkim. Unless a
259 transport-filter is involved we can buffer the headers in memory for
260 creating the signature, and read the spool data file once for the
261 signature and again for transmission.
263 JH/10 Enable use of sendfile in Linux builds as default. It was disabled in
264 4.77 as the kernel support then wasn't solid, having issues in 64bit
265 mode. Now, it's been long enough. Add support for FreeBSD also.
267 JH/11 Bug 2104: Fix continued use of a transport connection with TLS. In the
268 case where the routing stage had gathered several addresses to send to
269 a host before calling the transport for the first, we previously failed
270 to close down TLS in the old transport process before passing the TCP
271 connection to the new process. The new one sent a STARTTLS command
272 which naturally failed, giving a failed delivery and bloating the retry
273 database. Investigation and fix prototype from Wolfgang Breyha.
275 JH/12 Fix check on SMTP command input synchronisation. Previously there were
276 false-negatives in the check that the sender had not preempted a response
277 or prompt from Exim (running as a server), due to that code's lack of
278 awareness of the SMTP input buffering.
280 PP/04 Add commandline_checks_require_admin option.
281 Exim drops privileges sanely, various checks such as -be aren't a
282 security problem, as long as you trust local users with access to their
283 own account. When invoked by services which pass untrusted data to
284 Exim, this might be an issue. Set this option in main configuration
285 AND make fixes to the calling application, such as using `--` to stop
288 JH/13 Do pipelining under TLS. Previously, although safe, no advantage was
289 taken. Now take care to pack both (client) MAIL,RCPT,DATA, and (server)
290 responses to those, into a single TLS record each way (this usually means
291 a single packet). As a side issue, smtp_enforce_sync now works on TLS
294 PP/05 OpenSSL/1.1: use DH_bits() for more accurate DH param sizes. This
295 affects you only if you're dancing at the edge of the param size limits.
296 If you are, and this message makes sense to you, then: raise the
297 configured limit or use OpenSSL 1.1. Nothing we can do for older
300 JH/14 For the "sock" variant of the malware scanner interface, accept an empty
301 cmdline element to get the documented default one. Previously it was
304 JH/15 Fix a crash in the smtp transport caused when two hosts in succession
305 are unsuable for non-message-specific reasons - eg. connection timeout,
306 banner-time rejection.
308 JH/16 Fix logging of delivery remote port, when specified by router, under
311 PP/06 Repair manualroute's ability to take options in any order, even if one
312 is the name of a transport.
315 HS/01 Cleanup, prevent repeated use of -p/-oMr (CVE-2017-1000369)
317 JH/17 Change the list-building routines interface to use the expanding-string
318 triplet model, for better allocation and copying behaviour.
320 JH/18 Prebuild the data-structure for "builtin" macros, for faster startup.
321 Previously it was constructed the first time a possibly-matching string
322 was met in the configuration file input during startup; now it is done
325 JH/19 Bug 2141: Use the full-complex API for Berkeley DB rather than the legacy-
326 compatible one, to avoid the (poorly documented) possibility of a config
327 file in the working directory redirecting the DB files, possibly correpting
328 some existing file. CVE-2017-10140 assigned for BDB.
330 JH/20 Bug 2147: Do not defer for a verify-with-callout-and-random which is not
331 cache-hot. Previously, although the result was properly cached, the
332 initial verify call returned a defer.
334 JH/21 Bug 2151: Avoid using SIZE on the MAIL for a callout verify, on any but
335 the main verify for receipient in uncached-mode.
337 JH/22 Retire historical build files to an "unsupported" subdir. These are
338 defined as "ones for which we have no current evidence of testing".
340 JH/23 DKIM: enforce the DNS pubkey record "h" permitted-hashes optional field,
341 if present. Previously it was ignored.
343 JH/24 Start using specified-initialisers in C structure init coding. This is
344 a C99 feature (it's 2017, so now considered safe).
346 JH/25 Use one-bit bitfields for flags in the "addr" data structure. Previously
347 if was a fixed-sized field and bitmask ops via macros; it is now more
350 PP/07 GitHub PR 56: Apply MariaDB build fix.
351 Patch provided by Jaroslav Škarvada.
353 PP/08 Bug 2161: Fix regression in sieve quoted-printable handling introduced
354 during Coverity cleanups [4.87 JH/47]
355 Diagnosis and fix provided by Michael Fischer v. Mollard.
357 JH/26 Fix DKIM bug: when the pseudoheader generated for signing was exactly
358 the right size to place the terminating semicolon on its own folded
359 line, the header hash was calculated to an incorrect value thanks to
360 the (relaxed) space the fold became.
362 HS/02 Fix Bug 2130: large writes from the transport subprocess where chunked
363 and confused the parent.
365 JH/27 Fix SOCKS bug: an unitialized pointer was deref'd by the transport process
366 which could crash as a result. This could lead to undeliverable messages.
368 JH/28 Logging: "next input sent too soon" now shows where input was truncated
371 JH/29 Fix queue_run_in_order to ignore the PID portion of the message ID. This
372 matters on fast-turnover and PID-randomising systems, which were getting
373 out-of-order delivery.
375 JH/30 Fix a logging bug on aarch64: an unsafe routine was previously used for
376 a possibly-overlapping copy. The symptom was that "Remote host closed
377 connection in response to HELO" was logged instead of the actual 4xx
380 JH/31 Fix CHUNKING code to properly flush the unwanted chunk after an error.
381 Previously only that bufferd was discarded, resulting in SYMTP command
384 JH/32 DKIM: when a message has multiple signatures matching an identity given
385 in dkim_verify_signers, run the dkim acl once for each. Previously only
386 one run was done. Bug 2189.
388 JH/33 Downgrade an unfound-list name (usually a typo in the config file) from
389 "panic the current process" to "deliberately defer". The panic log is
390 still written with the problem list name; the mail and reject logs now
391 get a temp-reject line for the message that was being handled, saying
392 something like "domains check lookup or other defer". The SMTP 451
393 message is still "Temporary local problem".
395 JH/34 Bug 2199: Fix a use-after-free while reading smtp input for header lines.
396 A crafted sequence of BDAT commands could result in in-use memory beeing
397 freed. CVE-2017-16943.
399 HS/03 Bug 2201: Fix checking for leading-dot on a line during headers reading
400 from SMTP input. Previously it was always done; now only done for DATA
401 and not BDAT commands. CVE-2017-16944.
403 JH/35 Bug 2201: Flush received data in BDAT mode after detecting an error fatal
404 to the message (such as an overlong header line). Previously this was
405 not done and we did not exit BDAT mode. Followon from the previous item
406 though a different problem.
412 JH/01 Bug 1922: Support IDNA2008. This has slightly different conversion rules
413 than -2003 did; needs libidn2 in addition to libidn.
415 JH/02 The path option on a pipe transport is now expanded before use.
417 PP/01 GitHub PR 50: Do not call ldap_start_tls_s on ldapi:// connections.
418 Patch provided by "Björn", documentation fix added too.
420 JH/03 Bug 2003: fix Proxy Protocol v2 handling: the address size field was
421 missing a wire-to-host endian conversion.
423 JH/04 Bug 2004: fix CHUNKING in non-PIPELINEING mode. Chunk data following
424 close after a BDAT command line could be taken as a following command,
425 giving a synch failure. Fix by only checking for synch immediately
426 before acknowledging the chunk.
428 PP/02 GitHub PR 52: many spelling fixes, which include fixing parsing of
429 no_require_dnssec option and creation of _HAVE_TRANSPORT_APPEND_MAILDIR
430 macro. Patches provided by Josh Soref.
432 JH/05 Have the EHLO response advertise VRFY, if there is a vrfy ACL defined.
433 Previously we did not; the RFC seems ambiguous and VRFY is not listed
434 by IANA as a service extension. However, John Klensin suggests that we
437 JH/06 Bug 2017: Fix DKIM verification in -bh test mode. The data feed into
438 the dkim code may be unix-mode line endings rather than smtp wire-format
439 CRLF, so prepend a CR to any bare LF.
441 JH/07 Rationalise the coding for callout smtp conversations and transport ones.
442 As a side-benfit, callouts can now use PIPELINING hence fewer round-trips.
444 JH/08 Bug 2016: Fix DKIM verification vs. CHUNKING. Any BDAT commands after
445 the first were themselves being wrongly included in the feed into dkim
446 processing; with most chunk sizes in use this resulted in an incorrect
447 body hash calculated value.
449 JH/09 Bug 2014: permit inclusion of a DKIM-Signature header in a received
450 DKIM signature block, for verification. Although advised against by
451 standards it is specifically not ruled illegal.
453 JH/10 Bug 2025: Fix reception of (quoted) local-parts with embedded spaces.
455 JH/11 Bug 2029: Fix crash in DKIM verification when a message signature block is
456 missing a body hash (the bh= tag).
458 JH/12 Bug 2018: Re-order Proxy Protocol startup versus TLS-on-connect startup.
459 It seems that HAProxy sends the Proxy Protocol information in clear and
460 only then does a TLS startup, so do the same.
462 JH/13 Bug 2027: Avoid attempting to use TCP Fast Open for non-transport client
463 TCP connections (such as for Spamd) unless the daemon successfully set
464 Fast Open mode on its listening sockets. This fixes breakage seen on
465 too-old kernels or those not configured for Fast Open, at the cost of
466 requiring both directions being enabled for TFO, and TFO never being used
467 by non-daemon-related Exim processes.
469 JH/14 Bug 2000: Reject messages recieved with CHUNKING but with malformed line
470 endings, at least on the first header line. Try to canonify any that get
471 past that check, despite the cost.
473 JH/15 Angle-bracket nesting (an error inserted by broken sendmails) levels are
474 now limited to an arbitrary five deep, while parsing addresses with the
475 strip_excess_angle_brackets option enabled.
477 PP/03 Bug 2018: For Proxy Protocol and TLS-on-connect, do not over-read and
478 instead leave the unprompted TLS handshake in socket buffer for the
479 TLS library to consume.
481 PP/04 Bug 2018: Also handle Proxy Protocol v2 safely.
483 PP/05 FreeBSD compat: handle that Ports no longer create /usr/bin/perl
485 JH/16 Drop variables when they go out of scope. Memory management drops a whole
486 region in one operation, for speed, and this leaves assigned pointers
487 dangling. Add checks run only under the testsuite which checks all
488 variables at a store-reset and panics on a dangling pointer; add code
489 explicitly nulling out all the variables discovered. Fixes one known
490 bug: a transport crash, where a dangling pointer for $sending_ip_address
491 originally assigned in a verify callout, is re-used.
493 PP/06 Drop '.' from @INC in various Perl scripts.
495 PP/07 Switch FreeBSD iconv to always use the base-system libc functions.
497 PP/08 Reduce a number of compilation warnings under clang; building with
498 CC=clang CFLAGS+=-Wno-dangling-else -Wno-logical-op-parentheses
499 should be warning-free.
501 JH/17 Fix inbound CHUNKING when DKIM disabled at runtime.
503 HS/01 Fix portability problems introduced by PP/08 for platforms where
504 realloc(NULL) is not equivalent to malloc() [SunOS et al].
506 HS/02 Bug 1974: Fix missing line terminator on the last received BDAT
507 chunk. This allows us to accept broken chunked messages. We need a more
508 general solution here.
510 PP/09 Wrote util/chunking_fixqueue_finalnewlines.pl to help recover
511 already-broken messages in the queue.
513 JH/18 Bug 2061: Fix ${extract } corrupting an enclosing ${reduce } $value.
515 JH/19 Fix reference counting bug in routing-generated-address tracking.
521 JH/01 Use SIZE on MAIL FROM in a cutthrough connection, if the destination
522 supports it and a size is available (ie. the sending peer gave us one).
524 JH/02 The obsolete acl condition "demime" is removed (finally, after ten
525 years of being deprecated). The replacements are the ACLs
526 acl_smtp_mime and acl_not_smtp_mime.
528 JH/03 Upgrade security requirements imposed for hosts_try_dane: previously
529 a downgraded non-dane trust-anchor for the TLS connection (CA-style)
530 or even an in-clear connection were permitted. Now, if the host lookup
531 was dnssec and dane was requested then the host is only used if the
532 TLSA lookup succeeds and is dnssec. Further hosts (eg. lower priority
533 MXs) will be tried (for hosts_try_dane though not for hosts_require_dane)
534 if one fails this test.
535 This means that a poorly-configured remote DNS will make it incommunicado;
536 but it protects against a DNS-interception attack on it.
538 JH/04 Bug 1810: make continued-use of an open smtp transport connection
539 non-noisy when a race steals the message being considered.
541 JH/05 If main configuration option tls_certificate is unset, generate a
542 self-signed certificate for inbound TLS connections.
544 JH/06 Bug 165: hide more cases of password exposure - this time in expansions
545 in rewrites and routers.
547 JH/07 Retire gnutls_require_mac et.al. These were nonfunctional since 4.80
548 and logged a warning sing 4.83; now they are a configuration file error.
550 JH/08 Bug 1836: Fix crash in VRFY handling when handed an unqualified name
551 (lacking @domain). Apply the same qualification processing as RCPT.
553 JH/09 Bug 1804: Avoid writing msglog files when in -bh or -bhc mode.
555 JH/10 Support ${sha256:} applied to a string (as well as the previous
558 JH/11 Cutthrough: avoid using the callout hints db on a verify callout when
559 a cutthrough deliver is pending, as we always want to make a connection.
560 This also avoids re-routing the message when later placing the cutthrough
561 connection after a verify cache hit.
562 Do not update it with the verify result either.
564 JH/12 Cutthrough: disable when verify option success_on_redirect is used, and
565 when routing results in more than one destination address.
567 JH/13 Cutthrough: expand transport dkim_domain option when testing for dkim
568 signing (which inhibits the cutthrough capability). Previously only
569 the presence of an option was tested; now an expansion evaluating as
570 empty is permissible (obviously it should depend only on data available
571 when the cutthrough connection is made).
573 JH/14 Fix logging of errors under PIPELINING. Previously the log line giving
574 the relevant preceding SMTP command did not note the pipelining mode.
576 JH/15 Fix counting of empty lines in $body_linecount and $message_linecount.
577 Previously they were not counted.
579 JH/16 DANE: treat a TLSA lookup response having all non-TLSA RRs, the same
580 as one having no matching records. Previously we deferred the message
581 that needed the lookup.
583 JH/17 Fakereject: previously logged as a normal message arrival "<="; now
584 distinguished as "(=".
586 JH/18 Bug 1867: make the fail_defer_domains option on a dnslookup router work
587 for missing MX records. Previously it only worked for missing A records.
589 JH/19 Bug 1850: support Radius libraries that return REJECT_RC.
591 JH/20 Bug 1872: Ensure that acl_smtp_notquit is run when the connection drops
592 after the data-go-ahead and data-ack. Patch from Jason Betts.
594 JH/21 Bug 1846: Send DMARC forensic reports for reject and quarantine results,
595 even for a "none" policy. Patch from Tony Meyer.
597 JH/22 Fix continued use of a connection for further deliveries. If a port was
598 specified by a router, it must also match for the delivery to be
601 JH/23 Bug 1874: fix continued use of a connection for further deliveries.
602 When one of the recipients of a message was unsuitable for the connection
603 (has no matching addresses), we lost track of needing to mark it
604 deferred. As a result mail would be lost.
606 JH/24 Bug 1832: Log EHLO response on getting conn-close response for HELO.
608 JH/25 Decoding ACL controls is now done using a binary search; the source code
609 takes up less space and should be simpler to maintain. Merge the ACL
610 condition decode tables also, with similar effect.
612 JH/26 Fix problem with one_time used on a redirect router which returned the
613 parent address unchanged. A retry would see the parent address marked as
614 delivered, so not attempt the (identical) child. As a result mail would
617 JH/27 Fix a possible security hole, wherein a process operating with the Exim
618 UID can gain a root shell. Credit to http://www.halfdog.net/ for
619 discovery and writeup. Ubuntu bug 1580454; no bug raised against Exim
622 JH/28 Enable {spool,log} filesystem space and inode checks as default.
623 Main config options check_{log,spool}_{inodes,space} are now
624 100 inodes, 10MB unless set otherwise in the configuration.
626 JH/29 Fix the connection_reject log selector to apply to the connect ACL.
627 Previously it only applied to the main-section connection policy
630 JH/30 Bug 1897: fix callouts connection fallback from TLS to cleartext.
632 PP/01 Changed default Diffie-Hellman parameters to be Exim-specific, created
633 by me. Added RFC7919 DH primes as an alternative.
635 PP/02 Unbreak build via pkg-config with new hash support when crypto headers
636 are not in the system include path.
638 JH/31 Fix longstanding bug with aborted TLS server connection handling. Under
639 GnuTLS, when a session startup failed (eg because the client disconnected)
640 Exim did stdio operations after fclose. This was exposed by a recent
641 change which nulled out the file handle after the fclose.
643 JH/32 Bug 1909: Fix OCSP proof verification for cases where the proof is
644 signed directly by the cert-signing cert, rather than an intermediate
645 OCSP-signing cert. This is the model used by LetsEncrypt.
647 JH/33 Bug 1914: Ensure socket is nonblocking before draining after SMTP QUIT.
649 HS/01 Fix leak in verify callout under GnuTLS, about 3MB per recipient on
650 an incoming connection.
652 HS/02 Bug 1802: Do not half-close the connection after sending a request
655 HS/03 Use "auto" as the default EC curve parameter. For OpenSSL < 1.0.2
656 fallback to "prime256v1".
658 JH/34 SECURITY: Use proper copy of DATA command in error message.
659 Could leak key material. Remotely exploitable. CVE-2016-9963.
665 JH/01 Bug 1664: Disable OCSP for GnuTLS library versions at/before 3.3.16
666 and 3.4.4 - once the server is enabled to respond to an OCSP request
667 it does even when not requested, resulting in a stapling non-aware
668 client dropping the TLS connection.
670 TF/01 Code cleanup: Overhaul the debug_selector and log_selector machinery to
671 support variable-length bit vectors. No functional change.
673 TF/02 Improve the consistency of logging incoming and outgoing interfaces.
674 The I= interface field on outgoing lines is now after the H= remote
675 host field, same as incoming lines. There is a separate
676 outgoing_interface log selector which allows you to disable the
679 JH/02 Bug 728: Close logfiles after a daemon-process "exceptional" log write.
680 If not running log_selector +smtp_connection the mainlog would be held
681 open indefinitely after a "too many connections" event, including to a
682 deleted file after a log rotate. Leave the per net connection logging
683 leaving it open for efficiency as that will be quickly detected by the
684 check on the next write.
686 HS/01 Bug 1671: Fix post transport crash.
687 Processing the wait-<transport> messages could crash the delivery
688 process if the message IDs didn't exist for some reason. When
689 using 'split_spool_directory=yes' the construction of the spool
690 file name failed already, exposing the same netto behaviour.
692 JH/03 Bug 425: Capture substrings in $regex1, $regex2 etc from regex &
693 mime_regex ACL conditions.
695 JH/04 Bug 1686: When compiled with EXPERIMENTAL_DSN_INFO: Add extra information
696 to DSN fail messages (bounces): remote IP, remote greeting, remote response
697 to HELO, local diagnostic string.
699 JH/05 Downgrade message for a TLS-certificate-based authentication fail from
700 log line to debug. Even when configured with a tls authenticator many
701 client connections are expected to not authenticate in this way, so
702 an authenticate fail is not an error.
704 HS/02 Add the Exim version string to the process info. This way exiwhat
705 gives some more detail about the running daemon.
707 JH/06 Bug 1395: time-limit caching of DNS lookups, to the TTL value. This may
708 matter for fast-change records such as DNSBLs.
710 JH/07 Bug 1678: Always record an interface option value, if set, as part of a
711 retry record, even if constant. There may be multiple transports with
712 different interface settings and the retry behaviour needs to be kept
715 JH/08 Bug 1586: exiqgrep now refuses to run if there are unexpected arguments.
717 JH/09 Bug 1700: ignore space & tab embedded in base64 during decode.
719 JH/10 Bug 840: fix log_defer_output option of pipe transport
721 JH/11 Bug 830: use same host for all RCPTS of a message, even under
722 hosts_randomize. This matters a lot when combined with mua_wrapper.
724 JH/12 Bug 1706: percent and underbar characters are no longer escaped by the
725 ${quote_pgsql:<string>} operator.
727 JH/13 Bug 1708: avoid misaligned access in cached lookup.
729 JH/14 Change header file name for freeradius-client. Relevant if compiling
730 with Radius support; from the Gentoo tree and checked under Fedora.
732 JH/15 Bug 1712: Introduce $prdr_requested flag variable
734 JH/16 Bug 1714: Permit an empty string as expansion result for transport
735 option transport_filter, meaning no filtering.
737 JH/17 Bug 1713: Fix non-PDKIM_DEBUG build. Patch from Jasen Betts.
739 JH/18 Bug 1709: When built with TLS support, the tls_advertise_hosts option now
740 defaults to "*" (all hosts). The variable is now available when not built
741 with TLS, default unset, mainly to enable keeping the testsuite sane.
742 If a server certificate is not supplied (via tls_certificate) an error is
743 logged, and clients will find TLS connections fail on startup. Presumably
744 they will retry in-clear.
745 Packagers of Exim are strongly encouraged to create a server certificate
746 at installation time.
748 HS/03 Add -bP config_file as a synonym for -bP configure_file, for consistency
749 with the $config_file variable.
751 JH/19 Two additional event types: msg:rcpt:defer and msg:rcpt:host:defer. Both
752 in transport context, after the attempt, and per-recipient. The latter type
753 is per host attempted. The event data is the error message, and the errno
754 information encodes the lookup type (A vs. MX) used for the (first) host,
755 and the trailing two digits of the smtp 4xx response.
757 GF/01 Bug 1715: Fix for race condition in exicyclog, where exim could attempt
758 to write to mainlog (or rejectlog, paniclog) in the window between file
759 creation and permissions/ownership being changed. Particularly affects
760 installations where exicyclog is run as root, rather than exim user;
761 result is that the running daemon panics and dies.
763 JH/20 Bug 1701: For MySQL lookups, support MySQL config file option group names.
765 JH/21 Bug 1720: Add support for priority groups and weighted-random proxy
766 selection for the EXPERIMENTAL_SOCKS feature, via new per-proxy options
767 "pri" and "weight". Note that the previous implicit priority given by the
768 list order is no longer honoured.
770 JH/22 Bugs 963, 1721: Fix some corner cases in message body canonicalization
773 JH/23 Move SOCKS5 support from Experimental to mainline, enabled for a build
774 by defining SUPPORT_SOCKS.
776 JH/26 Move PROXY support from Experimental to mainline, enabled for a build
777 by defining SUPPORT_PROXY. Note that the proxy_required_hosts option
778 is renamed to hosts_proxy, and the proxy_{host,target}_{address,port}.
779 variables are renamed to proxy_{local,external}_{address,port}.
781 JH/27 Move Internationalisation support from Experimental to mainline, enabled
782 for a build by defining SUPPORT_I18N
784 JH/28 Bug 1745: Fix redis lookups to handle (quoted) spaces embedded in parts
785 of the query string, and make ${quote_redis:} do that quoting.
787 JH/29 Move Events support from Experimental to mainline, enabled by default
788 and removable for a build by defining DISABLE_EVENT.
790 JH/30 Updated DANE implementation code to current from Viktor Dukhovni.
792 JH/31 Fix bug with hosts_connection_nolog and named-lists which were wrongly
793 cached by the daemon.
795 JH/32 Move Redis support from Experimental to mainline, enabled for a build
796 by defining LOOKUP_REDIS. The libhiredis library is required.
798 JH/33 Bug 1748: Permit ACL dnslists= condition in non-smtp ACLs if explicit
799 keys are given for lookup.
801 JH/34 Bug 1192: replace the embedded copy of PolarSSL RSA routines in the DKIM
802 support, by using OpenSSL or GnuTLS library ones. This means DKIM is
803 only supported when built with TLS support. The PolarSSL SHA routines
804 are still used when the TLS library is too old for convenient support.
806 JH/35 Require SINGLE_DH_USE by default in OpenSSL (main config option
807 openssl_options), for security. OpenSSL forces this from version 1.1.0
808 server-side so match that on older versions.
810 JH/36 Bug 1778: longstanding bug in memory use by the ${run } expansion: A fresh
811 allocation for $value could be released as the expansion processing
812 concluded, but leaving the global pointer active for it.
814 JH/37 Bug 1769: Permit a VRFY ACL to override the default 252 response,
815 and to use the domains and local_parts ACL conditions.
817 JH/38 Fix cutthrough bug with body lines having a single dot. The dot was
818 incorrectly not doubled on cutthrough transmission, hence seen as a
819 body-termination at the receiving system - resulting in truncated mails.
820 Commonly the sender saw a TCP-level error, and retransmitted the message
821 via the normal store-and-forward channel. This could result in duplicates
822 received - but deduplicating mailstores were liable to retain only the
823 initial truncated version.
825 JH/39 Bug 1781: Fix use of DKIM private-keys having trailing '=' in the base-64.
827 JH/40 Fix crash in queryprogram router when compiled with EXPERIMENTAL_SRS.
829 JH/41 Bug 1792: Fix selection of headers to sign for DKIM: bottom-up. While
830 we're in there, support oversigning also; bug 1309.
832 JH/42 Bug 1796: Fix error logged on a malware scanner connection failure.
834 HS/04 Add support for keep_environment and add_environment options.
836 JH/43 Tidy coding issues detected by gcc --fsanitize=undefined. Some remain;
837 either intentional arithmetic overflow during PRNG, or testing config-
840 JH/44 Bug 1800: The combination of a -bhc commandline option and cutthrough
841 delivery resulted in actual delivery. Cancel cutthrough before DATA
844 JH/45 Fix cutthrough, when connection not opened by verify and target hard-
845 rejects a recipient: pass the reject to the originator.
847 JH/46 Multiple issues raised by Coverity. Some were obvious or plausible bugs.
848 Many were false-positives and ignorable, but it's worth fixing the
851 JH/47 Fix build on HP-UX and older Solaris, which need (un)setenv now also
852 for the new environment-manipulation done at startup. Move the routines
853 from being local to tls.c to being global via the os.c file.
855 JH/48 Bug 1807: Fix ${extract } for the numeric/3-string case. While preparsing
856 an extract embedded as result-arg for a map, the first arg for extract
857 is unavailable so we cannot tell if this is a numbered or keyed
858 extraction. Accept either.
864 JH/01 Bug 1545: The smtp transport option "retry_include_ip_address" is now
867 JH/02 The smtp transport option "multi_domain" is now expanded.
869 JH/03 The smtp transport now requests PRDR by default, if the server offers
872 JH/04 Certificate name checking on server certificates, when exim is a client,
873 is now done by default. The transport option tls_verify_cert_hostnames
874 can be used to disable this per-host. The build option
875 EXPERIMENTAL_CERTNAMES is withdrawn.
877 JH/05 The value of the tls_verify_certificates smtp transport and main options
878 default to the word "system" to access the system default CA bundle.
879 For GnuTLS, only version 3.0.20 or later.
881 JH/06 Verification of the server certificate for a TLS connection is now tried
882 (but not required) by default. The verification status is now logged by
883 default, for both outbound TLS and client-certificate supplying inbound
886 JH/07 Changed the default rfc1413 lookup settings to disable calls. Few
889 JH/08 The EXPERIMENTAL_DSN compile option is no longer needed; all Delivery
890 Status Notification (bounce) messages are now MIME format per RFC 3464.
891 Support for RFC 3461 DSN options NOTIFY,ENVID,RET,ORCPT can be advertised
892 under the control of the dsn_advertise_hosts option, and routers may
893 have a dsn_lasthop option.
895 JH/09 A timeout of 2 minutes is now applied to all malware scanner types by
896 default, modifiable by a malware= option. The list separator for
897 the options can now be changed in the usual way. Bug 68.
899 JH/10 The smtp_receive_timeout main option is now expanded before use.
901 JH/11 The incoming_interface log option now also enables logging of the
902 local interface on delivery outgoing connections.
904 JH/12 The cutthrough-routing facility now supports multi-recipient mails,
905 if the interface and destination host and port all match.
907 JH/13 Bug 344: The verify = reverse_host_lookup ACL condition now accepts a
910 JH/14 Bug 1573: The spam= ACL condition now additionally supports Rspamd.
911 Patch from Andrew Lewis.
913 JH/15 Bug 670: The spamd_address main option (for the spam= ACL condition)
914 now supports optional time-restrictions, weighting, and priority
915 modifiers per server. Patch originally by <rommer@active.by>.
917 JH/16 The spamd_address main option now supports a mixed list of local
918 and remote servers. Remote servers can be IPv6 addresses, and
919 specify a port-range.
921 JH/17 Bug 68: The spamd_address main option now supports an optional
922 timeout value per server.
924 JH/18 Bug 1581: Router and transport options headers_add/remove can
925 now have the list separator specified.
927 JH/19 Bug 392: spamd_address, and clamd av_scanner, now support retry
930 JH/20 Bug 1571: Ensure that $tls_in_peerdn is set, when verification fails
933 JH/21 Support for the A6 type of dns record is withdrawn.
935 JH/22 Bug 608: The result of a QUIT or not-QUIT toplevel ACL now matters
936 rather than the verbs used.
938 JH/23 Bug 1572: Increase limit on SMTP confirmation message copy size
939 from 255 to 1024 chars.
941 JH/24 Verification callouts now attempt to use TLS by default.
943 HS/01 DNSSEC options (dnssec_require_domains, dnssec_request_domains)
944 are generic router options now. The defaults didn't change.
946 JH/25 Bug 466: Add RFC2322 support for MIME attachment filenames.
947 Original patch from Alexander Shikoff, worked over by JH.
949 HS/02 Bug 1575: exigrep falls back to autodetection of compressed
950 files if ZCAT_COMMAND is not executable.
952 JH/26 Bug 1539: Add timeout/retry options on dnsdb lookups.
954 JH/27 Bug 286: Support SOA lookup in dnsdb lookups.
956 JH/28 Bug 1588: Do not use the A lookup following an AAAA for setting the FQDN.
957 Normally benign, it bites when the pair was led to by a CNAME;
958 modern usage is to not canonicalize the domain to a CNAME target
959 (and we were inconsistent anyway for A-only vs AAAA+A).
961 JH/29 Bug 1632: Removed the word "rejected" from line logged for ACL discards.
963 JH/30 Check the forward DNS lookup for DNSSEC, in addition to the reverse,
964 when evaluating $sender_host_dnssec.
966 JH/31 Check the HELO verification lookup for DNSSEC, adding new
967 $sender_helo_dnssec variable.
969 JH/32 Bug 1397: Enable ECDHE on OpenSSL, just the NIST P-256 curve.
971 JH/33 Bug 1346: Note MAIL cmd seen in -bS batch, to avoid smtp_no_mail log.
973 JH/34 Bug 1648: Fix a memory leak seen with "mailq" and large queues.
975 JH/35 Bug 1642: Fix support of $spam_ variables at delivery time. Was
976 documented as working, but never had. Support all but $spam_report.
978 JH/36 Bug 1659: Guard checking of input smtp commands again pseudo-command
979 added for tls authenticator.
981 HS/03 Add perl_taintmode main config option
987 TL/01 When running the test suite, the README says that variables such as
988 no_msglog_check are global and can be placed anywhere in a specific
989 test's script, however it was observed that placement needed to be near
990 the beginning for it to behave that way. Changed the runtest perl
991 script to read through the entire script once to detect and set these
992 variables, reset to the beginning of the script, and then run through
993 the script parsing/test process like normal.
995 TL/02 The BSD's have an arc4random API. One of the functions to induce
996 adding randomness was arc4random_stir(), but it has been removed in
997 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
998 function when detected.
1000 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
1001 cause callback expansion.
1003 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
1004 syntax errors in an expansion can be treated as a string instead of
1005 logging or causing an error, due to the internal use of bool_lax
1006 instead of bool when processing it.
1008 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
1009 server certificates when making smtp deliveries.
1011 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
1013 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
1015 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
1017 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
1018 Merged patch from Sebastian Wiedenroth.
1020 JH/05 Fix results-pipe from transport process. Several recipients, combined
1021 with certificate use, exposed issues where response data items split
1022 over buffer boundaries were not parsed properly. This eventually
1023 resulted in duplicates being sent. This issue only became common enough
1024 to notice due to the introduction of connection certificate information,
1025 the item size being so much larger. Found and fixed by Wolfgang Breyha.
1027 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
1028 size buffer was used, resulting in syntax errors when an expansion
1031 JH/07 Add support for directories of certificates when compiled with a GnuTLS
1032 version 3.3.6 or later.
1034 JH/08 Rename the TPDA experimental facility to Event Actions. The #ifdef
1035 is EXPERIMENTAL_EVENT, the main-configuration and transport options
1036 both become "event_action", the variables become $event_name, $event_data
1037 and $event_defer_errno. There is a new variable $verify_mode, usable in
1038 routers, transports and related events. The tls:cert event is now also
1039 raised for inbound connections, if the main configuration event_action
1042 TL/06 In test suite, disable OCSP for old versions of openssl which contained
1043 early OCSP support, but no stapling (appears to be less than 1.0.0).
1045 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
1046 server certificate names available under the smtp transport option
1047 "tls_verify_cert_hostname" now do not permit multi-component wildcard
1050 JH/10 Time-related extraction expansions from certificates now use the main
1051 option "timezone" setting for output formatting, and are consistent
1052 between OpenSSL and GnuTLS compilations. Bug 1541.
1054 JH/11 Fix a crash in mime ACL when meeting a zero-length, quoted or RFC2047-
1055 encoded parameter in the incoming message. Bug 1558.
1057 JH/12 Bug 1527: Autogrow buffer used in reading spool files. Since they now
1058 include certificate info, eximon was claiming there were spoolfile
1061 JH/13 Bug 1521: Fix ldap lookup for single-attr request, multiple-attr return.
1063 JH/14 Log delivery-related information more consistently, using the sequence
1064 "H=<name> [<ip>]" wherever possible.
1066 TL/07 Bug 1547: Omit RFCs from release. Draft and RFCs have licenses which
1067 are problematic for Debian distribution, omit them from the release
1070 JH/15 Updates and fixes to the EXPERIMENTAL_DSN feature.
1072 JH/16 Fix string representation of time values on 64bit time_t architectures.
1075 JH/17 Fix a null-indirection in certextract expansions when a nondefault
1076 output list separator was used.
1081 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
1082 checkers that were complaining about end of non-void function with no
1085 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
1086 This was a regression introduced in 4.83 by another bugfix.
1088 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
1090 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
1091 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
1097 TF/01 Correctly close the server side of TLS when forking for delivery.
1099 When a message was received over SMTP with TLS, Exim failed to clear up
1100 the incoming connection properly after forking off the child process to
1101 deliver the message. In some situations the subsequent outgoing
1102 delivery connection happened to have the same fd number as the incoming
1103 connection previously had. Exim would try to use TLS and fail, logging
1104 a "Bad file descriptor" error.
1106 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
1107 utilities have not been installed.
1109 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
1110 temporary space as the ACL may create new global variables.
1112 TL/01 LDAP support uses per connection or global context settings, depending
1113 upon the detected version of the libraries at build time.
1115 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
1116 to extract and use the src ip:port in logging and expansions as if it
1117 were a direct connection from the outside internet. PPv2 support was
1118 updated based on HAProxy spec change in May 2014.
1120 JH/02 Add ${listextract {number}{list}{success}{fail}}.
1122 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
1123 Properly escape header and check for NULL return.
1125 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
1128 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
1130 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
1131 characters in header names, implemented as a verify condition.
1132 Contributed by Michael Fischer v. Mollard.
1134 TL/05 Rename SPF condition results err_perm and err_temp to standardized
1135 results permerror and temperror. Previous values are deprecated but
1136 still accepted. In a future release, err_perm and err_temp will be
1137 completely removed, which will be a backward incompatibility if the
1138 ACL tests for either of these two old results. Patch contributed by
1139 user bes-internal on the mailing list.
1141 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
1143 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
1144 selectors, in both main and reject logs.
1146 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
1149 JH/07 Add malware type "sock" for talking to simple daemon.
1151 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
1153 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
1154 routers/transports under cutthrough routing.
1156 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
1157 numbers. Touch up "bool" conditional to keep the same definition.
1159 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
1161 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
1163 JH/12 Expand items in router/transport headers_add or headers_remove lists
1164 individually rather than the list as a whole. Bug 1452.
1166 Required for reasonable handling of multiple headers_ options when
1167 they may be empty; requires that headers_remove items with embedded
1168 colons must have them doubled (or the list-separator changed).
1170 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
1171 view the policy declared in the DMARC record. Currently, $dmarc_status
1172 is a combined value of both the record presence and the result of the
1175 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
1177 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
1178 dnslookup router and the smtp transport (applying to the forward
1181 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
1182 of ldap servers used for a specific lookup. Patch provided by Heiko
1185 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
1186 New variable $lookup_dnssec_authenticated for observability.
1188 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
1189 Patch submitted by Lars Timman.
1191 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
1193 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
1194 Requires trusted mode and valid format message id, aborts otherwise.
1195 Patch contributed by Heiko Schlichting.
1197 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
1198 certextract with support for various fields. Bug 1358.
1200 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
1201 is requested by default, modifiable by smtp transport option
1204 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
1205 operate on certificate variables to give certificate fingerprints
1206 Also new ${sha256:cert_variable}.
1208 JH/23 The PRDR feature is moved from being Experimental into the mainline.
1210 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
1211 Christian Aistleitner.
1213 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
1215 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
1216 file. Patch from Wolfgang Breyha.
1218 JH/25 Expand the coverage of the delivery $host and $host_address to
1219 client authenticators run in verify callout. Bug 1476.
1221 JH/26 Port service names are now accepted for tls_on_connect_ports, to
1222 align with daemon_smtp_ports. Bug 72.
1224 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
1225 support and error reporting did not work properly.
1227 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
1228 and is readable. Patch from Andrew Colin Kissa.
1230 TL/14 Enhance documentation of ${run expansion and how it parses the
1231 commandline after expansion, particularly in the case when an
1232 unquoted variable expansion results in an empty value.
1234 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
1236 PP/02 Fix internal collision of T_APL on systems which support RFC3123
1237 by renaming away from it. Addresses GH issue 15, reported by
1240 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
1242 TL/15 SECURITY: prevent double expansion in math comparison functions
1243 (can expand unsanitized data). Not remotely exploitable.
1250 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
1252 PP/02 Make -n do something, by making it not do something.
1253 When combined with -bP, the name of an option is not output.
1255 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
1258 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
1259 $sender_host_name and config options to manage this, and basic check
1262 PP/05 DSCP support for outbound connections and control modifier for inbound.
1264 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
1265 (Only plugin which currently uses this is kerberos4, which nobody should
1266 be using, but we should make it available and other future plugins might
1267 conceivably use it, even though it would break NAT; stuff *should* be
1268 using channel bindings instead).
1270 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
1271 name; added for Sendmail compatibility; requires admin caller.
1272 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
1273 just ignore it); requires trusted caller.
1274 Also parse but ignore: -Ac -Am -X<logfile>
1277 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
1279 TL/02 Add +smtp_confirmation as a default logging option.
1281 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
1282 Patch by Magnus Holmgren from 2007-02-20.
1284 TL/04 Bugzilla 1281 - Spec typo.
1285 Bugzilla 1283 - Spec typo.
1286 Bugzilla 1290 - Spec grammar fixes.
1288 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
1290 TL/06 Add Experimental DMARC support using libopendmarc libraries.
1292 TL/07 Fix an out of order global option causing a segfault. Reported to dev
1293 mailing list by by Dmitry Isaikin.
1295 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
1297 JH/02 Support "G" suffix to numbers in ${if comparisons.
1299 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
1301 NM/01 Bugzilla 1197 - Spec typo
1302 Bugzilla 1196 - Spec examples corrections
1304 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
1306 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
1307 gnutls_enable_pkcs11, but renamed to more accurately indicate its
1310 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
1311 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
1313 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
1314 "acl {{name}{arg}...}", and optional args on acl condition
1317 JH/05 Permit multiple router/transport headers_add/remove lines.
1319 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
1321 JH/07 Avoid using a waiting database for a single-message-only transport.
1322 Performance patch from Paul Fisher. Bugzilla 1262.
1324 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
1327 JH/09 Add $headers_added variable, with content from use of ACL modifier
1328 add_header (but not yet added to the message). Bugzilla 199.
1330 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
1331 Pulled from Bugzilla 817 by Wolfgang Breyha.
1333 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
1335 (nb: this is the same fix as in Exim 4.80.1)
1337 JH/11 Add A= logging on delivery lines, and a client_set_id option on
1340 JH/12 Add optional authenticated_sender logging to A= and a log_selector
1343 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
1345 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
1346 advertise SMTP AUTH mechanism to us, instead of a generic
1347 protocol violation error. Also, make Exim more robust to bad
1348 data from the Dovecot auth socket.
1350 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
1352 When a queue runner is handling a message, Exim first routes the
1353 recipient addresses, during which it prunes them based on the retry
1354 hints database. After that it attempts to deliver the message to
1355 any remaining recipients. It then updates the hints database using
1358 So if a recipient address works intermittently, it can get repeatedly
1359 deferred at routing time. The retry hints record remains fresh so the
1360 address never reaches the final cutoff time.
1362 This is a fairly common occurrence when a user is bumping up against
1363 their storage quota. Exim had some logic in its local delivery code
1364 to deal with this. However it did not apply to per-recipient defers
1365 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
1367 This change adds a proper retry rule check during routing so that the
1368 final cutoff time is checked against the message's age. We only do
1369 this check if there is an address retry record and there is not a
1370 domain retry record; this implies that previous attempts to handle
1371 the address had the retry_use_local_parts option turned on. We use
1372 this as an approximation for the destination being like a local
1373 delivery, as in LMTP.
1375 I suspect this new check makes the old local delivery cutoff check
1376 redundant, but I have not verified this so I left the code in place.
1378 TF/02 Correct gecos expansion when From: is a prefix of the username.
1380 Test 0254 submits a message to Exim with the header
1384 When I ran the test suite under the user fanf2, Exim expanded
1385 the header to contain my full name, whereas it should have added
1386 a Resent-Sender: header. It erroneously treats any prefix of the
1387 username as equal to the username.
1389 This change corrects that bug.
1391 GF/01 DCC debug and logging tidyup
1392 Error conditions log to paniclog rather than rejectlog.
1393 Debug lines prefixed by "DCC: " to remove any ambiguity.
1395 TF/03 Avoid unnecessary rebuilds of lookup-related code.
1397 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
1398 Bug spotted by Jeremy Harris; was flawed since initial commit.
1399 Would have resulted in OCSP responses post-SNI triggering an Exim
1400 NULL dereference and crash.
1402 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
1404 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
1405 Bug detection, analysis and fix by Samuel Thibault.
1406 Bugzilla 1331, Debian bug #698092.
1408 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
1410 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
1411 Server implementation by Todd Lyons, client by JH.
1412 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
1413 config variable "prdr_enable" controls whether the server
1414 advertises the facility. If the client requests PRDR a new
1415 acl_data_smtp_prdr ACL is called once for each recipient, after
1416 the body content is received and before the acl_smtp_data ACL.
1417 The client is controlled by both of: a hosts_try_prdr option
1418 on the smtp transport, and the server advertisement.
1419 Default client logging of deliveries and rejections involving
1420 PRDR are flagged with the string "PRDR".
1422 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
1423 fclose(). Diagnosis by Todd Lyons.
1425 PP/17 Update configure.default to handle IPv6 localhost better.
1426 Patch by Alain Williams (plus minor tweaks).
1429 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
1430 This is now consistent with GnuTLS, and is now documented: the
1431 previous undocumented portable approach to treating the option as
1432 unset was to force an expansion failure. That still works, and
1433 an empty string is now equivalent.
1435 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
1436 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
1437 not performing validation itself.
1439 PP/20 Added force_command boolean option to pipe transport.
1440 Patch from Nick Koston, of cPanel Inc.
1442 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
1445 TF/04 Added udpsend ACL modifier and hexquote expansion operator
1447 PP/21 Fix eximon continuous updating with timestamped log-files.
1448 Broken in a format-string cleanup in 4.80, missed when I repaired the
1449 other false fix of the same issue.
1450 Report and fix from Heiko Schlichting.
1453 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
1454 Report from Prashanth Katuri.
1456 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
1457 It's SecureTransport, so affects any MacOS clients which use the
1458 system-integrated TLS libraries, including email clients.
1460 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
1461 using a MIME ACL for non-SMTP local injection.
1462 Report and assistance in diagnosis by Warren Baker.
1464 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
1466 JH/16 Fix comparisons for 64b. Bugzilla 1385.
1468 TL/09 Add expansion variable $authenticated_fail_id to keep track of
1469 last id that failed so it may be referenced in subsequent ACL's.
1471 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
1474 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
1475 ldap library initialization, allowing self-signed CA's to be
1476 used. Also properly sets require_cert option later in code by
1477 using NULL (global ldap config) instead of ldap handle (per
1478 session). Bug diagnosis and testing by alxgomz.
1480 TL/12 Enhanced documentation in the ratelimit.pl script provided in
1481 the src/util/ subdirectory.
1483 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
1484 renamed to Transport Post Delivery Action by Jeremy Harris, as
1487 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
1488 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
1489 redis_servers = needs to be configured which will be used by the redis
1490 lookup. Patch from Warren Baker, of The Packet Hub.
1492 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
1494 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
1495 hostname or reverse DNS when processing a host list. Used suggestions
1496 from multiple comments on this bug.
1498 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
1500 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
1501 Missed a few lines, added it to make the runtest require no keyboard
1504 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
1505 contains upper case chars. Make router use caseful_local_part.
1507 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
1508 support when GnuTLS has been built with p11-kit.
1514 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
1516 This, or similar/improved, will also be change PP/11 of 4.82.
1522 PP/01 Handle short writes when writing local log-files.
1523 In practice, only affects FreeBSD (8 onwards).
1524 Bugzilla 1053, with thanks to Dmitry Isaikin.
1526 NM/01 Bugzilla 949 - Documentation tweak
1528 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
1531 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
1533 PP/02 Implemented gsasl authenticator.
1535 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
1537 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
1538 `pkg-config foo` for cflags/libs.
1540 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
1541 with rest of GSASL and with heimdal_gssapi.
1543 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
1544 `pkg-config foo` for cflags/libs for the TLS implementation.
1546 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
1547 properties get this fed in as external SSF. A number of robustness
1548 and debugging improvements to the cyrus_sasl authenticator.
1550 PP/08 cyrus_sasl server now expands the server_realm option.
1552 PP/09 Bugzilla 1214 - Log authentication information in reject log.
1553 Patch by Jeremy Harris.
1555 PP/10 Added dbmjz lookup type.
1557 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
1559 PP/12 MAIL args handles TAB as well as SP, for better interop with
1560 non-compliant senders.
1561 Analysis and variant patch by Todd Lyons.
1563 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
1564 Bug report from Lars Müller <lars@samba.org> (via SUSE),
1565 Patch from Dirk Mueller <dmueller@suse.com>
1567 PP/13 tls_peerdn now print-escaped for spool files.
1568 Observed some $tls_peerdn in wild which contained \n, which resulted
1569 in spool file corruption.
1571 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
1572 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
1573 or write after TLS renegotiation, which otherwise led to messages
1576 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
1577 as a tracking header (ie: a signed header comes before the signature).
1578 Patch from Wolfgang Breyha.
1580 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
1581 comma-sep list; embedded commas doubled.
1583 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
1585 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
1587 Report and patch from Dmitry Banschikov.
1589 PP/16 Removed "dont_insert_empty_fragments" from "openssl_options".
1590 Removed SSL_clear() after SSL_new() which led to protocol negotiation
1591 failures. We appear to now support TLS1.1+ with Exim.
1593 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
1594 lets Exim select keys and certificates based upon TLS SNI from client.
1595 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
1596 before an outbound SMTP session. New log_selector, +tls_sni.
1598 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
1599 NULL dereference. Report and patch from Alun Jones.
1601 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
1602 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
1603 Not seeing resolver debug output on NetBSD, but suspect this is a
1604 resolver implementation change.
1606 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
1607 Left warnings. Added "eximon gdb" invocation mode.
1609 PP/21 Defaulting "accept_8bitmime" to true, not false.
1611 PP/22 Added -bw for inetd wait mode support.
1613 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
1614 locate the relevant includes and libraries. Made this the default.
1616 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
1617 Bugzilla 1246, report and most of solution from Tomasz Kusy.
1619 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
1620 This may cause build issues on older platforms.
1622 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
1623 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
1624 gnutls_require_mac & gnutls_require_protocols (no longer supported).
1625 Added SNI support via GnuTLS too.
1626 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
1628 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
1630 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
1631 Applied second patch from Janne, implementing suggestion to default
1632 multiple-strings-in-record handling to match SPF spec.
1634 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
1636 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
1637 read-only, out of scope).
1638 Patch from Wolfgang Breyha, report from Stuart Northfield.
1640 PP/29 Fix three issues highlighted by clang analyser static analysis.
1641 Only crash-plausible issue would require the Cambridge-specific
1642 iplookup router and a misconfiguration.
1643 Report from Marcin Mirosław.
1645 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
1647 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
1648 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
1649 As part of this, removing so much warning spew let me fix some minor
1650 real issues in debug logging.
1652 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
1653 assignment on my part. Fixed.
1655 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
1656 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
1657 Janne Snabb (who went above and beyond: thank you).
1659 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
1660 string otherwise requires a connection and a bunch more work and it's
1661 relatively easy to get wrong. Should also expose TLS library linkage
1664 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
1665 64-bit ${eval} (JH/03).
1667 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
1668 GNU libc to support some of the 64-bit stuff, should not lead to
1669 conflicts. Defined before os.h is pulled in, so if a given platform
1670 needs to override this, it can.
1672 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
1673 protection layer was required, which is not implemented.
1674 Bugzilla 1254, patch from Wolfgang Breyha.
1676 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
1677 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
1678 tls_dhparam take prime identifiers. Also unbreak combination of
1679 OpenSSL+DH_params+TLSSNI.
1681 PP/39 Disable SSLv2 by default in OpenSSL support.
1687 PP/01 Solaris build fix for Oracle's LDAP libraries.
1688 Bugzilla 1109, patch from Stephen Usher.
1690 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
1692 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
1695 TF/02 Fix a couple more cases where we did not log the error message
1696 when unlink() failed. See also change 4.74-TF/03.
1698 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
1699 lock up or crash if it happened to be inside a call to libc when it
1700 got a SIGUSR1 from exiwhat.
1702 The SIGUSR1 handler appends the current process status to the process
1703 log which is later printed by exiwhat. It used to use the general
1704 purpose logging code to do this, but several functions it calls are
1705 not safe for signals.
1707 The new output code in the SIGUSR1 handler is specific to the process
1708 log, and simple enough that it's easy to inspect for signal safety.
1709 Removing some special cases also simplifies the general logging code.
1710 Removing the spurious timestamps from the process log simplifies
1713 TF/04 Improved ratelimit ACL condition.
1715 The /noupdate option has been deprecated in favour of /readonly which
1716 has clearer semantics. The /leaky, /strict, and /readonly update modes
1717 are mutually exclusive. The update mode is no longer included in the
1718 database key; it just determines when the database is updated. (This
1719 means that when you upgrade Exim will forget old rate measurements.)
1721 Exim now checks that the per_* options are used with an update mode that
1722 makes sense for the current ACL. For example, when Exim is processing a
1723 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
1724 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
1725 must specify per_mail/readonly. If you omit the update mode it defaults to
1726 /leaky where that makes sense (as before) or /readonly where required.
1728 The /noupdate option is now undocumented but still supported for
1729 backwards compatibility. It is equivalent to /readonly except that in
1730 ACLs where /readonly is required you may specify /leaky/noupdate or
1731 /strict/noupdate which are treated the same as /readonly.
1733 A useful new feature is the /count= option. This is a generalization
1734 of the per_byte option, so that you can measure the throughput of other
1735 aggregate values. For example, the per_byte option is now equivalent
1736 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
1738 The per_rcpt option has been generalized using the /count= mechanism
1739 (though it's more complicated than the per_byte equivalence). When it is
1740 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
1741 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
1742 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
1743 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
1744 Note that using per_rcpt with a non-readonly update mode in more than
1745 one ACL will cause the recipients to be double-counted. (The per_mail
1746 and per_byte options don't have this problem.)
1748 The handling of very low rates has changed slightly. If the computed rate
1749 is less than the event's count (usually one) then this event is the first
1750 after a long gap. In this case the rate is set to the same as this event's
1751 count, so that the first message of a spam run is counted properly.
1753 The major new feature is a mechanism for counting the rate of unique
1754 events. The new per_addr option counts the number of different
1755 recipients that someone has sent messages to in the last time period. It
1756 behaves like per_rcpt if all the recipient addresses are different, but
1757 duplicate recipient addresses do not increase the measured rate. Like
1758 the /count= option this is a general mechanism, so the per_addr option
1759 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
1760 example, measure the rate that a client uses different sender addresses
1761 with the options per_mail/unique=$sender_address. There are further
1762 details in the main documentation.
1764 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
1766 TF/06 Removed a few PCRE remnants.
1768 TF/07 Automatically extract Exim's version number from tags in the git
1769 repository when doing development or release builds.
1771 PP/02 Raise smtp_cmd_buffer_size to 16kB.
1772 Bugzilla 879. Patch from Paul Fisher.
1774 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
1775 Heavily based on revision 40f9a89a from Simon Arlott's tree.
1778 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
1780 PP/05 Variable $av_failed, true if the AV scanner deferred.
1781 Bugzilla 1078. Patch from John Horne.
1783 PP/06 Stop make process more reliably on build failure.
1784 Bugzilla 1087. Patch from Heiko Schlittermann.
1786 PP/07 Make maildir_use_size_file an _expandable_ boolean.
1787 Bugzilla 1089. Patch from Heiko Schlittermann.
1789 PP/08 Handle ${run} returning more data than OS pipe buffer size.
1790 Bugzilla 1131. Patch from Holger Weiß.
1792 PP/09 Handle IPv6 addresses with SPF.
1793 Bugzilla 860. Patch from Wolfgang Breyha.
1795 PP/10 GnuTLS: support TLS 1.2 & 1.1.
1797 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
1800 PP/11 match_* no longer expand right-hand-side by default.
1801 New compile-time build option, EXPAND_LISTMATCH_RHS.
1802 New expansion conditions, "inlist", "inlisti".
1804 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
1806 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
1808 PP/14 fix log_write() format string regression from TF/03.
1809 Bugzilla 1152. Patch from Dmitry Isaikin.
1815 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
1817 PP/02 Harmonised TLS library version reporting; only show if debugging.
1818 Layout now matches that introduced for other libraries in 4.74 PP/03.
1820 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
1822 PP/04 New "dns_use_edns0" global option.
1824 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
1827 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
1828 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
1830 TK/01 Updated PolarSSL code to 0.14.2.
1831 Bugzilla 1097. Patch from Andreas Metzler.
1833 PP/07 Catch divide-by-zero in ${eval:...}.
1834 Fixes bugzilla 1102.
1836 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
1839 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
1840 format-string attack -- SECURITY: remote arbitrary code execution.
1842 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
1843 time unintentionally subject to list matching rules, letting the header
1844 cause arbitrary Exim lookups (of items which can occur in lists, *not*
1845 arbitrary string expansion). This allowed for information disclosure.
1847 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
1848 INT_MIN/-1 -- value coerced to INT_MAX.
1854 NM/01 Workaround for PCRE version dependency in version reporting
1857 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
1858 This fixes portability to compilers other than gcc, notably
1859 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
1861 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
1862 makefiles for portability to HP-UX and POSIX correctness.
1864 PP/01 Permit LOOKUP_foo enabling on the make command-line.
1865 Also via indented variable definition in the Makefile.
1866 (Debugging by Oliver Heesakkers).
1868 PP/02 Restore caching of spamd results with expanded spamd_address.
1869 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
1871 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
1872 Improves build reliability. Fix from: Frank Elsner
1874 NM/02 Fix wide character breakage in the rfc2047 coding
1875 Fixes bug 1064. Patch from Andrey N. Oktyabrski
1877 NM/03 Allow underscore in dnslist lookups
1878 Fixes bug 1026. Patch from Graeme Fowler
1880 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
1881 Code patches from Adam Ciarcinski of NetBSD.
1883 NM/04 Fixed exiqgrep to cope with mailq missing size issue
1886 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
1887 is logged, to avoid truncation. Patch from John Horne.
1889 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
1890 Patch from Jakob Hirsch.
1892 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
1893 SQL string expansion failure details.
1894 Patch from Andrey Oktyabrski.
1896 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
1897 Patch from Simon Arlott.
1899 PP/09 New lookups functionality failed to compile on old gcc which rejects
1900 extern declarations in function scope.
1901 Patch from Oliver Fleischmann
1903 PP/10 Use sig_atomic_t for flags set from signal handlers.
1904 Check getgroups() return and improve debugging.
1905 Fixed developed for diagnosis in bug 927 (which turned out to be
1908 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
1909 Patch from Mark Zealey.
1911 PP/12 Bugzilla 1056: Improved spamd server selection.
1912 Patch from Mark Zealey.
1914 PP/13 Bugzilla 1086: Deal with maildir quota file races.
1915 Based on patch from Heiko Schlittermann.
1917 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
1918 Patch from Uwe Doering, sign-off by Michael Haardt.
1920 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
1921 variable declaration deep within a block. Bug and patch from
1924 PP/15 lookups-Makefile IRIX compatibility coercion.
1926 PP/16 Make DISABLE_DKIM build knob functional.
1928 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
1929 Patch by Simon Arlott
1931 TF/03 Fix valgrind.h portability to C89 compilers that do not support
1932 variable argument macros. Our copy now differs from upstream.
1938 TF/01 Failure to get a lock on a hints database can have serious
1939 consequences so log it to the panic log.
1941 TF/02 Log LMTP confirmation messages in the same way as SMTP,
1942 controlled using the smtp_confirmation log selector.
1944 TF/03 Include the error message when we fail to unlink a spool file.
1946 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
1947 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
1948 for maintaining out-of-tree patches for some time.
1950 PP/01 Bugzilla 139: Documentation and portability issues.
1951 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
1952 Handle per-OS dynamic-module compilation flags.
1954 PP/02 Let /dev/null have normal permissions.
1955 The 4.73 fixes were a little too stringent and complained about the
1956 permissions on /dev/null. Exempt it from some checks.
1957 Reported by Andreas M. Kirchwitz.
1959 PP/03 Report version information for many libraries, including
1960 Exim version information for dynamically loaded libraries. Created
1961 version.h, now support a version extension string for distributors
1962 who patch heavily. Dynamic module ABI change.
1964 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
1965 privilege escalation vulnerability whereby the Exim run-time user
1966 can cause root to append content of the attacker's choosing to
1969 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
1972 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
1973 If dropping privileges for untrusted macros, we disabled normal logging
1974 on the basis that it would fail; for the Exim run-time user, this is not
1975 the case, and it resulted in successful deliveries going unlogged.
1976 Fixed. Reported by Andreas Metzler.
1982 PP/01 Date: & Message-Id: revert to normally being appended to a message,
1983 only prepend for the Resent-* case. Fixes regression introduced in
1984 Exim 4.70 by NM/22 for Bugzilla 607.
1986 PP/02 Include check_rfc2047_length in configure.default because we're seeing
1987 increasing numbers of administrators be bitten by this.
1989 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
1991 PP/03 Bugzilla 994: added openssl_options main configuration option.
1993 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
1995 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
1997 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
1999 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
2000 without a peer certificate, leading to a segfault because of an
2001 assumption that peers always have certificates. Be a little more
2002 paranoid. Problem reported by Martin Tscholak.
2004 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
2005 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
2006 NB: ClamAV planning to remove STREAM in "middle of 2010".
2007 CL also introduces -bmalware, various -d+acl logging additions and
2008 more caution in buffer sizes.
2010 PP/09 Implemented reverse_ip expansion operator.
2012 PP/10 Bugzilla 937: provide a "debug" ACL control.
2014 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
2016 PP/12 Bugzilla 973: Implement --version.
2018 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
2020 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
2022 PP/15 Bugzilla 816: support multiple condition rules on Routers.
2024 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
2025 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
2026 ignore trailing whitespace.
2028 JJ/02 prevent non-panic DKIM error from being sent to paniclog
2030 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
2033 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
2034 Notification from Dr Andrew Aitchison.
2036 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
2037 ExtendedDetectionInfo response format.
2038 Notification from John Horne.
2040 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
2043 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
2044 XSL and documented dependency on system catalogs, with examples of how
2047 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
2050 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
2051 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
2052 configuration file which is writeable by the Exim user or group.
2054 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
2055 of configuration files to cover files specified with the -C option if
2056 they are going to be used with root privileges, not just the default
2059 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
2060 option (effectively making it always true).
2062 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
2063 files to be used while preserving root privileges.
2065 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
2066 that rogue child processes cannot use them.
2068 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
2069 run-time user, instead of root.
2071 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
2072 Exim run-time user without dropping privileges.
2074 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
2075 result string, instead of calling string_vformat() twice with the same
2078 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
2079 for other users. Others should always drop root privileges if they use
2080 -C on the command line, even for a whitelisted configure file.
2082 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
2084 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
2090 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
2091 $data_path, and $header_path variables; fixed documentation bugs and
2094 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
2095 exipick to access non-standard spools, including the "frozen" queue
2098 NM/01 Bugzilla 965: Support mysql stored procedures.
2099 Patch from Alain Williams
2101 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
2103 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
2104 Patch from Andreas Metzler
2106 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
2107 Patch from Kirill Miazine
2109 NM/05 Bugzilla 671: Added umask to procmail example.
2111 JJ/03 installed exipick 20100323.0, fixing doc bug
2113 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
2114 directory. Notification and patch from Dan Rosenberg.
2116 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
2118 TK/02 Improve log output when DKIM signing operation fails.
2120 MH/01 Treat the transport option dkim_domain as a colon separated
2121 list, not as a single string, and sign the message with each element,
2122 omitting multiple occurences of the same signer.
2124 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
2125 Bugzilla 985, 986. Patch by Simon Arlott
2127 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
2128 Patch by Simon Arlott
2130 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
2131 MBX locking. Notification from Dan Rosenberg.
2137 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
2139 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
2141 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
2143 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
2145 NM/04 Bugzilla 915: Flag broken perl installation during build.
2151 TK/01 Added patch by Johannes Berg that expands the main option
2152 "spamd_address" if it starts with a dollar sign.
2154 TK/02 Write list of recipients to X-Envelope-Sender header when building
2155 the mbox-format spool file for content scanning (suggested by Jakob
2158 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
2159 (http://www.dcc-servers.net/) support via dccifd. Activated by
2160 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
2162 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
2163 by Mark Daniel Reidel <mr@df.eu>.
2165 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
2166 When building exim an external PCRE library is now needed -
2167 PCRE is a system library on the majority of modern systems.
2168 See entry on PCRE_LIBS in EDITME file.
2170 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
2171 conversation. Added nologin parameter to request.
2172 Patch contributed by Kirill Miazine.
2174 TF/01 Do not log submission mode rewrites if they do not change the address.
2176 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
2178 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
2179 log files in place. Contributed by Roberto Lima.
2181 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
2183 TF/03 Bugzilla 615: When checking the local_parts router precondition
2184 after a local_part_suffix or local_part_prefix option, Exim now
2185 does not use the address's named list lookup cache, since this
2186 contains cached lookups for the whole local part.
2188 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
2189 Robert Millan. Documentation is in experimental-spec.txt.
2191 TF/04 Bugzilla 668: Fix parallel build (make -j).
2193 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
2195 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
2196 Patch provided by Jan Srzednicki.
2198 TF/05 Leading white space used to be stripped from $spam_report which
2199 wrecked the formatting. Now it is preserved.
2201 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
2202 that they are available at delivery time.
2204 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
2206 TF/08 TLS error reporting now respects the incoming_interface and
2207 incoming_port log selectors.
2209 TF/09 Produce a more useful error message if an SMTP transport's hosts
2210 setting expands to an empty string.
2212 NM/06 Bugzilla 744: EXPN did not work under TLS.
2213 Patch provided by Phil Pennock.
2215 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
2216 Patch provided by Richard Godbee.
2218 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
2219 acl_smtp_notquit, added index entry.
2221 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
2222 Patch provided by Eugene Bujak.
2224 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
2225 accept(). Patch provided by Maxim Dounin.
2227 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
2228 Patch provided by Phil Pennock.
2230 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
2232 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
2233 Patch provided by Brad "anomie" Jorsch.
2235 NM/14 Bugzilla 622: Added timeout setting to transport filter.
2236 Patch provided by Dean Brooks.
2238 TK/05 Add native DKIM support (does not depend on external libraries).
2240 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
2241 Patch provided by Graeme Fowler.
2243 NM/16 Bugzilla 851: Documentation example syntax fix.
2245 NM/17 Changed NOTICE file to remove references to embedded PCRE.
2247 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
2250 NM/19 Bugzilla 745: TLS version reporting.
2251 Patch provided by Phil Pennock.
2253 NM/20 Bugzilla 167: bool: condition support.
2254 Patch provided by Phil Pennock.
2256 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
2257 clients. Patch provided by Phil Pennock.
2259 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
2260 Patch provided by Brad "anomie" Jorsch.
2262 NM/23 Bugzilla 687: Fix misparses in eximstats.
2263 Patch provided by Heiko Schlittermann.
2265 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
2266 Patch provided by Heiko Schlittermann.
2268 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
2269 plus update to original patch.
2271 NM/26 Bugzilla 799: Documentation correction for ratelimit.
2273 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
2274 Patch provided by David Brownlee.
2276 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
2278 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
2280 NM/30 Bugzilla 888: TLS documentation bugfixes.
2282 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
2284 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
2285 Unlike the original bugzilla I have changed all shell scripts in src tree.
2287 NM/33 Bugzilla 898: Transport filter timeout fix.
2288 Patch by Todd Rinaldo.
2290 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
2291 Patch by Serge Demonchaux.
2293 NM/35 Bugzilla 39: Base64 decode bug fixes.
2294 Patch by Jakob Hirsch.
2296 NM/36 Bugzilla 909: Correct connect() call in dcc code.
2298 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
2300 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
2302 NM/39 Bugzilla 911: Fixed MakeLinks build script.
2308 TK/01 Add preliminary DKIM support. Currently requires a forked version of
2309 ALT-N's libdkim that I have put here:
2310 http://duncanthrax.net/exim-experimental/
2312 Note to Michael Haardt: I had to rename some vars in sieve.c. They
2313 were called 'true' and it seems that C99 defines that as a reserved
2314 keyword to be used with 'bool' variable types. That means you could
2315 not include C99-style headers which use bools without triggering
2316 build errors in sieve.c.
2318 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
2319 as mailq or other aliases. Changed the --help handling significantly
2320 to do whats expected. exim_usage() emits usage/help information.
2322 SC/01 Added the -bylocaldomain option to eximstats.
2324 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
2326 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
2328 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
2334 PH/01 Another patch from the Sieve maintainer.
2336 PH/02 When an IPv6 address is converted to a string for single-key lookup
2337 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
2338 dots are used instead of colons so that keys in lsearch files need not
2339 contain colons. This was done some time before quoting was made available
2340 in lsearch files. However, iplsearch files do require colons in IPv6 keys
2341 (notated using the quote facility) so as to distinguish them from IPv4
2342 keys. This meant that lookups for IP addresses in host lists did not work
2343 for iplsearch lookups.
2345 This has been fixed by arranging for IPv6 addresses to be expressed with
2346 colons if the lookup type is iplsearch. This is not incompatible, because
2347 previously such lookups could never work.
2349 The situation is now rather anomalous, since one *can* have colons in
2350 ordinary lsearch keys. However, making the change in all cases is
2351 incompatible and would probably break a number of configurations.
2353 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
2356 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
2357 conversion specification without a maximum field width, thereby enabling
2358 a rogue spamd server to cause a buffer overflow. While nobody in their
2359 right mind would setup Exim to query an untrusted spamd server, an
2360 attacker that gains access to a server running spamd could potentially
2361 exploit this vulnerability to run arbitrary code as the Exim user.
2363 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
2364 $primary_hostname instead of what libspf2 thinks the hosts name is.
2366 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
2367 a directory entry by the name of the lookup key. Previously, if a
2368 symlink pointed to a non-existing file or a file in a directory that
2369 Exim lacked permissions to read, a lookup for a key matching that
2370 symlink would fail. Now it is enough that a matching directory entry
2371 exists, symlink or not. (Bugzilla 503.)
2373 PH/03 The body_linecount and body_zerocount variables are now exported in the
2376 PH/04 Added the $dnslist_matched variable.
2378 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
2379 This means they are set thereafter only if the connection becomes
2382 PH/06 Added the client_condition to authenticators so that some can be skipped
2383 by clients under certain conditions.
2385 PH/07 The error message for a badly-placed control=no_multiline_responses left
2386 "_responses" off the end of the name.
2388 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
2390 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
2391 (without spaces) instead of just copying the configuration text.
2393 PH/10 Added the /noupdate option to the ratelimit ACL condition.
2395 PH/11 Added $max_received_linelength.
2397 PH/12 Added +ignore_defer and +include_defer to host lists.
2399 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
2400 way in which PCRE > 7.0 is built.
2402 PH/14 Implemented queue_only_load_latch.
2404 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
2405 MAIL command. The effect was to mangle the value on 64-bit systems.
2407 PH/16 Another patch from the Sieve maintainer.
2409 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
2411 PH/18 If a system quota error occurred while trying to create the file for
2412 a maildir delivery, the message "Mailbox is full" was not appended to the
2413 bounce if the delivery eventually timed out. Change 4.67/27 below applied
2414 only to a quota excession during the actual writing of the file.
2416 PH/19 It seems that peer DN values may contain newlines (and other non-printing
2417 characters?) which causes problems in log lines. The DN values are now
2418 passed through string_printing() before being added to log lines.
2420 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
2421 and InterBase are left for another time.)
2423 PH/21 Added message_body_newlines option.
2425 PH/22 Guard against possible overflow in moan_check_errorcopy().
2427 PH/23 POSIX allows open() to be a macro; guard against that.
2429 PH/24 If the recipient of an error message contained an @ in the local part
2430 (suitably quoted, of course), incorrect values were put in $domain and
2431 $local_part during the evaluation of errors_copy.
2437 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
2438 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
2441 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
2442 issue a MAIL command.
2444 PH/02 In an ACL statement such as
2446 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
2448 if a client was not listed at all, or was listed with a value other than
2449 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
2450 the condition was not true (as it should be), so access was not denied.
2451 The bug was that the ! inversion was incorrectly passed on to the second
2452 item. This has been fixed.
2454 PH/03 Added additional dnslists conditions == and =& which are different from
2455 = and & when the dns lookup returns more than one IP address.
2457 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
2458 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
2460 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
2461 FSYNC, which compiles an option called disable_fsync that allows for
2462 bypassing fsync(). The documentation is heavily laced with warnings.
2464 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
2466 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
2467 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
2468 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
2469 including adding "make clean"; (3) Added -fPIC when compiling the test
2470 dynamically loaded module, to get rid of a warning.
2472 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
2473 message fails, move_frozen_messages = true and ignore_bounce_errors_after
2474 = 0s. The bug is otherwise harmless.
2476 PH/07 There was a bug in the dovecot authenticator such that the value of
2477 $auth1 could be overwritten, and so not correctly preserved, after a
2478 successful authentication. This usually meant that the value preserved by
2479 the server_setid option was incorrect.
2481 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
2483 PH/09 Installed PCRE release 7.0.
2485 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
2486 run for batched SMTP input. It is now run at the start of every message
2487 in the batch. While fixing this I discovered that the process information
2488 (output by running exiwhat) was not always getting set for -bs and -bS
2489 input. This is fixed, and it now also says "batched" for BSMTP.
2491 PH/11 Added control=no_pipelining.
2493 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
2494 patch, slightly modified), and move the expansion of helo_data till after
2495 the connection is made in the smtp transport (so it can use these
2498 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
2500 PH/14 Added log_selector = +pid.
2502 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
2504 PH/16 Add ${if forany and ${if forall.
2506 PH/17 Added dsn_from option to vary the From: line in DSNs.
2508 PH/18 Flush SMTP output before performing a callout, unless control =
2509 no_callout_flush is set.
2511 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
2512 was true (the default) a successful delivery failed to delete the retry
2513 item, thus causing premature timeout of the address. The bug is now
2516 PH/20 Added hosts_avoid_pipelining to the smtp transport.
2518 PH/21 Long custom messages for fakedefer and fakereject are now split up
2519 into multiline responses in the same way that messages for "deny" and
2520 other ACL rejections are.
2522 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
2523 with slight modification.
2525 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
2526 draft, changing the syntax and factoring some duplicate code".
2528 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
2529 for deliveries of the second and subsequent messages over the same SMTP
2532 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
2533 ${reduce, with only minor "tidies".
2535 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
2537 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
2538 expansion side effects.
2540 PH/27 When a message times out after an over-quota error from an Exim-imposed
2541 quota, the bounce message says "mailbox is full". This message was not
2542 being given when it was a system quota that was exceeded. It now should
2545 MH/03 Made $recipients available in local_scan(). local_scan() already has
2546 better access to the recipient list through recipients_list[], but
2547 $recipients can be useful in postmaster-provided expansion strings.
2549 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
2550 in the case of a MAIL command with additional options following the
2551 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
2552 were accidentally chopped off.
2554 PH/29 SMTP synchronization checks are implemented when a command is read -
2555 there is a check that no more input is waiting when there shouldn't be
2556 any. However, for some commands, a delay in an ACL can mean that it is
2557 some time before the response is written. In this time, more input might
2558 arrive, invalidly. So now there are extra checks after an ACL has run for
2559 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
2560 pipelining has not been advertised.
2562 PH/30 MH's patch to allow iscntrl() characters to be list separators.
2564 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
2565 returned in the SMTP response when smtp_return_error_details was false.
2566 This has been fixed.
2568 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
2569 instead of the C I/O that was originally supplied, because problems were
2570 reported on Solaris.
2572 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
2573 Exim which did not show up earlier: it was assuming that a call to
2574 SSL_CTX_set_info_callback() might give an error value. In fact, there is
2575 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
2576 was a macro that became an assignment, so it seemed to work. This has
2577 changed to a proper function call with a void return, hence the compile
2578 error. Exim's code has been fixed.
2580 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
2583 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
2585 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
2587 PH/37 If a message is not accepted after it has had an id assigned (e.g.
2588 because it turns out to be too big or there is a timeout) there is no
2589 "Completed" line in the log. When some messages of this type were
2590 selected by exigrep, they were listed as "not completed". Others were
2591 picked up by some special patterns. I have improved the selection
2592 criteria to be more general.
2594 PH/38 The host_find_failed option in the manualroute router can now be set
2595 to "ignore", to completely ignore a host whose IP address cannot be
2596 found. If all hosts are ignored, the behaviour is controlled by the new
2597 host_all_ignored option.
2599 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
2600 homing or because of multiple MX records with /mx) generated more than
2601 one IP address, and the following item turned out to be the local host,
2602 all the secondary addresses of the first item were incorrectly removed
2603 from the list, along with the local host and any following hosts (which
2604 is what is supposed to happen).
2606 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
2607 whoever called Exim into the -H file. In the case of the daemon it was
2608 behaving confusingly. When first started, it used values for whoever
2609 started the daemon, but after a SIGHUP it used the Exim user (because it
2610 calls itself on a restart). I have changed the code so that it now always
2613 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
2614 message are rejected with the same error (e.g. no authentication or bad
2615 sender address), and a DATA command is nevertheless sent (as can happen
2616 with PIPELINING or a stupid MUA), the error message that was given to the
2617 RCPT commands is included in the rejection of the DATA command. This is
2618 intended to be helpful for MUAs that show only the final error to their
2621 PH/42 Another patch from the Sieve maintainer.
2623 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
2624 Eximstats - Fixed some broken HTML links and added missing column headers
2626 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
2627 columns for Rejects, Temp Rejects, Ham, and Spam rows.
2629 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
2631 PH/43 Yet another patch from the Sieve maintainer.
2633 PH/44 I found a way to check for a TCP/IP connection going away before sending
2634 the response to the final '.' that terminates a message, but only in the
2635 case where the client has not sent further data following the '.'
2636 (unfortunately, this is allowed). However, in many cases there won't be
2637 any further data because there won't be any more messages to send. A call
2638 to select() can be used: if it shows that the input is "ready", there is
2639 either input waiting, or the socket has been closed. An attempt to read
2640 the next input character can distinguish the two cases. Previously, Exim
2641 would have sent an OK response which the client would never have see.
2642 This could lead to message repetition. This fix should cure that, at
2643 least in a lot of common cases.
2645 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
2646 advertised in response to EHLO.
2652 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
2653 fixed by 4.65/MH/01 (is this a record?) are fixed:
2655 (i) An empty string was always treated as zero by the numeric comparison
2656 operators. This behaviour has been restored.
2658 (ii) It is documented that the numeric comparison operators always treat
2659 their arguments as decimal numbers. This was broken in that numbers
2660 starting with 0 were being interpreted as octal.
2662 While fixing these problems I realized that there was another issue that
2663 hadn't been noticed. Values of message_size_limit (both the global option
2664 and the transport option) were treated as octal if they started with 0.
2665 The documentation was vague. These values are now always treated as
2666 decimal, and I will make that clear in the documentation.
2672 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
2673 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
2676 MH/01 Don't check that the operands of numeric comparison operators are
2677 integers when their expansion is in "skipping" mode (fixes bug
2678 introduced by 4.64-PH/07).
2680 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
2681 child addresses, Exim now panics and dies. Previously, because the count
2682 is held in a short int, deliveries were likely to be lost. As such a
2683 large number of recipients for a single message is ridiculous
2684 (performance will be very, very poor), I have chosen to impose a limit
2685 rather than extend the field.
2691 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
2692 leftover -K file (the existence of which was triggered by #402).
2693 While we were at it, introduced process PID as part of the -K
2694 filename. This should rule out race conditions when creating
2697 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
2698 processing considerably. Previous code took too long for large mails,
2699 triggering a timeout which in turn triggers #401.
2701 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
2702 in the DK code in transports.c. sendfile() is not really portable,
2703 hence the _LINUX specificness.
2705 TF/01 In the add_headers option to the mail command in an Exim filter,
2706 there was a bug that Exim would claim a syntax error in any
2707 header after the first one which had an odd number of characters
2710 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
2711 callout verification, Exim cached a "reject" for the entire domain. This
2712 is correct for most verifications, but it is not correct for a recipient
2713 verification with use_sender or use_postmaster set, because in that case
2714 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
2715 case of MAIL FROM:<> rejection from other early rejections (e.g.
2716 rejection of HELO). When verifying a recipient using a non-null MAIL
2717 address, the cache is ignored if it shows MAIL FROM:<> rejection.
2718 Whatever the result of the callout, the value of the domain cache is
2719 left unchanged (for any other kind of callout, getting as far as trying
2720 RCPT means that the domain itself is ok).
2722 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
2725 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
2726 manifest itself as EPIPE rather than ECONNECT. When tidying away a
2727 session, the daemon ignores ECONNECT errors and logs others; it now
2728 ignores EPIPE as well.
2730 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
2731 (quoted-printable decoding).
2733 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
2734 later the small subsequent patch to fix an introduced bug.
2736 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
2738 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
2740 PH/08 An error is now given if message_size_limit is specified negative.
2742 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
2743 to be given (somewhat) arbitrary names.
2745 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
2748 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
2749 miscellaneous code fixes
2751 PH/10 Added the log_reject_target ACL modifier to specify where to log
2754 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
2755 hostname. This is wrong, because it relates to the incoming message (and
2756 probably the interface on which it is arriving) and not to the outgoing
2757 callout (which could be using a different interface). This has been
2758 changed to use the value of the helo_data option from the smtp transport
2759 instead - this is what is used when a message is actually being sent. If
2760 there is no remote transport (possible with a router that sets up host
2761 addresses), $smtp_active_hostname is used.
2763 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
2764 tweaks were necessary in order to get it to work (see also 21 below):
2765 (a) The code assumed that strncpy() returns a negative number on buffer
2766 overflow, which isn't the case. Replaced with Exim's string_format()
2768 (b) There were several signed/unsigned issues. I just did the minimum
2769 hacking in of casts. There is scope for a larger refactoring.
2770 (c) The code used strcasecmp() which is not a standard C function.
2771 Replaced with Exim's strcmpic() function.
2772 (d) The code set only $1; it now sets $auth1 as well.
2773 (e) A simple test gave the error "authentication client didn't specify
2774 service in request". It would seem that Dovecot has changed its
2775 interface. Fortunately there's a specification; I followed it and
2776 changed what the client sends and it appears to be working now.
2778 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
2781 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
2782 address A is aliased to B and C, where B exists and C does not. Without
2783 -v the output is "A verified" because verification stops after a
2784 successful redirection if more than one address is generated. However,
2785 with -v the child addresses are also verified. Exim was outputting "A
2786 failed to verify" and then showing the successful verification for C,
2787 with its parentage. It now outputs "B failed to verify", showing B's
2788 parentage before showing the successful verification of C.
2790 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
2791 look up a TXT record in a specific list after matching in a combined
2794 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
2795 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
2796 they consult the DNS. I had assumed they would set it the way they
2797 wanted; and indeed my experiments on Linux seem to show that in some
2798 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
2799 To be on the safe side, however, I have now made the interface to
2800 host_find_byname() similar to host_find_bydns(), with an argument
2801 containing the DNS resolver options. The host_find_byname() function now
2802 sets these options at its start, just as host_find_bydns() does. The smtp
2803 transport options dns_qualify_single and dns_search_parents are passed to
2804 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
2805 of host_find_byname() use the default settings of RES_DEFNAMES
2806 (qualify_single) but not RES_DNSRCH (search_parents).
2808 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
2809 spool_read_header() do less string testing, by means of a preliminary
2810 switch on the second character of optional "-foo" lines. (This is
2811 overdue, caused by the large number of possibilities that now exist.
2812 Originally there were few.) While I was there, I also converted the
2813 str(n)cmp tests so they don't re-test the leading "-" and the first
2814 character, in the hope this might squeeze out yet more improvement.
2816 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
2817 flag allowing group syntax was set by the header_syntax check but not
2818 turned off, possible causing trouble later; (2) The flag was not being
2819 set at all for the header_verify test, causing "group"-style headers to
2820 be rejected. I have now set it in this case, and also caused header_
2821 verify to ignore an empty address taken from a group. While doing this, I
2822 came across some other cases where the code for allowing group syntax
2823 while scanning a header line wasn't quite right (mostly, not resetting
2824 the flag correctly in the right place). These bugs could have caused
2825 trouble for malformed header lines. I hope it is now all correct.
2827 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
2828 with the "reply" argument non-NULL. The code, however (which originally
2829 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
2830 but it didn't always do it. This confused somebody who was copying the
2831 code for some other use. I have removed all the tests.
2833 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
2834 feature that was used to support insecure browsers during the U.S. crypto
2835 embargo. It requires special client support, and Exim is probably the
2836 only MTA that supported it -- and would never use it because real RSA is
2837 always available. This code has been removed, because it had the bad
2838 effect of slowing Exim down by computing (never used) parameters for the
2839 RSA_EXPORT functionality.
2841 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
2842 authenticator to fail if there's a tab character in the incoming data
2843 (there should never be unless someone is messing about, as it's supposed
2844 to be base64-encoded). Also added, on Timo's advice, the "secured" option
2845 if the connection is using TLS or if the remote IP is the same as the
2846 local IP, and the "valid-client-cert option" if a client certificate has
2849 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
2850 authenticators. This can be used for authorization after authentication
2851 succeeds. (In the case of plaintext, it servers for both authentication
2854 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
2855 if any retry times were supplied.
2857 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
2858 connection, where there is no client IP address to check. In this
2859 situation, the verify now always succeeds.
2861 PH/25 Applied John Jetmore's -Mset patch.
2863 PH/26 Added -bem to be like -Mset, but loading a message from a file.
2865 PH/27 In a string expansion for a processed (not raw) header when multiple
2866 headers of the same name were present, leading whitespace was being
2867 removed from all of them, but trailing whitespace was being removed only
2868 from the last one. Now trailing whitespace is removed from each header
2869 before concatenation. Completely empty headers in a concatenation (as
2870 before) are ignored.
2872 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
2873 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
2875 PH/29 [Removed. This was a change that I later backed out, and forgot to
2876 correct the ChangeLog entry (that I had efficiently created) before
2877 committing the later change.]
2879 PH/30 Exim was sometimes attempting to deliver messages that had suffered
2880 address errors (4xx response to RCPT) over the same connection as other
2881 messages routed to the same hosts. Such deliveries are always "forced",
2882 so retry times are not inspected. This resulted in far too many retries
2883 for the affected addresses. The effect occurred only when there were more
2884 hosts than the hosts_max_try setting in the smtp transport when it had
2885 the 4xx errors. Those hosts that it had tried were not added to the list
2886 of hosts for which the message was waiting, so if all were tried, there
2887 was no problem. Two fixes have been applied:
2889 (i) If there are any address or message errors in an SMTP delivery, none
2890 of the hosts (tried or untried) are now added to the list of hosts
2891 for which the message is waiting, so the message should not be a
2892 candidate for sending over the same connection that was used for a
2893 successful delivery of some other message. This seems entirely
2894 reasonable: after all the message is NOT "waiting for some host".
2895 This is so "obvious" that I'm not sure why it wasn't done
2896 previously. Hope I haven't missed anything, but it can't do any
2897 harm, as the worst effect is to miss an optimization.
2899 (ii) If, despite (i), such a delivery is accidentally attempted, the
2900 routing retry time is respected, so at least it doesn't keep
2901 hammering the server.
2903 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
2904 in ${readsocket because some servers need this prod.
2906 PH/32 Added some extra debug output when updating a wait-xxx database.
2908 PH/33 The hint "could be header name not terminated by colon", which has been
2909 given for certain expansion errors for a long time, was not being given
2910 for the ${if def:h_colon_omitted{... case.
2912 PH/34 The spec says: "With one important exception, whenever a domain list is
2913 being scanned, $domain contains the subject domain." There was at least
2914 one case where this was not true.
2916 PH/35 The error "getsockname() failed: connection reset by peer" was being
2917 written to the panic log as well as the main log, but it isn't really
2918 panic-worthy as it just means the connection died rather early on. I have
2919 removed the panic log writing for the ECONNRESET error when getsockname()
2922 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
2923 runs only) independently of the message's sender address. This meant
2924 that, if the 4xx error was in fact related to the sender, a different
2925 message to the same recipient with a different sender could confuse
2926 things. In particular, this can happen when sending to a greylisting
2927 server, but other circumstances could also provoke similar problems.
2928 I have changed the default so that the retry time for these errors is now
2929 based a combination of the sender and recipient addresses. This change
2930 can be overridden by setting address_retry_include_sender=false in the
2933 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
2934 remote server are returned as part of bounce messages. This was not
2935 happening for LMTP over a pipe (the lmtp transport), but now it is the
2936 same for both kinds of LMTP.
2938 PH/38 Despite being documented as not happening, Exim was rewriting addresses
2939 in header lines that were in fact CNAMEs. This is no longer the case.
2941 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
2942 and queue runs started by the daemon processed all messages. This has
2943 been fixed so that -R and -S can now usefully be given with -q<time>.
2945 PH/40 Import PCRE release 6.7 (fixes some bugs).
2947 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
2949 PH/42 Give an error if -q is specified more than once.
2951 PH/43 Renamed the variables $interface_address and $interface_port as
2952 $received_ip_address and $received_port, to make it clear that these
2953 values apply to message reception, and not to the outgoing interface when
2954 a message is delivered. (The old names remain recognized, of course.)
2956 PH/44 There was no timeout on the connect() call when using a Unix domain
2957 socket in the ${readsocket expansion. There now is.
2959 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
2960 be meaningful with "accept".
2962 SC/01 Eximstats V1.43
2963 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
2965 SC/02 Eximstats V1.44
2966 Use a glob alias rather than an array ref in the generated
2967 parser. This improves both readability and performance.
2969 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
2970 Collect SpamAssassin and rejection statistics.
2971 Don't display local sender or destination tables unless
2972 there is data to show.
2973 Added average volumes into the top table text output.
2975 SC/04 Eximstats V1.46
2976 Collect data on the number of addresses (recipients)
2977 as well as the number of messages.
2979 SC/05 Eximstats V1.47
2980 Added 'Message too big' to the list of mail rejection
2981 reasons (thanks to Marco Gaiarin).
2983 SC/06 Eximstats V1.48
2984 Mainlog lines which have GMT offsets and are too short to
2985 have a flag are now skipped.
2987 SC/07 Eximstats V1.49 (Alain Williams)
2988 Added the -emptyok flag.
2990 SC/08 Eximstats V1.50
2991 Fixes for obtaining the IP address from reject messages.
2993 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
2994 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
2995 whitespace changes from 4.64-PH/27
2997 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
3000 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
3001 are found, allow negative numbers in numeric criteria)
3003 JJ/06 exipick.20061117.2, added new $message_body_missing variable
3005 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
3006 to match changes made in 4.64-PH/43
3008 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
3010 PH/47 Put in an explicit test for a DNS lookup of an address record where the
3011 "domain" is actually an IP address, and force a failure. This locks out
3012 those revolvers/nameservers that support "A-for-A" lookups, in
3013 contravention of the specifications.
3015 PH/48 When a host name was looked up from an IP address, and the subsequent
3016 forward lookup of the name timed out, the host name was left in
3017 $sender_host_name, contrary to the specification.
3019 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
3020 restricted to single-key lookups, Exim was not diagnosing an error if
3021 * or *@ was used with a query-style lookup.
3023 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
3025 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
3026 long ago, but noone interested enough thought of it. Let's just say that
3027 the "1.1" means that there are some new functions that weren't there at
3028 some point in the past.
3030 PH/51 Error processing for expansion failure of helo_data from an smtp
3031 transport during callout processing was broken.
3033 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
3034 tested/used via the -bh/-bhc/-bs options.
3036 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
3037 bug, fixed in subsequent PCRE releases).
3039 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
3040 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
3042 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
3048 SC/01 Use a glob alias rather than an array ref in eximstats generated
3049 parser. This improves both readability and performance.
3051 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
3052 Don't display local sender or destination tables in eximstats unless
3053 there is data to show.
3054 Added average volumes into the eximstats top table text output.
3056 SC/03 Collect data on the number of addresses (recipients) as well
3057 as the number of messages in eximstats.
3059 TF/01 Correct an error in the documentation for the redirect router. Exim
3060 does (usually) call initgroups() when daemonizing.
3062 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
3063 with consistent privilege compared to when running as a daemon.
3065 TF/03 Note in the spec that $authenticated_id is not set for local
3066 submissions from trusted users.
3068 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
3069 Thanks to Dean Brooks <dean@iglou.com> for the patch.
3071 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
3072 by adding some example configuration directives to the default
3073 configuration file. A little bit of work is required to uncomment the
3074 directives and define how usernames and passwords are checked, but
3075 there is now a framework to start from.
3077 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
3078 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
3079 without this. I don't know how relevant this is to other LDAP libraries.
3081 PH/02 Add the verb name to the "unknown ACL verb" error.
3083 PH/03 Magnus Holmgren's patch for filter_prepend_home.
3085 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
3087 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
3088 directory not expanded when it should be if an expanded home directory
3089 was set for the address (which is overridden by the transport).
3091 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
3094 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
3095 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
3096 because it is too late at that time, and has no effect.
3098 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
3099 security issue with \' (bugzilla #107). I could not use the
3100 PQescapeStringConn() function, because it needs a PGconn value as one of
3103 PH/08 When testing addresses using -bt, indicate those final addresses that
3104 are duplicates that would not cause an additional delivery. At least one
3105 person was confused, thinking that -bt output corresponded to deliveries.
3106 (Suppressing duplicates isn't a good idea as you lose the information
3107 about possibly different redirections that led to the duplicates.)
3109 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
3110 systems where poll() doesn't work, in particular OS X.
3112 PH/10 Added more information to debugging output for retry time not reached.
3114 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
3115 operations in malware.c.
3117 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
3120 PH/13 If write_rejectlog was set false when logging was sent to syslog with
3121 syslog_duplication set false, log lines that would normally be written
3122 both the the main log and to the reject log were not written to syslog at
3125 PH/14 In the default configuration, change the use of "message" in ACL warn
3126 statements to "add_header".
3128 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
3129 not followed by a command (e.g. "seen endif").
3131 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
3132 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
3135 PH/17 Added extra conditions to the default value of delay_warning_condition
3139 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
3140 { match{$h_precedence:}{(?i)bulk|list|junk} } \
3141 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
3144 The Auto-Submitted: and various List- headers are standardised, whereas I
3145 don't think Precedence: ever was.
3147 PH/18 Refactored debugging code in route_finduser() to show more information,
3148 in particular, the error code if getpwnam() issues one.
3150 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
3151 This is apparently needed in addition to the PH/07 change above to avoid
3152 any possible encoding problems.
3154 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
3155 but not after initializing Perl.
3157 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
3158 output them only if debugging. By default they are written stderr,
3159 apparently, which is not desirable.
3161 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
3164 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
3167 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
3169 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
3170 authenticated or an ident call has been made. Suppress the default
3171 values for $authenticated_id and $authenticated_sender (but permit -oMai
3172 and -oMas) when testing with -bh.
3174 PH/24 Re-jigged the order of the tests in the default configuration so that the
3175 tests for valid domains and recipients precede the DNS black list and CSA
3176 tests, on the grounds that those ones are more expensive.
3178 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
3179 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
3180 This bug exists in every version of Exim that I still have, right back to
3183 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
3184 However, an attempt to turn on case-sensitivity in a regex key by
3185 including (?-i) didn't work because the subject string was already
3186 lowercased, and the effects were non-intuitive. It turns out that a
3187 one-line patch can be used to allow (?-i) to work as expected.
3193 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
3194 other effects) broke the use of negated acl sub-conditions.
3196 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
3199 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
3200 "Deny" causes Exim to reject the incoming connection with a 554 error.
3201 Unfortunately, if there is a major crisis, such as a disk failure,
3202 tcp-wrappers gives "deny", whereas what one would like would be some
3203 kind of temporary error. A kludge has been added to help with this.
3204 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
3205 554 error is used if errno is still zero or contains ENOENT (which occurs
3206 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
3209 PH/03 Add -lutil to the default FreeBSD LIBS setting.
3211 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
3212 errors. Otherwise a message that provokes a temporary error (when other
3213 messages do not) can cause a whole host to time out.
3215 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
3216 addresses were routed directly to files or pipes from a redirect router.
3217 File deliveries just didn't batch; pipe deliveries might have suffered
3220 PH/06 A failure to get a lock for a hints database would erroneously always say
3221 "Failed to get write lock", even when it was really a read lock.
3223 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
3224 of 0600. This has been changed to use the value of the lockfile_mode
3225 option (which defaults to 0600).
3227 PH/08 Applied small patch from the Sieve maintainer.
3229 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
3230 folder from quota calculations, a direct delivery into this folder messed
3231 up the contents of the maildirsize file. This was because the regex was
3232 used only to exclude .Trash (or whatever) when the size of the mailbox
3233 was calculated. There was no check that a delivery was happening into an
3234 excluded directory. This bug has been fixed by ignoring all quota
3235 processing for deliveries into excluded directories.
3237 PH/10 Added the maildirfolder_create_regex option to appendfile.
3243 PH/01 The code for finding all the local interface addresses on a FreeBSD
3244 system running IPv6 was broken. This may well have applied to all BSD
3245 systems, as well as to others that have similar system calls. The broken
3246 code found IPv4 interfaces correctly, but gave incorrect values for the
3247 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
3248 that it would not match correctly against @[] and not recognize the IPv6
3251 PH/02 The ipliteral router was not recognizing addresses of the form user@
3252 [ipv6:....] because it didn't know about the "ipv6:" prefix.
3254 PH/03 Added disable_ipv6.
3256 PH/04 Changed $reply_address to use the raw form of the headers instead of the
3257 decoded form, because it is most often used to construct To: headers
3258 lines in autoreplies, and the decoded form may well be syntactically
3259 invalid. However, $reply_address has leading white space removed, and all
3260 newlines turned into spaces so that the autoreply transport does not
3263 PH/05 If group was specified without a user on a router, and no group or user
3264 was specified on a transport, the group from the router was ignored.
3266 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
3267 for visible compile-time settings that can be used to change these
3268 numbers, for those that want even more. Backwards compatibility with old
3269 spool files has been maintained. However, going back to a previous Exim
3270 release will lost any variables that are in spool files.
3272 PH/07 Two small changes when running in the test harness: increase delay when
3273 passing a TCP/IP connection to a new process, in case the original
3274 process has to generate a bounce, and remove special handling of
3275 127.0.0.2 (sic), which is no longer necessary.
3277 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
3278 be the same on different OS.
3280 PH/09 Moved a debug statement in filter processing to avoid a race problem when
3283 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
3284 whether --show-vars was specified or not
3286 JJ/02 exipick: Added support for new ACL variable spool format introduced
3289 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
3290 syntactically invalid From: or Reply-to: line, and a filter used this to
3291 generate an autoreply, and therefore failed to obtain an address for the
3292 autoreply, Exim could try to deliver to a non-existent relative file
3293 name, causing unrelated and misleading errors. What now happens is that
3294 it logs this as a hard delivery error, but does not attempt to create a
3297 PH/11 The exinext utility has a -C option for testing purposes, but although
3298 the given file was scanned by exinext itself; it wasn't being passed on
3299 when Exim was called.
3301 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
3302 an end-of-file indication when reading a command response.
3304 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
3305 compiled. In many other places in Exim, IPv6 addresses are always
3306 recognized, so I have changed this. It also means that IPv4 domain
3307 literals of the form [IPV4:n.n.n.n] are now always recognized.
3309 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
3310 used if the router is not running as root, for example, when verifying at
3311 ACL time, or when using -bh. The debugging output from this situation was
3312 non-existent - all you got was a failure to exec. I have made two
3315 (a) Failures to set uid/gid, the current directory, or a process leader
3316 in a subprocess such as that created by queryprogram now generate
3317 suitable debugging output when -d is set.
3319 (b) The queryprogram router detects when it is not running as root,
3320 outputs suitable debugging information if -d is set, and then runs
3321 the subprocess without attempting to change uid/gid.
3323 PH/15 Minor change to Makefile for building test_host (undocumented testing
3326 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
3327 additional section of a DNS packet that returns MX or SRV records.
3328 Instead, it always explicitly searches for A/AAAA records. This avoids
3329 major problems that occur when a DNS server includes only records of one
3330 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
3331 fixed another bug: if SRV records were looked up and the corresponding
3332 address records were *not* found in the additional section, the port
3333 values from the SRV records were lost.
3335 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
3336 using the correct key (the original address) when searching the retry
3337 rules in order to find which one to use for generating the retry hint.
3339 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
3340 adding the default one. Similarly, if it contains a Reply-To: header, the
3341 errors_reply_to option, if set, is not used.
3343 PH/19 When calculating a retry time, Exim used to measure the "time since
3344 failure" by looking at the "first failed" field in the retry record. Now
3345 it does not use this if it is later than than the arrival time of the
3346 message. Instead it uses the arrival time. This makes for better
3347 behaviour in cases where some deliveries succeed, thus re-setting the
3348 "first failed" field. An example is a quota failure for a huge message
3349 when small messages continue to be delivered. Without this change, the
3350 "time since failure" will always be short, possible causing more frequent
3351 delivery attempts for the huge message than are intended.
3352 [Note: This change was subsequently modified - see PH/04 for 4.62.]
3354 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
3355 $1, $2, $3) because the numerical variables can be reset during some
3356 expansion items (e.g. "match"), thereby losing the authentication data.
3358 PH/21 Make -bV show the size of off_t variables so that the test suite can
3359 decide whether to run tests for quotas > 2G.
3361 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
3362 mailbox_size, and mailbox_filecount in the appendfile transport. If a
3363 filecount value is greater than 2G or if a quota value is greater than 2G
3364 on a system where the size of off_t is not greater than 4, a panic error
3367 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
3368 never match. The debug and -bh output now contains an explicit error
3369 message indicating a malformed IPv4 address or mask.
3371 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
3372 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
3373 PH/23 above applies.
3375 PH/25 Do not write to syslog when running in the test harness. The only
3376 occasion when this arises is a failure to open the main or panic logs
3377 (for which there is an explicit test).
3379 PH/26 Added the /no_tell option to "control=freeze".
3381 PH/27 If a host name lookup failed very early in a connection, for example, if
3382 the IP address matched host_lookup and the reverse lookup yielded a name
3383 that did not have a forward lookup, an error message of the form "no IP
3384 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
3385 could be logged. Now it outputs the IP address instead of "NULL".
3387 PH/28 An enabling patch from MH: add new function child_open_exim2() which
3388 allows the sender and the authenticated sender to be set when
3389 submitting a message from within Exim. Since child_open_exim() is
3390 documented for local_scan(), the new function should be too.
3392 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
3393 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
3394 results in an empty string is now treated as unset.
3396 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
3398 PH/31 Added sender_verify_fail logging option.
3400 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
3401 needed by RFC 822 but not by RFC 2822 was commented out. I have now
3402 tidied the source and removed it altogether.
3404 PH/33 When a queue run was abandoned because the load average was too high, a
3405 log line was always written; now it is written only if the queue_run log
3406 selector is set. In addition, the log line for abandonment now contains
3407 information about the queue run such as the pid. This is always present
3408 in "start" and "stop" lines but was omitted from the "abandon" line.
3410 PH/34 Omit spaces between a header name and the colon in the error message that
3411 is given when verify = headers_syntax fails (if there are lots of them,
3412 the message gets confusing).
3414 PH/35 Change the default for dns_check_names_pattern to allow slashes within
3415 names, as there are now some PTR records that contain slashes. This check
3416 is only to protect against broken name servers that fall over on strange
3417 characters, so the fact that it applies to all lookups doesn't matter.
3419 PH/36 Now that the new test suite is complete, we can remove some of the
3420 special code in Exim that was needed for the old test suite. For example,
3421 sorting DNS records because real resolvers return them in an arbitrary
3422 order. The new test suite's fake resolver always returns records in the
3425 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
3426 bounces) except when queue_only is set, to avoid logging races between
3427 the different processes.
3429 PH/38 Panic-die if .include specifies a non-absolute path.
3431 PH/39 A tweak to the "H" retry rule from its user.
3433 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
3434 a label. They prevented compilation on older perls.
3436 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
3437 a warning to be raised on newish perls.
3439 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
3440 on queue. Changes to match documented behaviour of showing count of
3441 messages matching specified criteria.
3443 PH/40 Changed the default ident timeout from 30s to 5s.
3445 PH/41 Added support for the use of login_cap features, on those BSD systems
3446 that have them, for controlling the resources used by pipe deliveries.
3448 PH/42 The content-scanning code uses fopen() to create files in which to put
3449 message data. Previously it was not paying any attention to the mode of
3450 the files. Exim runs with umask(0) because the rest of the code creates
3451 files with open(), and sets the required mode explicitly. Thus, these
3452 files were ending up world-writeable. This was not a big issue, because,
3453 being within the spool directory, they were not world-accessible. I have
3454 created a function called modefopen, which takes an additional mode
3455 argument. It sets umask(777), creates the file, chmods it to the required
3456 mode, then resets the umask. All the relevant calls to fopen() in the
3457 content scanning code have been changed to use this function.
3459 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
3460 to 24 hours. This avoids potential overflow problems when processing G
3461 and H retry rules. I suspect nobody ever tinkers with this value.
3463 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
3465 PH/45 When the plaintext authenticator is running as a client, the server's
3466 challenges are checked to ensure they are valid base64 strings. By
3467 default, the authentication attempt is cancelled if an invalid string is
3468 received. Setting client_ignore_invalid_base64 true ignores these errors.
3469 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
3470 they are received. Thus, the responses can be made to depend on the
3471 challenges. If an invalid string is ignored, an empty string is placed in
3474 PH/46 Messages that are created by the autoreply transport now contains a
3475 References: header, in accordance with RFCs 2822 and 3834.
3477 PH/47 Added authenticated_sender_force to the smtp transport.
3479 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
3481 PH/49 Installed latest patch from the Sieve maintainer.
3483 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
3484 was also set, the appendfile transport was unnecessarily scanning a
3485 directory of message files (e.g. for maildir delivery) to find the count
3486 of files (along with the size), even though it did not need this
3487 information. It now does the scan only if it needs to find either the
3488 size of the count of files.
3490 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
3492 PH/52 Two bugs concerned with error handling when the smtp transport is
3495 (i) Exim was not creating retry information for temporary errors given
3496 for individual recipients after the DATA command when the smtp transport
3497 was used in LMTP mode. This meant that they could be retried too
3498 frequently, and not timed out correctly.
3500 (ii) Exim was setting the flag that allows error details to be returned
3501 for LMTP errors on RCPT commands, but not for LMTP errors for individual
3502 recipients that were returned after the DATA command.
3504 PH/53 This is related to PH/52, but is more general: for any failing address,
3505 when detailed error information was permitted to be returned to the
3506 sender, but the error was temporary, then after the final timeout, only
3507 "retry timeout exceeded" was returned. Now it returns the full error as
3508 well as "retry timeout exceeded".
3510 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
3511 do this, and (what is worse) MTAs that accept it.
3513 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
3514 will now be deprecated.
3516 PH/56 New os.c-cygwin from the Cygwin maintainer.
3518 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
3519 formats (previously only available in exim formats via -bpr, -bpru,
3520 and -bpra. Now also available in native and exiqgrep formats)
3522 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
3523 with very large, slow to parse queues
3525 JJ/08 exipick: added ! as generic prefix to negate any criteria format
3527 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
3529 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
3530 responses to authentication challenges, though it was showing the
3531 challenges; (ii) I've removed the CR characters from the debug output for
3534 PH/58 Allow for the insertion of a newline as well as a space when a string
3535 is turned into more than one encoded-word during RFC 2047 encoding. The
3536 Sieve code now uses this.
3538 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
3539 data_4xx, lost_connection, tls_required.
3541 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
3542 message was being sent as an SMTP response.
3544 PH/61 Add -l and -k options to exicyclog.
3546 PH/62 When verifying, if an address was redirected to one new address, so that
3547 verification continued, and the new address failed or deferred after
3548 having set something in $address_data, the value of $address_data was not
3549 passed back to the ACL. This was different to the case when no
3550 redirection occurred. The value is now passed back in both cases.
3552 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
3553 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
3554 use login_cap.h, so on its own it isn't the distinguishing feature. The
3555 new name refers directly to the setclassresources() function.
3557 PH/65 Added configuration files for NetBSD3.
3559 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
3561 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
3562 is preferred over IPv4.
3564 PH/68 The bounce_return_message and bounce_return_body options were not being
3565 honoured for bounces generated during the reception of non-SMTP messages.
3566 In particular, this applied to messages rejected by the ACL. This bug has
3567 been fixed. However, if bounce_return_message is true and bounce_return_
3568 body is false, the headers that are returned for a non-SMTP message
3569 include only those that have been read before the error was detected.
3570 (In the case of an ACL rejection, they have all been read.)
3572 PH/69 The HTML version of the specification is now built in a directory called
3573 spec_html instead of spec.html, because the latter looks like a path with
3574 a MIME-type, and this confuses some software.
3576 PH/70 Catch two compiler warnings in sieve.c.
3578 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
3579 function verify_get_ident() calls ip_connect() to connect a socket, but
3580 if the "connect()" function timed out, ip_connect() used to close the
3581 socket. However, verify_get_ident() also closes the socket later, and in
3582 between Exim writes to the log, which may get opened at this point. When
3583 the socket was closed in ip_connect(), the log could get the same file
3584 descriptor number as the socket. This naturally causes chaos. The fix is
3585 not to close the socket in ip_connect(); the socket should be closed by
3586 the function that creates it. There was only one place in the code where
3587 this was missing, in the iplookup router, which I don't think anybody now
3588 uses, but I've fixed it anyway.
3590 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
3591 well as to direct DNS lookups. Otherwise the handling of names in host
3592 lists is inconsistent and therefore confusing.
3598 PH/01 Two changes to the default runtime configuration:
3600 (1) Move the checks for relay_from_hosts and authenticated clients from
3601 after to before the (commented out) DNS black list checks.
3603 (2) Add control=submission to the relay_from_hosts and authenticated
3604 clients checks, on the grounds that messages accepted by these
3605 statements are most likely to be submissions.
3607 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
3609 (1) Generate an error if the third argument for the ${prvs expansion is
3612 (2) Treat a missing third argument of ${prvscheck as if it were an empty
3615 (3) Reset the variables that are obtained from the first argument of
3616 ${prvscheck and used in the second argument before leaving the code,
3617 because their memory is reclaimed, so using them afterwards may do
3620 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
3621 one (it's much easier than Tom thought :-).
3623 (5) Because of (4), we can now allow for the use of $prvscheck_result
3624 inside the third argument.
3626 PH/03 For some reason, the default setting of PATH when running a command from
3627 a pipe transport was just "/usr/bin". I have changed it to
3630 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
3631 anything to be listed in the output from -bV.
3633 PH/05 When a filter generated an autoreply, the entire To: header line was
3634 quoted in the delivery log line, like this:
3636 => >A.N.Other <ano@some.domain> <original@ddress> ...
3638 This has been changed so that it extracts the operative address. There
3639 may be more than one such address. If so, they are comma-separated, like
3642 => >ano@some.domain,ona@other.domain <original@ddress> ...
3644 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
3645 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
3646 not being looked up in the rDNS to get a host name, Exim was showing the
3647 IP address twice in Received: lines, even though the IP addresses were
3648 identical. For example:
3650 Received: from [1.2.3.4] (helo=[1.2.3.4])
3652 However, if the real host name was known, it was omitting the HELO data
3653 if it matched the actual IP address. This has been tidied up so that it
3654 doesn't show the same IP address twice.
3656 PH/07 When both +timestamp and +memory debugging was on, the value given by
3657 $tod_xxx expansions could be wrong, because the tod_stamp() function was
3658 called by the debug printing, thereby overwriting the timestamp buffer.
3659 Debugging no longer uses the tod_stamp() function when +timestamp is set.
3661 PH/08 When the original message was included in an autoreply transport, it
3662 always said "this is a copy of the message, including all the headers",
3663 even if body_only or headers_only was set. It now gives an appropriate
3666 PH/09 Applied a patch from the Sieve maintainer which:
3668 o fixes some comments
3669 o adds the (disabled) notify extension core
3670 o adds some debug output for the result of if/elsif tests
3671 o points to the current vacation draft in the documentation
3672 and documents the missing references header update
3676 o fixes a bug in processing the envelope test (when testing
3677 multiple envelope elements, the last element determined the
3680 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
3681 Electronic Mail") by including:
3683 Auto-submitted: auto-generated
3685 in the messages that it generates (bounce messages and others, such as
3686 warnings). In the case of bounce messages for non-SMTP messages, there was
3687 also a typo: it was using "Auto_submitted" (underscore instead of
3688 hyphen). Since every message generated by Exim is necessarily in response
3689 to another message, thes have all been changed to:
3691 Auto-Submitted: auto-replied
3693 in accordance with these statements in the RFC:
3695 The auto-replied keyword:
3697 - SHOULD be used on messages sent in direct response to another
3698 message by an automatic process,
3700 - MUST NOT be used on manually-generated messages,
3702 - MAY be used on Delivery Status Notifications (DSNs) and Message
3703 Disposition Notifications (MDNs),
3705 - MUST NOT be used on messages generated by automatic or periodic
3706 processes, except for messages which are automatic responses to
3709 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
3710 to the default Received: header definition.
3712 PH/12 Added log selector acl_warn_skipped (default on).
3714 PH/13 After a successful wildlsearch lookup, discard the values of numeric
3715 variables because (a) they are in the wrong storage pool and (b) even if
3716 they were copied, it wouldn't work properly because of the caching.
3718 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
3719 checking when decoding. Apparently there are clients that generate
3720 overlong encoded strings. Why am I not surprised?
3722 PH/15 If the first argument of "${if match_address" was not empty, but did not
3723 contain an "@" character, Exim crashed. Now it writes a panic log message
3724 and treats the condition as false.
3726 PH/16 In autoreply, treat an empty string for "once" the same as unset.
3728 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
3729 extension "envelope-auth". The code is finished and in agreement with
3730 other implementations, but there is no documentation so far and in fact,
3731 nobody wrote the draft yet. This extension is currently #undef'ed, thus
3732 not changing the active code.
3734 Print executed "if" and "elsif" statements when debugging is used. This
3735 helps a great deal to understand what a filter does.
3737 Document more things not specified clearly in RFC3028. I had all this
3738 sorted out, when out of a sudden new issues came to my mind. Oops."
3740 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
3743 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
3744 canonical form (as documented). However, after a host name lookup from
3745 the IP address, check_host() was doing a simple string comparison with
3746 addresses acquired from the DNS when checking that the found name did
3747 have the original IP as one of its addresses. Since any found IPv6
3748 addresses are likely to be in abbreviated form, the comparison could
3749 fail. Luckily, there already exists a function for doing the comparison
3750 by converting both addresses to binary, so now that is used instead of
3751 the text comparison.
3753 PH/20 There was another similar case to PH/19, when a complete host name was
3754 given in a host list; looking up its IP address could give an abbreviated
3755 form, whereas the current host's name might or might not be abbreviated.
3756 The same fix has been applied.
3762 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
3763 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
3766 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
3767 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
3769 PH/03 Typo: missing ".o" in src/pcre/Makefile.
3771 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
3772 header line, restrict the check to what is listed in RFCs 2369 and 2929.
3773 Also, for "Auto-Submitted", treat anything other than "no" as
3774 non-personal, in accordance with RFC 3834. (Previously it treated
3775 anything starting "auto-" as non-personal.)
3777 TF/01 The control=submission/name=... option had a problem with syntax
3778 errors if the name included a slash character. The /name= option
3779 now slurps the rest of the string, so it can include any characters
3780 but it must come last in the list of options (after /sender_retain
3783 PH/05 Some modifications to the interface to the fake nameserver for the new
3791 TK/01 Added the "success_on_redirect" address verification option. See
3792 NewStuff for rationale and an example.
3794 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
3796 PH/02 Patch to exigrep to allow it to work on syslog lines.
3798 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
3799 fread() to skip over the body file's header line, because in Cygwin the
3800 header line is locked and is inaccessible.
3802 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
3803 co-exist for some time) to make it clear that it is the Exim ID that is
3804 referenced, not the Message-ID: header line.
3806 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
3807 string_format() function, because snprintf() does not exist on all
3810 PH/06 The use of forbid_filter_existstest now also locks out the use of the
3811 ${stat: expansion item.
3813 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
3814 protocol synchronization error", to keep the pedants happy.
3816 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
3817 well as for IRIX systems, when gcc is being used. See the host.c source
3820 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
3822 PH/10 Named domain lists were not working if used in a queue_smtp_domains
3825 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
3826 transport and to the smtp transport in LMTP mode.
3828 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
3830 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
3831 run a filter in a subprocess. This could lead to confusion in subsequent
3832 lookups in the parent process. There should also be a search_tidyup() at
3833 the end of the subprocess.
3835 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
3836 only if the host matched helo_try_verify_hosts, which caused the
3837 verification to occur when the EHLO/HELO command was issued. The ACL just
3838 tested the remembered result. Now, if a previous verification attempt has
3839 not happened, "verify = helo" does it there and then.
3841 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
3843 TK/03 Fix log output including CR from clamd.
3845 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
3846 exist provoked a memory error which could cause a segfault.
3848 PH/15 Installed PCRE 6.2
3850 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
3852 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
3853 of the problem. Specifically, suggested +O2 rather than +O1 for the
3856 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
3858 PH/20 If a delivery was routed to a non-standard port by means of an SRV
3859 record, the port was not correctly logged when the outgoing_port log
3860 selector was set (it logged the transort's default port).
3862 PH/21 Added support for host-specific ports to manualroute, queryprogram,
3863 fallback_hosts, and "hosts" in the smtp transport.
3865 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
3866 host errors such as "Connection refused".
3868 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
3869 authentication with radiusclient 0.4.9:
3871 - Error returned from rc_read_config was caught wrongly
3872 - Username/password not passed on to radius server due to wrong length.
3874 The presumption is that some radiusclient API changes for 4.51/PH/17
3875 were not taken care of correctly. The code is still untested by me (my
3876 Linux distribution still has 0.3.2 of radiusclient), but it was
3877 contributed by a Radius user.
3879 PH/24 When doing a callout, the value of $domain wasn't set correctly when
3880 expanding the "port" option of the smtp transport.
3882 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
3883 while reading a MIME header. Thanks to Tom Hughes for a patch.
3885 PH/24 Include config.h inside local_scan.h so that configuration settings are
3888 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
3889 that in an ACL for RCPT (for example), you can examine exactly what was
3892 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
3893 commands, but it was not correctly comparing the address with the actual
3894 client host address. Thus, it would show the EHLO address in Received:
3895 header lines when this was not necessary.
3897 PH/27 Added the % operator to ${eval:}.
3899 PH/28 Exim tries to create and chdir to its spool directory when it starts;
3900 it should be ignoring failures (because with -C, for example, it has lost
3901 privilege). It wasn't ignoring creation failures other than "already
3904 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
3907 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
3908 because an input line was too long, either on its own, or by virtue of
3909 too many continuations, the temporary file was not being removed, and the
3910 return code was incorrect.
3912 PH/31 Missing "BOOL" in function definition in filtertest.c.
3914 PH/32 Applied Sieve patches from the maintainer.
3916 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
3918 PH/33 Added "verify = not_blind".
3920 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
3921 Local/Makefile (with some defaults set). These are used in built scripts
3922 such as exicyclog, but they have never been used in the exim_install
3923 script (though there are many overriding facilities there). I have
3924 arranged that the exim_install script now takes note of these two
3927 PH/35 Installed configuration files for Dragonfly.
3929 PH/36 When a locally submitted message by a trusted user did not contain a
3930 From: header, and the sender address was obtained from -f or from an SMTP
3931 MAIL command, and the trusted user did not use -F to supply a sender
3932 name, $originator_name was incorrectly used when constructing a From:
3933 header. Furthermore, $originator_name was used for submission mode
3934 messages from external hosts without From: headers in a similar way,
3935 which is clearly wrong.
3937 PH/37 Added control=suppress_local_fixups.
3939 PH/38 When log_selector = +received_sender was set, and the addition of the
3940 sender made the log line's construction buffer exactly full, or one byte
3941 less than full, an overflow happened when the terminating "\n" was
3944 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
3945 when the result of a list match is failure because a DNS lookup failed.
3947 PH/40 RM_COMMAND is now used in the building process.
3949 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
3950 the "build-* directories that it finds.
3952 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
3953 address in a domain literal was a prefix of an interface address.
3955 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
3956 when verifying a sender address, unless rewrite_headers is false.
3958 PH/44 Wrote a long comment about why errors_to addresses are verified as
3959 recipients, not senders.
3961 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
3962 the ratelimit ACL was added.
3964 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
3966 PH/46 Added extra information about PostgreSQL errors to the error string.
3968 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
3969 avoiding the need to install special zones in a real server. This is
3970 backwards compatible; if it can't find the fake resolver, it drops back.
3971 Thus, both old and new test suites can be run.
3973 TF/02 Added util/ratelimit.pl
3975 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
3976 clock is set back in time.
3978 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
3979 Candler <B.Candler@pobox.com>.
3981 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
3982 OK for addresses that are the result of redirections.
3984 PH/48 A number of further additions for the benefit of the new test suite,
3985 including a fake gethostbyname() that interfaces to the fake DNS resolver
3988 TF/06 The fix for widen_domains has also been applied to qualify_single and
3989 search_parents which are the other dnslookup options that can cause
3992 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
3995 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
3997 TF/07 Exim produced the error message "an SRV record indicated no SMTP
3998 service" if it encountered an MX record with an empty target hostname.
3999 The message is now "an MX or SRV record indicated no SMTP service".
4001 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
4002 if the DNS of the sending site is misconfigured. This is quite a
4003 common situation. This change restores the behaviour of treating a
4004 helo verification defer as a failure.
4006 PH/51 If self=fail was set on a router, the bounce message did not include the
4007 actual error message.
4013 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
4015 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
4016 command itself ended in error, the underlying message about the transport
4017 filter timeout was being overwritten with the pipe command error. Now the
4018 underlying error message should be appended to the second error message.
4020 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
4022 PH/02 Reduce the amount of output that "make" produces by default. Full output
4023 can still be requested.
4025 PH/03 The warning log line about a condition test deferring for a "warn" verb
4026 was being output only once per connection, rather than after each
4027 occurrence (because it was using the same function as for successful
4028 "warn" verbs). This seems wrong, so I have changed it.
4030 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
4031 it should not have, which might have caused a crash in the right
4032 circumstances, but probably never did.
4034 PH/04 Installed a modified version of Tony Finch's patch to make submission
4035 mode fix the return path as well as the Sender: header line, and to
4036 add a /name= option so that you can make the user's friendly name appear
4039 TF/03 Added the control = fakedefer ACL modifier.
4041 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
4042 Mark Lowes for thorough testing.
4044 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
4046 TK/03 Merged latest SRS patch from Miles Wilton.
4048 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
4049 with the definition in sysexits.h (which is #included earlier).
4050 Fortunately, Exim does not actually use EX_OK. The code used to try to
4051 preserve the sysexits.h value, by assuming that macro definitions were
4052 scanned for macro replacements. I have been disabused of this notion,
4053 so now the code just undefines EX_OK before #including unistd.h.
4055 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
4056 in the smtp transport. When a block could not be written in a single
4057 write() function, the timeout was being re-applied to each part-write.
4058 This seems wrong - if the receiver was accepting one byte at a time it
4059 would take for ever. The timeout is now adjusted when this happens. It
4060 doesn't have to be particularly precise.
4062 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
4063 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
4065 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
4066 without a domain if the check to <postmaster@domain> fails.
4068 SC/01 Eximstats: added -xls and the ability to specify output files
4069 (patch written by Frank Heydlauf).
4071 SC/02 Eximstats: use FileHandles for outputting results.
4073 SC/03 Eximstats: allow any combination of xls, txt, and html output.
4075 SC/04 Eximstats: fixed display of large numbers with -nvr option
4077 SC/05 Eximstats: fixed merging of reports with empty tables.
4079 SC/06 Eximstats: added the -include_original_destination flag
4081 SC/07 Eximstats: removed tabs and trailing whitespace.
4083 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
4085 TK/06 MBOX spool code: Add real "From " MBOX separator line
4086 so the .eml file is really in mbox format (even though
4087 most programs do not really care). Patch from Alex Miller.
4089 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
4090 The latter is generated from $received_to and is only set if the
4091 message has one envelope recipient. SA can use these headers,
4092 obviously out-of-the-box. Patch from Alex Miller.
4094 PH/08 The ${def test on a variable was returning false if the variable's
4095 value was "0", contrary to what the specification has always said!
4096 The result should be true unless the variable is empty.
4098 PH/09 The syntax error of a character other than { following "${if
4099 def:variable_name" (after optional whitespace) was not being diagnosed.
4100 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
4101 accidental colon was present, for example, could give incorrect results.
4103 PH/10 Tidied the code in a number of places where the st_size field of a stat()
4104 result is used (not including appendfile, where other changes are about
4107 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
4108 This involved changing a lot of size variables from int to off_t. It
4109 should work with maildirs and everything.
4111 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
4112 spamd dying while we are connected to it.
4114 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
4117 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
4118 function for BDB 4.3.
4120 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
4122 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
4123 the sources for PCRE have been re-arranged and more files are now
4126 PH/15 The code I had for printing potentially long long variables in PH/11
4127 above was not the best (it lost precision). The length of off_t variables
4128 is now inspected at build time, and an appropriate printing format (%ld
4129 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
4130 to be "long long int" or "long int". This is needed for the internal
4131 formatting function string_vformat().
4133 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
4134 the configuration file to be ":syslog", then the script "guesses" where
4135 the logs files are, rather than using the compiled in default. In our
4136 case the guess is not the same as the compiled default, so the script
4137 suddenly stopped working when I started to use syslog. The patch checks
4138 to see if log_file_path is "". If so, it attempts to read it from exim
4139 with no configuration file to get the compiled in version, before it
4140 falls back to the previous guessing code."
4142 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
4143 implementing BATV in an Exim configuration. See NewStuff for the gory
4146 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
4147 Makefile that are specific to HP-UX.
4149 PH/18 If the "use_postmaster" option was set for a recipient callout together
4150 with the "random" option, the postmaster address was used as the MAIL
4151 FROM address for the random test, but not for the subsequent recipient
4152 test. It is now used for both.
4154 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
4155 patch removes a few documentation additions to RFC 3028, because the
4156 latest draft now contains them. It adds the new en;ascii-case comparator
4157 and a new error check for 8bit text in MIME parts. Comparator and
4158 require names are now matched exactly. I enabled the subaddress
4159 extension, but it is not well tested yet (read: it works for me)."
4161 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
4162 rework some of the code of TK/09 above to avoid the hardwired use of
4163 "%lld" and "long long". Replaced the call to snprintf() with a call to
4166 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
4167 records point to non-existent hosts", "retry timeout exceeded", and
4168 "retry time not reached for any host after a long failure period".
4170 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
4171 experimental DomainKeys support:
4173 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
4174 (2) On an error such as an illegally used "control", the wrong name for
4175 the control was given.
4177 These problems did NOT occur unless DomainKeys support was compiled.
4179 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
4181 PH/24 Added ${if match_ip condition.
4183 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
4184 databases so that it will be absolutely obvious if a crash occurs in the
4185 DB library. This is a regular occurrence (often caused by mis-matched
4188 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
4189 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
4190 picked up on a user's system that detects such things. There doesn't seem
4191 to be a gcc warning option for this - only an attribute that has to be
4192 put on the function's prototype. It seems that in Fedora Core 4 they have
4193 set this on a number of new functions. No doubt there will be more in due
4196 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
4197 specify a transport. However, if an address that was verified by such a
4198 router was the subject of a callout, Exim crashed because it tried to
4199 read the rcpt_include_affixes from the non-existent transport. Now it
4200 just assumes that the setting of that option is false. This bug was
4201 introduced by 4.51/PH/31.
4203 PH/28 Changed -d+all to exclude +memory, because that information is very
4204 rarely of interest, but it makes the output a lot bigger. People tend to
4205 do -d+all out of habit.
4207 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
4208 code in os-type was giving problems when libc.so lives in lib64, like on
4211 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
4212 aren't the modern standard, and it seems that some systems' include files
4213 don't always have them. Exim was already checking for some of the newer
4214 ones like T_AAAA, and defining it itself. I've added checks for all the
4215 record types that Exim uses.
4217 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
4218 not automatically generating a new one, as it is supposed to. This
4219 prevented TLS from working. If the file did exist, but contained invalid
4220 data, a new version was generated, as expected. It was only the case of a
4221 non-existent file that was broken.
4223 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
4224 with a change in libdomainkeys > 0.64.
4226 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
4227 from DNS. If the selector record carries the flag, it now has
4228 precedence over the domain-wide flag.
4230 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
4232 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
4233 the use of a port name that isn't defined in /etc/services) occurred, the
4234 message was deferred as in a normal delivery, and thus remained on the
4235 spool, instead of being failed because of the mua_wrapper setting. This
4236 is now fixed, and I tidied up some of the mua_wrapper messages at the
4239 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
4240 the messages in a hash of arrays rather than using individual hashes.
4241 This is a bit cleaner and results in dramatic memory savings, albeit
4242 at a slight CPU cost.
4244 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
4245 as requested by Marc Sherman.
4247 SC/10 Eximstats: added histograms for user specified patterns as requested
4250 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
4252 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
4253 fopen() in the content-scanning modules that did not already have it.
4259 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
4260 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
4262 TK/02 Fix ACL "control" statement not being available in MIME ACL.
4264 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
4266 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
4267 to test Sieve filters that use "vacation".
4269 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
4270 that changes the way the GnuTLS parameters are stored in the cache file.
4271 The new format can be generated externally. For backward compatibility,
4272 if the data in the cache doesn't make sense, Exim assumes it has read an
4273 old-format file, and it generates new data and writes a new file. This
4274 means that you can't go back to an older release without removing the
4277 PH/03 A redirect router that has both "unseen" and "one_time" set does not
4278 work if there are any delivery delays because "one_time" forces the
4279 parent to be marked "delivered", so its unseen clone is never tried
4280 again. For this reason, Exim now forbids the simultaneous setting of
4283 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
4284 redirected to themselves ("homonym" addresses). Read the long ChangeLog
4285 entry if you want to know the details. The fix, however, neglected to
4286 consider the case when local delivery batching is involved. The test for
4287 "previously delivered" was not happening when checking to see if an
4288 address could be batched with a previous (undelivered) one; under
4289 certain circumstances this could lead to multiple deliveries to the same
4292 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
4293 in its include files, and this causes problems building Exim.
4295 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
4296 header_syntax) but Exim was just ignoring anything given after a slash.
4297 In particular, this caused confusion with an attempt to use "verify =
4298 reverse_host_lookup/defer_ok". An error is now given when options are
4299 supplied for verify items that do not have them. (Maybe reverse_host_
4300 lookup should have a defer_ok option, but that's a different point.)
4302 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
4303 defined by RFC 821) to 2048, because there were problems with some AUTH
4304 commands, and RFC 1869 says the size should be increased for extended
4305 SMTP commands that take arguments.
4307 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
4310 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
4311 "unknown" error; now it says that the functionality isn't in the binary.
4313 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
4314 an address' error message when a string expansion fails (syntax or
4315 whatever). Otherwise the password may appear in the log. Following change
4316 PH/42 below, there is no longer a chance of it appearing in a bounce
4319 PH/11 Installed exipick version 20050225.0 from John Jetmore.
4321 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
4322 of its addresses was ever tried. (Bugzilla bug #2.)
4324 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
4325 the result incorrectly in the debug output. (It correctly added a newline
4326 to what was transported.)
4328 TF/01 Added $received_time.
4330 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
4331 commented out examples of how to interface to a virus scanner and to
4332 SpamAssassin. Also added commented examples of av_scanner and
4333 spamd_address settings.
4335 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
4336 and controls are allowed in which ACLs. There were a couple of minor
4337 errors. Some of the entries in the conditions table (which is a table of
4338 where they are NOT allowed) were getting very unwieldy; rewrote them as a
4339 negation of where the condition IS allowed.
4341 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
4343 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
4344 header file does not have a version number, so I've had to invent a new
4345 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
4346 API. The code is untested by me (my Linux distribution still has 0.3.2 of
4347 radiusclient), but it was contributed by a Radius user.
4349 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
4350 files or directories, for OpenSSL.
4352 PH/19 When an Exim process that is running as root has to create an Exim log
4353 file, it does so in a subprocess that runs as exim:exim so as to get the
4354 ownership right at creation (otherwise, other Exim processes might see
4355 the file with the wrong ownership). There was no test for failure of this
4356 fork() call, which would lead to the process getting stuck as it waited
4357 for a non-existent subprocess. Forks do occasionally fail when resources
4358 run out. I reviewed all the other calls to fork(); they all seem to check
4361 PH/20 When checking for unexpected SMTP input at connect time (before writing
4362 the banner), Exim was not dealing correctly with a non-positive return
4363 from the read() function. If the client had disconnected by this time,
4364 the result was a log entry for a synchronization error with an empty
4365 string after "input=" when read() returned zero. If read() returned -1
4366 (an event I could not check), uninitialized data bytes were printed.
4367 There were reports of junk text (parts of files, etc) appearing after
4370 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
4372 PH/22 Added support for macro redefinition, and (re)definition in between
4373 driver and ACL definitions.
4375 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
4376 forgetting to use the resulting value; it was using the unexpanded value.
4378 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
4379 hadn't been configured. The fix is from Juergen Kreileder, who
4380 understands it better than I do:
4382 "Here's what I see happening with three configured cyrus_sasl
4383 authenticators configured (plain, login, cram-md5):
4385 On startup auth_cyrus_sasl_init() gets called for each of these.
4386 This means three calls to sasl_listmech() without a specified mech_list.
4387 => SASL tests which mechs of all available mechs actually work
4388 => three warnings about OTP not working
4389 => the returned list contains: plain, login, cram-md5, digest-md5, ...
4391 With the patch, sasl_listmech() also gets called three times. But now
4392 SASL's mech_list option is set to the server_mech specified in the the
4393 authenticator. Or in other words, the answer from sasl_listmech()
4394 gets limited to just the mech you're testing for (which is different
4396 => the return list contains just 'plain' or 'login', 'cram-md5' or
4397 nothing depending on the value of ob->server_mech.
4399 I've just tested the patch: Authentication still works fine,
4400 unavailable mechs specified in the exim configuration are still
4401 caught, and the auth.log warnings about OTP are gone."
4403 PH/25 When debugging is enabled, the contents of the command line are added
4404 to the debugging output, even when log_selector=+arguments is not
4407 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
4408 answer is "GNU", and only if the return is "GNU/something" is the answer
4411 PH/27 $acl_verify_message is now set immediately after the failure of a
4412 verification in an ACL, and so is available in subsequent modifiers. In
4413 particular, the message can be preserved by coding like this:
4415 warn !verify = sender
4416 set acl_m0 = $acl_verify_message
4418 Previously, $acl_verify_message was set only while expanding "message"
4419 and "log_message" when a very denied access.
4421 PH/28 Modified OS/os.c-Linux with
4423 -#ifndef OS_LOAD_AVERAGE
4424 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
4426 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
4427 nomenclature these days.)
4429 PH/29 Installed patch from the Sieve maintainer that adds the options
4430 sieve_useraddress and sieve_subaddress to the redirect router.
4432 PH/30 In these circumstances:
4433 . Two addresses routed to the same list of hosts;
4434 . First host does not offer TLS;
4435 . First host accepts first address;
4436 . First host gives temporary error to second address;
4437 . Second host offers TLS and a TLS session is established;
4438 . Second host accepts second address.
4439 Exim incorrectly logged both deliveries with the TLS parameters (cipher
4440 and peerdn, if requested) that were in fact used only for the second
4443 PH/31 When doing a callout as part of verifying an address, Exim was not paying
4444 attention to any local part prefix or suffix that was matched by the
4445 router that accepted the address. It now behaves in the same way as it
4446 does for delivery: the affixes are removed from the local part unless
4447 rcpt_include_affixes is set on the transport.
4449 PH/32 Add the sender address, as F=<...>, to the log line when logging a
4450 timeout during the DATA phase of an incoming message.
4452 PH/33 Sieve envelope tests were broken for match types other than :is. I have
4453 applied a patch sanctioned by the Sieve maintainer.
4455 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
4456 the uid or gid is negative. A case of a negative gid caused this to be
4457 noticed. The fix allows for either to be negative.
4459 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
4460 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
4463 PH/36 The change PH/12 above was broken. Fixed it.
4465 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
4466 the grounds that routing the same address twice would always produce the
4467 same answer. This might have been true once, but it is certainly no
4468 longer true now. Routing a child address may depend on the previous
4469 routing that produced that child. Some complicated redirection strategies
4470 went wrong when messages had multiple recipients, and made Exim's
4471 behaviour dependent on the order in which the addresses were given.
4473 I have moved the duplicate checking until after the routing is complete.
4474 Exim scans the addresses that are assigned to local and remote
4475 transports, and removes any duplicates. This means that more work will be
4476 done, as duplicates will always all be routed, but duplicates are
4477 presumably rare, so I don't expect this is of any significance.
4479 For deliveries to pipes, files, and autoreplies, the duplicate checking
4480 still happens during the routing process, since they are not going to be
4483 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
4484 It corrects a timeout issue with spamd. This is Ian's comment: "The
4485 background is that sometimes spamd either never reads data from a
4486 connection it has accepted, or it never writes response data. The exiscan
4487 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
4488 blindly assumes that writes won't block so it may never time out."
4490 PH/39 Allow G after quota size as well as K and M.
4492 PH/40 The value set for $authenticated_id in an authenticator may not contain
4493 binary zeroes or newlines because the value is written to log lines and
4494 to spool files. There was no check on this. Now the value is run through
4495 the string_printing() function so that such characters are converted to
4496 printable escape sequences.
4498 PH/41 $message_linecount is a new variable that contains the total number of
4499 lines in the message. Compare $body_linecount, which is the count for the
4502 PH/42 Exim no longer gives details of delivery errors for specific addresses in
4503 bounce and delay warning messages, except in certain special cases, which
4506 (a) An SMTP error message from a remote host;
4507 (b) A message specified in a :fail: redirection;
4508 (c) A message specified in a "fail" command in a system filter;
4509 (d) A message specified in a FAIL return from the queryprogram router;
4510 (e) A message specified by the cannot_route_message router option.
4512 In these cases only, Exim does include the error details in bounce and
4513 warning messages. There are also a few cases where bland messages such
4514 as "unrouteable address" or "local delivery error" are given.
4516 PH/43 $value is now also set for the "else" part of a ${run expansion.
4518 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
4519 being worked on, but at least Exim now implements the latest version to
4522 PH/45 In a pipe transport, although a timeout while waiting for the pipe
4523 process to complete was treated as a delivery failure, a timeout while
4524 writing the message to the pipe was logged, but erroneously treated as a
4525 successful delivery. Such timeouts include transport filter timeouts. For
4526 consistency with the overall process timeout, these timeouts are now
4527 treated as errors, giving rise to delivery failures by default. However,
4528 there is now a new Boolean option for the pipe transport called
4529 timeout_defer, which, if set TRUE, converts the failures into defers for
4530 both kinds of timeout. A transport filter timeout is now identified in
4533 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
4534 systems where "make" and "gmake" are different, calling "gmake" at top
4535 level broke things. I've arranged for the value of $(MAKE) to be passed
4536 from the Makefile to this script so that it can call the same version of
4540 A note about Exim versions 4.44 and 4.50
4541 ----------------------------------------
4543 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
4544 changes of various kinds. As a consequence, a big documentation update was
4545 needed. This delayed the release for rather longer than seemed good, especially
4546 in the light of a couple of (minor) security issues. Therefore, the changes
4547 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
4548 release. So 4.44 and 4.50 are in effect two different branches that both start
4551 I have left the 4.50 change log unchanged; it contains all the changes since
4552 4.43. The change log for 4.44 is below; many of its items are identical to
4553 those for 4.50. This seems to be the most sensible way to preserve the
4554 historical information.
4560 1. Minor wording change to the doc/README.SIEVE file.
4562 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
4563 computation of the current number of files was incorrect.
4565 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4566 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4569 4. Give more explanation in the error message when the command for a transport
4570 filter fails to execute.
4572 5. There are several places where Exim runs a non-Exim command in a
4573 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4574 was being done only for the command run by the queryprogram router. It is
4575 now done for all such subprocesses. The other cases are: ${run, transport
4576 filters, and the commands run by the lmtp and pipe transports.
4578 6. Added CONFIGURE_GROUP build-time option.
4580 7. Some older OS have a limit of 256 on the maximum number of file
4581 descriptors. Exim was using setrlimit() to set 1000 as a large value
4582 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4583 systems. I've change it so that if it can't get 1000, it tries for 256.
4585 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4586 was an oversight, and furthermore, ever since the addition of extra
4587 controls (e.g. 4.43/32), the checks on when to allow different forms of
4588 "control" were broken. There should now be diagnostics for all cases when a
4589 control that does not make sense is encountered.
4591 9. Added the /retain_sender option to "control=submission".
4593 10. $recipients is now available in the predata ACL (oversight).
4595 11. Tidy the search cache before the fork to do a delivery from a message
4596 received from the command line. Otherwise the child will trigger a lookup
4597 failure and thereby defer the delivery if it tries to use (for example) a
4598 cached ldap connection that the parent has called unbind on.
4600 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4601 of $address_data from the recipient verification was clobbered by the
4602 sender verification.
4604 13. The value of address_data from a sender verification is now available in
4605 $sender_address_data in subsequent conditions in the ACL statement.
4607 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
4609 15. Added a new option "connect=<time>" to callout options, to set a different
4612 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4613 was its contents. (It was OK if the option was not defined at all.)
4615 17. A "Completed" log line is now written for messages that are removed from
4616 the spool by the -Mrm option.
4618 18. New variables $sender_verify_failure and $recipient_verify_failure contain
4619 information about exactly what failed.
4621 19. Added -dd to debug only the daemon process.
4623 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
4624 handles timeouts, both on the server side and network timeouts. Renamed the
4625 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
4627 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
4628 It is now set to "smtps".
4630 22. $host_address is now set to the target address during the checking of
4631 ignore_target_hosts.
4633 23. When checking ignore_target_hosts for an ipliteral router, no host name was
4634 being passed; this would have caused $sender_host_name to have been used if
4635 matching the list had actually called for a host name (not very likely,
4636 since this list is usually IP addresses). A host name is now passed as
4639 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4640 code that specifies a non-restarting handler (typically sigaction() in
4641 modern systems) in an attempt to fix a rare and obscure crash bug.
4643 25. Narrowed the window for a race in the daemon that could cause it to ignore
4644 SIGCHLD signals. This is not a major problem, because they are used only to
4645 wake it up if nothing else does.
4647 26. A malformed maildirsize file could cause Exim to calculate negative values
4648 for the mailbox size or file count. Odd effects could occur as a result.
4649 The maildirsize information is now recalculated if the size or filecount
4652 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4653 support for a long time. Removed HAVE_SYS_VFS_H.
4655 28. Installed the latest version of exipick from John Jetmore.
4657 29. In an address list, if the pattern was not a regular expression, an empty
4658 subject address (from a bounce message) matched only if the pattern was an
4659 empty string. Non-empty patterns were not even tested. This was the wrong
4660 because it is perfectly reasonable to use an empty address as part of a
4661 database query. An empty address is now tested by patterns that are
4662 lookups. However, all the other forms of pattern expect the subject to
4663 contain a local part and a domain, and therefore, for them, an empty
4664 address still always fails if the pattern is not itself empty.
4666 30. Exim went into a mad DNS loop when attempting to do a callout where the
4667 host was specified on an smtp transport, and looking it up yielded more
4668 than one IP address.
4670 31. Re-factored the code for checking spool and log partition space into a
4671 function that finds that data and another that does the check. The former
4672 is then used to implement four new variables: $spool_space, $log_space,
4673 $spool_inodes, and $log_inodes.
4675 32. The RFC2047 encoding function was originally intended for short strings
4676 such as real names; it was not keeping to the 75-character limit for
4677 encoded words that the RFC imposes. It now respects the limit, and
4678 generates multiple encoded words if necessary. To be on the safe side, I
4679 have increased the buffer size for the ${rfc2047: expansion operator from
4682 33. It is now permitted to omit both strings after an "if" condition; if the
4683 condition is true, the result is "true". As before, when the second string
4684 is omitted, a false condition yields an empty string. This makes it less
4685 cumbersome to write custom ACL and router conditions.
4687 34. Failure to deliver a bounce message always caused it to be frozen, even if
4688 there was an errors_to setting on the router. The errors_to setting is now
4691 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4692 canonical form (fully expanded) before being placed in
4693 $sender_host_address.
4695 36. The table in the code that translates DNS record types into text (T_A to
4696 "A" for instance) was missing entries for NS and CNAME. It is just possible
4697 that this could have caused confusion if both these types were looked up
4698 for the same domain, because the text type is used as part of Exim's
4699 per-process caching. But the chance of anyone hitting this buglet seems
4702 37. The dnsdb lookup has been extended in a number of ways.
4704 (1) There is a new type, "zns", which walks up the domain tree until it
4705 finds some nameserver records. It should be used with care.
4707 (2) There is a new type, "mxh", which is like "mx" except that it returns
4708 just the host names, not the priorities.
4710 (3) It is now possible to give a list of domains (or IP addresses) to be
4711 looked up. The behaviour when one of the lookups defers can be
4712 controlled by a keyword.
4714 (4) It is now possible to specify the separator character for use when
4715 multiple records are returned.
4717 38. The dnslists ACL condition has been extended: it is now possible to supply
4718 a list of IP addresses and/or domains to be looked up in a particular DNS
4721 39. Added log_selector=+queue_time_overall.
4723 40. When running the queue in the test harness, wait just a tad after forking a
4724 delivery process, to get repeatability of debugging output.
4726 41. Include certificate and key file names in error message when GnuTLS fails
4727 to set them up, because the GnuTLS error message doesn't include the name
4728 of the failing file when there is a problem reading it.
4730 42. Allow both -bf and -bF in the same test run.
4732 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
4734 44. The "Exiscan patch" is now merged into the mainline Exim source.
4736 45. Sometimes the final signoff response after QUIT could fail to get
4737 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4738 < 0 before doing a fflush(). This bug looks as though it goes back to the
4739 introduction of TLS in release 3.20, but "sometimes" must have been rare
4740 because the tests only now provoked it.
4742 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
4743 (this can affect the format of dates).
4745 47. exim_tidydb, when checking for the continued existence of a message for
4746 which it has found a message-specific retry record, was not finding
4747 messages that were in split spool directories. Consequently, it was
4748 deleting retry records that should have stayed in existence.
4750 48. Steve fixed some bugs in eximstats.
4752 49. The SPA authentication driver was not abandoning authentication and moving
4753 on to the next authenticator when an expansion was forced to fail,
4754 contradicting the general specification for all authenticators. Instead it
4755 was generating a temporary error. It now behaves as specified.
4757 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
4758 (the order specifies the preference for clients). The order is now AES256,
4759 AES128, 3DES, ARCFOUR128.
4761 51. Small patch to Sieve code - explicitly set From: when generating an
4764 52. Exim crashed if a remote delivery caused a very long error message to be
4765 recorded - for instance if somebody sent an entire SpamAssassin report back
4766 as a large number of 550 error lines. This bug was coincidentally fixed by
4767 increasing the size of one of Exim's internal buffers (big_buffer) that
4768 happened as part of the Exiscan merge. However, to be on the safe side, I
4769 have made the code more robust (and fixed the comments that describe what
4772 53. Now that there can be additional text after "Completed" in log lines (if
4773 the queue_time_overall log selector is set), a one-byte patch to exigrep
4774 was needed to allow it to recognize "Completed" as not the last thing in
4777 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
4778 patch that reportedly fixes this has been added. I am not expert enough to
4779 create a test for it. This is what the patch creator wrote:
4781 "I found a little strange behaviour of ldap code when working with
4782 Windows 2003 AD Domain, where users was placed in more than one
4783 Organization Units. When I tried to give exim partial DN, the exit code
4784 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
4785 But simultaneously result of request was absolutely normal ldap result,
4786 so I produce this patch..."
4788 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
4789 REFERENCE, so I have modified the code to exclude the patch when that macro
4792 55. Some experimental protocols are using DNS PTR records for new purposes. The
4793 keys for these records are domain names, not reversed IP addresses. The
4794 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4795 leaves it alone. Component reversal etc. now happens only for IP addresses.
4798 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
4800 57. Double the size of the debug message buffer (to 2048) so that more of very
4801 long debug lines gets shown.
4803 58. The exicyclog utility now does better if the number of log files to keep
4804 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
4806 59. Two changes related to the smtp_active_hostname option:
4808 (1) $smtp_active_hostname is now available as a variable.
4809 (2) The default for smtp_banner uses $smtp_active_hostname instead
4810 of $primary_hostname.
4812 60. The host_aton() function is supposed to be passed a string that is known
4813 to be a valid IP address. However, in the case of IPv6 addresses, it was
4814 not checking this. This is a hostage to fortune. Exim now panics and dies
4815 if the condition is not met. A case was found where this could be provoked
4816 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4817 components; fortuitously, this particular loophole had already been fixed
4818 by change 4.50/55 above.
4820 If there are any other similar loopholes, the new check in host_aton()
4821 itself should stop them being exploited. The report I received stated that
4822 data on the command line could provoke the exploit when Exim was running as
4823 exim, but did not say which command line option was involved. All I could
4824 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4825 running as the user.
4828 61. There was a buffer overflow vulnerability in the SPA authentication code
4829 (which came originally from the Samba project). I have added a test to the
4830 spa_base64_to_bits() function which I hope fixes it.
4833 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
4834 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
4836 63. The daemon start-up calls getloadavg() while still root for those OS that
4837 need the first call to be done as root, but it missed one case: when
4838 deliver_queue_load_max is set with deliver_drop_privilege. This is
4839 necessary for the benefit of the queue runner, because there is no re-exec
4840 when deliver_drop_privilege is set.
4842 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
4843 This has been fixed.
4845 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4846 was in use, was not putting the data itself into the right store pool;
4847 consequently, it could be overwritten for a subsequent message in the same
4848 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4851 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
4853 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
4854 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
4855 the function were treating the return as a boolean value, which happened to
4856 work because 0=false and not-0=true, but is not correct code.
4858 68. The host_aton() function was not handling scoped IPv6 addresses (those
4859 with, for example, "%eth0" on the end) correctly.
4861 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
4862 negated items (that is, ~something) in unsigned ints. Some compilers
4863 apparently mutter when there is no cast.
4865 70. If an address verification called from an ACL failed, and did not produce a
4866 user-specific message (i.e. there was only a "system" message), nothing was
4867 put in $acl_verify_message. In this situation, it now puts the system
4870 71. Change 4.23/11 added synchronization checking at the start of an SMTP
4871 session; change 4.31/43 added the unwanted input to the log line - except
4872 that it did not do this in the start of session case. It now does.
4874 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
4875 This is wrong and can cause the other end to generate a synchronization
4876 error if it is another Exim or anything else that does the synchronization
4877 check. A QUIT command is no longer sent after a timeout.
4879 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
4880 during host lookups.
4882 74. The defer_ok option of callout verification was not working if it was used
4883 when verifying addresses in header lines, that is, for this case:
4885 verify = header_sender/callout=defer_ok
4887 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
4888 those file descriptors could be used for SMTP connections. If anything
4889 wrote to stderr (the example that came up was "warn" in embedded Perl), it
4890 could be sent to the SMTP client, causing chaos. The daemon now opens
4891 stdin, stdout, and stderr to /dev/null when it puts itself into the
4894 76. Arrange for output from Perl's "warn" command to be written to Exim's main
4895 log by default. The user can override this with suitable Perl magic.
4897 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
4898 the log message when discard triggers, was not working for the DATA ACL or
4899 for the non-SMTP ACL.
4901 78. Error message wording change in sieve.c.
4903 79. If smtp_accept_max_per_host was set, the number of connections could be
4904 restricted to fewer than expected, because the daemon was trying to set up
4905 a new connection before checking whether the processes handling previous
4906 connections had finished. The check for completed processes is now done
4907 earlier. On busy systems, this bug wouldn't be noticed because something
4908 else would have woken the daemon, and it would have reaped the completed
4911 80. If a message was submitted locally by a user whose login name contained one
4912 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
4913 It caused a spool format error. I have fixed the spool reading code. A
4914 related problem was that the "from" clause in the Received: line became
4915 illegal because of the space(s). It is now covered by ${quote_local_part.
4917 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
4920 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
4922 83. Patch from Sieve maintainer to fix unterminated string problem in
4923 "vacation" handling.
4925 84. Some minor changes to the Linux configuration files to help with other
4926 OS variants using glibc.
4928 85. One more patch for Sieve to update vacation handling to latest spec.
4931 ----------------------------------------------------
4932 See the note above about the 4.44 and 4.50 releases.
4933 ----------------------------------------------------
4939 1. Change 4.43/35 introduced a bug that caused file counts to be
4940 incorrectly computed when quota_filecount was set in an appendfile
4943 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4944 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4947 3. Give more explanation in the error message when the command for a transport
4948 filter fails to execute.
4950 4. There are several places where Exim runs a non-Exim command in a
4951 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4952 was being done only for the command run by the queryprogram router. It is
4953 now done for all such subprocesses. The other cases are: ${run, transport
4954 filters, and the commands run by the lmtp and pipe transports.
4956 5. Some older OS have a limit of 256 on the maximum number of file
4957 descriptors. Exim was using setrlimit() to set 1000 as a large value
4958 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4959 systems. I've change it so that if it can't get 1000, it tries for 256.
4961 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4962 was an oversight, and furthermore, ever since the addition of extra
4963 controls (e.g. 4.43/32), the checks on when to allow different forms of
4964 "control" were broken. There should now be diagnostics for all cases when a
4965 control that does not make sense is encountered.
4967 7. $recipients is now available in the predata ACL (oversight).
4969 8. Tidy the search cache before the fork to do a delivery from a message
4970 received from the command line. Otherwise the child will trigger a lookup
4971 failure and thereby defer the delivery if it tries to use (for example) a
4972 cached ldap connection that the parent has called unbind on.
4974 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4975 of $address_data from the recipient verification was clobbered by the
4976 sender verification.
4978 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4979 was its contents. (It was OK if the option was not defined at all.)
4981 11. A "Completed" log line is now written for messages that are removed from
4982 the spool by the -Mrm option.
4984 12. $host_address is now set to the target address during the checking of
4985 ignore_target_hosts.
4987 13. When checking ignore_target_hosts for an ipliteral router, no host name was
4988 being passed; this would have caused $sender_host_name to have been used if
4989 matching the list had actually called for a host name (not very likely,
4990 since this list is usually IP addresses). A host name is now passed as
4993 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4994 code that specifies a non-restarting handler (typically sigaction() in
4995 modern systems) in an attempt to fix a rare and obscure crash bug.
4997 15. Narrowed the window for a race in the daemon that could cause it to ignore
4998 SIGCHLD signals. This is not a major problem, because they are used only to
4999 wake it up if nothing else does.
5001 16. A malformed maildirsize file could cause Exim to calculate negative values
5002 for the mailbox size or file count. Odd effects could occur as a result.
5003 The maildirsize information is now recalculated if the size or filecount
5006 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
5007 support for a long time. Removed HAVE_SYS_VFS_H.
5009 18. Updated exipick to current release from John Jetmore.
5011 19. Allow an empty sender to be matched against a lookup in an address list.
5012 Previously the only cases considered were a regular expression, or an
5015 20. Exim went into a mad DNS lookup loop when doing a callout where the
5016 host was specified on the transport, if the DNS lookup yielded more than
5019 21. The RFC2047 encoding function was originally intended for short strings
5020 such as real names; it was not keeping to the 75-character limit for
5021 encoded words that the RFC imposes. It now respects the limit, and
5022 generates multiple encoded words if necessary. To be on the safe side, I
5023 have increased the buffer size for the ${rfc2047: expansion operator from
5026 22. Failure to deliver a bounce message always caused it to be frozen, even if
5027 there was an errors_to setting on the router. The errors_to setting is now
5030 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
5031 canonical form (fully expanded) before being placed in
5032 $sender_host_address.
5034 24. Updated eximstats to version 1.33
5036 25. Include certificate and key file names in error message when GnuTLS fails
5037 to set them up, because the GnuTLS error message doesn't include the name
5038 of the failing file when there is a problem reading it.
5040 26. Expand error message when OpenSSL has problems setting up cert/key files.
5043 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
5044 (this can affect the format of dates).
5046 28. exim_tidydb, when checking for the continued existence of a message for
5047 which it has found a message-specific retry record, was not finding
5048 messages that were in split spool directories. Consequently, it was
5049 deleting retry records that should have stayed in existence.
5051 29. eximstats updated to version 1.35
5052 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
5053 1.35 - bugfix such that pie charts by volume are generated correctly
5055 30. The SPA authentication driver was not abandoning authentication and moving
5056 on to the next authenticator when an expansion was forced to fail,
5057 contradicting the general specification for all authenticators. Instead it
5058 was generating a temporary error. It now behaves as specified.
5060 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
5061 (the order specifies the preference for clients). The order is now AES256,
5062 AES128, 3DES, ARCFOUR128.
5064 31. Small patch to Sieve code - explicitly set From: when generating an
5067 32. Exim crashed if a remote delivery caused a very long error message to be
5068 recorded - for instance if somebody sent an entire SpamAssassin report back
5069 as a large number of 550 error lines. This bug was coincidentally fixed by
5070 increasing the size of one of Exim's internal buffers (big_buffer) that
5071 happened as part of the Exiscan merge. However, to be on the safe side, I
5072 have made the code more robust (and fixed the comments that describe what
5075 33. Some experimental protocols are using DNS PTR records for new purposes. The
5076 keys for these records are domain names, not reversed IP addresses. The
5077 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
5078 leaves it alone. Component reversal etc. now happens only for IP addresses.
5081 34. The host_aton() function is supposed to be passed a string that is known
5082 to be a valid IP address. However, in the case of IPv6 addresses, it was
5083 not checking this. This is a hostage to fortune. Exim now panics and dies
5084 if the condition is not met. A case was found where this could be provoked
5085 from a dnsdb PTR lookup with an IPv6 address that had more than 8
5086 components; fortuitously, this particular loophole had already been fixed
5087 by change 4.50/55 or 4.44/33 above.
5089 If there are any other similar loopholes, the new check in host_aton()
5090 itself should stop them being exploited. The report I received stated that
5091 data on the command line could provoke the exploit when Exim was running as
5092 exim, but did not say which command line option was involved. All I could
5093 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
5094 running as the user.
5097 35. There was a buffer overflow vulnerability in the SPA authentication code
5098 (which came originally from the Samba project). I have added a test to the
5099 spa_base64_to_bits() function which I hope fixes it.
5102 36. The daemon start-up calls getloadavg() while still root for those OS that
5103 need the first call to be done as root, but it missed one case: when
5104 deliver_queue_load_max is set with deliver_drop_privilege. This is
5105 necessary for the benefit of the queue runner, because there is no re-exec
5106 when deliver_drop_privilege is set.
5108 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
5109 was in use, was not putting the data itself into the right store pool;
5110 consequently, it could be overwritten for a subsequent message in the same
5111 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
5114 38. Sometimes the final signoff response after QUIT could fail to get
5115 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
5116 < 0 before doing a fflush(). This bug looks as though it goes back to the
5117 introduction of TLS in release 3.20, but "sometimes" must have been rare
5118 because the tests only now provoked it.
5124 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
5125 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
5126 Now it returns an integer. A number of calls were still expecting a T/F
5127 return. Fortuitously, in all cases, the tests worked in OK situations,
5128 which is the norm. However, things would have gone wrong on any write
5129 failures on the smtp file descriptor. This function is used when sending
5130 messages over SMTP and also when doing verify callouts.
5132 2. When Exim is called to do synchronous delivery of a locally submitted
5133 message (the -odf or -odi options), it no longer closes stderr before doing
5136 3. Implemented the mua_wrapper option.
5138 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
5140 5. Implemented the functions header_remove(), header_testname(),
5141 header_add_at_position(), and receive_remove_recipient(), and exported them
5144 6. If an ACL "warn" statement specified the addition of headers, Exim already
5145 inserted X-ACL-Warn: at the start if there was no header name. However, it
5146 was not making this test for the second and subsequent header lines if
5147 there were newlines in the string. This meant that an invalid header could
5148 be inserted if Exim was badly configured.
5150 7. Allow an ACL "warn" statement to add header lines at the start or after all
5151 the Received: headers, as well as at the end.
5153 8. Added the rcpt_4xx retry error code.
5155 9. Added postmaster_mailfrom=xxx to callout verification option.
5157 10. Added mailfrom=xxxx to the callout verification option, for verify=
5160 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
5161 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
5163 12. Inserted some casts to stop certain compilers warning when using pointer
5164 differences as field lengths or precisions in printf-type calls (mostly
5165 affecting debugging statements).
5167 13. Added optional readline() support for -be (dynamically loaded).
5169 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
5170 same clock tick as a message's arrival, so that its received time was the
5171 same as the "first fail" time on the retry record, and that message
5172 remained on the queue past the ultimate address timeout, every queue runner
5173 would try a delivery (because it was past the ultimate address timeout) but
5174 after another failure, the ultimate address timeout, which should have then
5175 bounced the address, did not kick in. This was a "< instead of <=" error;
5176 in most cases the first failure would have been in the next clock tick
5177 after the received time, and all would be well.
5179 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
5180 being recognized when the domain list was tested by the match_domain
5181 condition in an expansion string.
5183 16. Added the ${str2b64: operator.
5185 17. Exim was always calling setrlimit() to set a large limit for the number of
5186 processes, without checking whether the existing limit was already
5187 adequate. (It did check for the limit on file descriptors.) Furthermore,
5188 errors from getrlimit() and setrlimit() were being ignored. Now they are
5189 logged to the main and panic logs, but Exim does carry on, to try to do its
5190 job under whatever limits there are.
5192 18. Imported PCRE 5.0.
5194 19. Trivial typo in log message " temporarily refused connection" (the leading
5197 20. If the log selector return_path_on_delivery was set and an address was
5198 redirected to /dev/null, the delivery process crashed because it assumed
5199 that a return path would always be set for a "successful" delivery. In this
5200 case, the whole delivery is bypassed as an optimization, and therefore no
5203 21. Internal re-arrangement: the function for sending a challenge and reading
5204 a response while authentication was assuming a zero-terminated challenge
5205 string. It's now changed to take a pointer and a length, to allow for
5206 binary data in such strings.
5208 22. Added the cyrus_sasl authenticator (code supplied by MBM).
5210 23. Exim was not respecting finduser_retries when seeking the login of the
5211 uid under which it was called; it was always trying 10 times. (The default
5212 setting of finduser_retries is zero.) Also, it was sleeping after the final
5213 failure, which is pointless.
5215 24. Implemented tls_on_connect_ports.
5217 25. Implemented acl_smtp_predata.
5219 26. If the domain in control=submission is set empty, Exim assumes that the
5220 authenticated id is a complete email address when it generates From: or
5221 Sender: header lines.
5223 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
5224 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
5225 chown and chgrp in /bin and hostname in /usr/bin.
5227 28. Exim was keeping the "process log" file open after each use, just as it
5228 does for the main log. This opens the possibility of it remaining open for
5229 long periods when the USR1 signal hits a daemon. Occasional processlog
5230 errors were reported, that could have been caused by this. Anyway, it seems
5231 much more sensible not to leave this file open at all, so that is what now
5234 29. The long-running daemon process does not normally write to the log once it
5235 has entered its main loop, and it closes the log before doing so. This is
5236 so that log files can straightforwardly be renamed and moved. However,
5237 there are a couple of unusual error situations where the daemon does write
5238 log entries, and I had neglected to close the log afterwards.
5240 30. The text of an SMTP error response that was received during a remote
5241 delivery was being truncated at 512 bytes. This is too short for some of
5242 the long messages that one sometimes sees. I've increased the limit to
5245 31. It is now possible to make retry rules that apply only when a message has a
5246 specific sender, in particular, an empty sender.
5248 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
5249 it possible to be selective about when SMTP synchronization is enforced.
5251 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
5253 32. Implemented hosts_connection_nolog.
5255 33. Added an ACL for QUIT.
5257 34. Setting "delay_warning=" to disable warnings was not working; it gave a
5260 35. Added mailbox_size and mailbox_filecount to appendfile.
5262 36. Added control = no_multiline_responses to ACLs.
5264 37. There was a bug in the logic of the code that waits for the clock to tick
5265 in the case where the clock went backwards by a substantial amount such
5266 that the microsecond fraction of "now" was more than the microsecond
5267 fraction of "then" (but the whole seconds number was less).
5269 38. Added support for the libradius Radius client library this is found on
5270 FreeBSD (previously only the radiusclient library was supported).
5276 1. When certain lookups returned multiple values in the form name=value, the
5277 quoting of the values was not always being done properly. Specifically:
5278 (a) If the value started with a double quote, but contained no whitespace,
5280 (b) If the value contained whitespace other than a space character (i.e.
5281 tabs or newlines or carriage returns) it was not quoted.
5282 This fix has been applied to the mysql and pgsql lookups by writing a
5283 separate quoting function and calling it from the lookup code. The fix
5284 should probably also be applied to nisplus, ibase and oracle lookups, but
5285 since I cannot test any of those, I have not disturbed their existing code.
5287 2. A hit in the callout cache for a specific address caused a log line with no
5288 reason for rejecting RCPT. Now it says "Previous (cached) callout
5289 verification failure".
5291 3. There was an off-by-one bug in the queryprogram router. An over-long
5292 return line was truncated at 256 instead of 255 characters, thereby
5293 overflowing its buffer with the terminating zero. As well as fixing this, I
5294 have increased the buffer size to 1024 (and made a note to document this).
5296 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
5297 when Exim is waiting for an SMTP response from a remote server, Exim
5298 restarts its select() call on the socket, thereby resetting its timeout.
5299 This is not a problem when such interrupts are rare. Somebody set up a cron
5300 job to run exiwhat every 2 minutes, which is less than the normal select()
5301 timeout (5 or 10 minutes). This meant that the select() timeout never
5302 kicked in because it was always reset. I have fixed this by comparing the
5303 time when an interrupt arrives with the time at the start of the first call
5304 to select(). If more time than the timeout has elapsed, the interrupt is
5305 treated as a timeout.
5307 5. Some internal re-factoring in preparation for the addition of Sieve
5308 extensions (by MH). In particular, the "personal" test is moved to a
5309 separate function, and given an option for scanning Cc: and Bcc: (which is
5310 not set for Exim filters).
5312 6. When Exim created an email address using the login of the caller as the
5313 local part (e.g. when creating a From: or Sender: header line), it was not
5314 quoting the local part when it contained special characters such as @.
5316 7. Installed new OpenBSD configuration files.
5318 8. Reworded some messages for syntax errors in "and" and "or" conditions to
5319 try to make them clearer.
5321 9. Callout options, other than the timeout value, were being ignored when
5322 verifying sender addresses in header lines. For example, when using
5324 verify = header_sender/callout=no_cache
5326 the cache was (incorrectly) being used.
5328 10. Added a missing instance of ${EXE} to the exim_install script; this affects
5329 only the Cygwin environment.
5331 11. When return_path_on_delivery was set as a log selector, if different remote
5332 addresses in the same message used different return paths and parallel
5333 remote delivery occurred, the wrong values would sometimes be logged.
5334 (Whenever a remote delivery process finished, the return path value from
5335 the most recently started remote delivery process was logged.)
5337 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
5338 lines when AUTH and/or TLS are in use. This is the "received protocol"
5339 field. Exim used to use "asmtp" for authenticated SMTP, without any
5340 indication (in the protocol name) for TLS use. Now it follows the RFC and
5341 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
5342 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
5343 names appear in log lines as well as in Received: header lines.
5345 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
5346 extensions, and comparison tests, and to fix some bugs.
5348 14. Changes to the "personal" filter test:
5350 (1) The test was buggy in that it was just doing the equivalent of
5351 "contains" tests on header lines. For example, if a user's address was
5352 anne@some.where, the "personal" test would incorrectly be true for
5354 To: susanne@some.where
5356 This test is now done by extracting each address from the header in turn,
5357 and checking the entire address. Other tests that are part of "personal"
5358 are now done using regular expressions (for example, to check local parts
5359 of addresses in From: header lines).
5361 (2) The list of non-personal local parts in From: addresses has been
5362 extended to include "listserv", "majordomo", "*-request", and "owner-*",
5363 taken from the Sieve specification recommendations.
5365 (3) If the message contains any header line starting with "List-" it is
5366 treated as non-personal.
5368 (4) The test for "circular" in the Subject: header line has been removed
5369 because it now seems ill-conceived.
5371 15. Minor typos in src/EDITME comments corrected.
5373 16. Installed latest exipick from John Jetmore.
5375 17. If headers_add on a router specified a text string that was too long for
5376 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
5377 of string_sprintf() is now avoided.
5379 18. $message_body_size was not set (it was always zero) when running the DATA
5380 ACL and the local_scan() function.
5382 19. For the "mail" command in an Exim filter, no default was being set for
5383 the once_repeat time, causing a random time value to be used if "once" was
5384 specified. (If the value happened to be <= 0, no repeat happened.) The
5385 default is now 0s, meaning "never repeat". The "vacation" command was OK
5386 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
5387 (I found it when inspecting the code).
5389 20. There is now an overall timeout for performing a callout verification. It
5390 defaults to 4 times the callout timeout, which applies to individual SMTP
5391 commands during the callout. The overall timeout applies when there is more
5392 than one host that can be tried. The timeout is checked before trying the
5393 next host. This prevents very long delays if there are a large number of
5394 hosts and all are timing out (e.g. when the network connections are timing
5395 out). The value of the overall timeout can be changed by specifying an
5396 additional sub-option for "callout", called "maxwait". For example:
5398 verify = sender/callout=5s,maxwait=20s
5400 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
5401 to the end before writing, but this should make it even safer).
5403 22. Exim was forgetting that it had advertised PIPELINING for the second and
5404 subsequent messages on an SMTP connection. It was also not resetting its
5405 memory on STARTTLS and an internal HELO.
5407 23. When Exim logs an SMTP synchronization error within a session, it now
5408 records whether PIPELINING has been advertised or not.
5410 24. Added 3 instances of "(long int)" casts to time_t variables that were being
5411 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
5412 rather than long int.
5414 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
5416 26. Added the never_mail option to autoreply.
5422 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
5423 crash if the getsockname() call failed; this can happen if a connection is
5424 closed very soon after it is established. The problem was simply in the
5425 order in which certain operations were done, causing Exim to try to write
5426 to the SMTP stream before it had set up the file descriptor. The bug has
5427 been fixed by making things happen in the correct order.
5433 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
5434 before the connection was closed, thus losing the rejection response.
5436 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
5437 some early Solaris releases, but causes trouble in current releases where
5438 socklen_t is defined.
5440 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
5443 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
5446 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
5447 that releasing the top of it at the end releases what was used for sub-
5448 expansions (unless the block got too big). However, discard this block if
5449 the first thing is a variable or header, so that we can use its block when
5450 it is dynamic (useful for very large $message_headers, for example).
5452 6. Lookups now cache *every* query, not just the most recent. A new, separate
5453 store pool is used for this. It can be recovered when all lookup caches are
5454 flushed. Lookups now release memory at the end of their result strings.
5455 This has involved some general refactoring of the lookup sources.
5457 7. Some code has been added to the store_xxx() functions to reduce the amount
5458 of flapping under certain conditions.
5460 8. log_incoming_interface used to affect only the <= reception log lines. Now
5461 it causes the local interface and port to be added to several more SMTP log
5462 lines, for example "SMTP connection from", and rejection lines.
5464 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
5466 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
5468 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
5469 could be overwritten at the end of the current message (or the start of a
5470 new message if it was set in a HELO ACL). The value is now preserved for
5471 the duration of the SMTP connection.
5473 12. If a transport had a headers_rewrite setting, and a matching header line
5474 contained an unqualified address, that address was qualified, even if it
5475 did not match any rewriting rules. The underlying bug was that the values
5476 of the flags that permit the existence of unqualified sender and recipient
5477 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
5478 non-local messages, and by -bnq for local messages) were not being
5479 preserved with the message after it was received.
5481 13. When Exim was logging an SMTP synchronization error, it could sometimes log
5482 "next input=" as part of the text comprising the host identity instead of
5483 the correct text. The code was using the same buffer for two different
5484 strings. However, depending on which order the printing function evaluated
5485 its arguments, the bug did not always show up. Under Linux, for example, my
5486 test suite worked just fine.
5488 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
5489 doesn't work with some older versions of Perl. It has been changed to "my",
5490 which in any case is probably the better facility to use.
5492 15. A really picky compiler found some instances of statements for creating
5493 error messages that either had too many or two few arguments for the format
5496 16. The size of the buffer for calls to the DNS resolver has been increased
5497 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
5498 for addresses that have a lot of PTR records. This alleviates a problem; it
5499 does not fully solve it.
5501 17. A dnsdb lookup for PTR records that receives more data than will fit in the
5502 buffer now truncates the list and logs the incident, which is the same
5503 action as happens when Exim is looking up a host name and its aliases.
5504 Previously in this situation something unpredictable would happen;
5505 sometimes it was "internal error: store_reset failed".
5507 18. If a server dropped the connection unexpectedly when an Exim client was
5508 using GnuTLS and trying to read a response, the client delivery process
5509 crashed while trying to generate an error log message.
5511 19. If a "warn" verb in an ACL added multiple headers to a message in a single
5512 string, for example:
5514 warn message = H1: something\nH2: something
5516 the text was added as a single header line from Exim's point of view
5517 though it ended up OK in the delivered message. However, searching for the
5518 second and subsequent header lines using $h_h2: did not work. This has been
5519 fixed. Similarly, if a system filter added multiple headers in this way,
5520 the routers could not see them.
5522 20. Expanded the error message when iplsearch is called with an invalid key to
5523 suggest using net-iplsearch in a host list.
5525 21. When running tests using -bh, any delays imposed by "delay" modifiers in
5526 ACLs are no longer actually imposed (and a message to that effect is
5529 22. If a "gecos" field in a passwd entry contained escaped characters, in
5530 particular, if it contained a \" sequence, Exim got it wrong when building
5531 a From: or a Sender: header from that name. A second bug also caused
5532 incorrect handling when an unquoted " was present following a character
5533 that needed quoting.
5535 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
5536 was not being matched caselessly.
5538 24. Arranged for all hyphens in the exim.8 source to be escaped with
5541 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
5542 properly. Recipient callout cache records were still being keyed to include
5543 the sender, even when use_sender was set false. This led to far more
5544 callouts that were necessary. The sender is no longer included in the key
5545 when use_sender is false.
5547 26. Added "control = submission" modifier to ACLs.
5549 27. Added the ${base62d: operator to decode base 62 numbers.
5551 28. dnsdb lookups can now access SRV records.
5553 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
5554 the configuration file.
5556 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
5557 (-v) mode. This makes the output for a verbose queue run more intelligible.
5559 31. Added a use_postmaster feature to recipient callouts.
5561 32. Added the $body_zerocount variable, containing the number of binary zero
5562 bytes in the message body.
5564 33. The time of last modification of the "new" subdirectory is now used as the
5565 "mailbox time last read" when there is a quota error for a maildir
5568 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
5570 35. Added +ignore_unknown as a special item in host lists.
5572 36. Code for decoding IPv6 addresses in host lists is now included, even if
5573 IPv6 support is not being compiled. This fixes a bug in which an IPv6
5574 address was recognized as an IP address, but was then not correctly decoded
5575 into binary, causing unexpected and incorrect effects when compared with
5582 1. Very minor rewording of debugging text in manualroute to say "list of
5583 hosts" instead of "hostlist".
5585 2. If verify=header_syntax was set, and a header line with an unqualified
5586 address (no domain) and a large number of spaces between the end of the
5587 name and the colon was received, the reception process suffered a buffer
5588 overflow, and (when I tested it) crashed. This was caused by some obsolete
5589 code that should have been removed. The fix is to remove it!
5591 3. When running in the test harness, delay a bit after writing a bounce
5592 message to get a bit more predictability in the log output.
5594 4. Added a call to search_tidyup() just before forking a reception process. In
5595 theory, someone could use a lookup in the expansion of smtp_accept_max_
5596 per_host which, without the tidyup, could leave open a database connection.
5598 5. Added the variables $recipient_data and $sender_data which get set from a
5599 lookup success in an ACL "recipients" or "senders" condition, or a router
5600 "senders" option, similar to $domain_data and $local_part_data.
5602 6. Moved the writing of debug_print from before to after the "senders" test
5605 7. Change 4.31/66 (moving the time when the Received: is generated) caused
5606 problems for message scanning, either using a data ACL, or using
5607 local_scan() because the Received: header was not generated till after they
5608 were called (in order to set the time as the time of reception completion).
5609 I have revised the way this works. The header is now generated after the
5610 body is received, but before the ACL or local_scan() are called. After they
5611 are run, the timestamp in the header is updated.
5617 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
5618 before starting a queue runner without re-exec. This happened only when
5619 deliver_drop_privilege was set or when the Exim user was set to root. The
5620 effect of the bug was that timeouts during subsequent deliveries caused
5621 crashes instead of being properly handled. The handler is now left at its
5622 default (and expected) setting.
5624 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
5625 message, again when deliver_drop_privilege is set or Exim is run as root.
5626 The bug described in (1) was not present in this case, but the tidying up
5627 of the other signals was missing. I have made the two cases consistent.
5629 3. The ignore_target_hosts setting on a manualroute router was being ignored
5630 for hosts that were looked up using the /MX notation.
5632 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
5635 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
5636 operated on the sender address. After changing the $sender_address to <>
5637 for the sender address verify, Exim was re-instated it as the original
5638 (before rewriting) address, but remembering that it had rewritten it, so it
5639 wasn't rewriting it again. This bug also had the effect of breaking the
5640 sender address verification caching when the sender address was rewritten.
5642 6. The ignore_target_hosts option was being ignored by the ipliteral router.
5643 This has been changed so that if the ip literal address matches
5644 ignore_target_hosts, the router declines.
5646 7. Added expansion conditions match_domain, match_address, and match_local_
5647 part (NOT match_host).
5649 8. The placeholder for the Received: header didn't have a length field set.
5651 9. Added code to Exim itself and to exim_lock to test for a specific race
5652 condition that could lead to file corruption when using MBX delivery. The
5653 issue is with the lockfile that is created in /tmp. If this file is removed
5654 after a process has opened it but before that process has acquired a lock,
5655 there is the potential for a second process to recreate the file and also
5656 acquire a lock. This could lead to two Exim processes writing to the file
5657 at the same time. The added code performs the same test as UW imapd; it
5658 checks after acquiring the lock that its file descriptor still refers to
5659 the same named file.
5661 10. The buffer for building added header lines was of fixed size, 8192 bytes.
5662 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
5665 11. Added the smtp_active_hostname option. If used, this will typically be made
5666 to depend on the incoming interface address. Because $interface_address is
5667 not set up until the daemon has forked a reception process, error responses
5668 that can happen earlier (such as "too many connections") no longer contain
5671 12. If an expansion in a condition on a "warn" statement fails because a lookup
5672 defers, the "warn" statement is abandoned, and the next ACL statement is
5673 processed. Previously this caused the whole ACL to be aborted.
5675 13. Added the iplsearch lookup type.
5677 14. Added ident_timeout as a log selector.
5679 15. Added tls_certificate_verified as a log selector.
5681 16. Added a global option tls_require_ciphers (compare the smtp transport
5682 option of the same name). This controls incoming TLS connections.
5684 17. I finally figured out how to make tls_require_ciphers do a similar thing
5685 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
5686 before starting the TLS session.
5688 18. Tabs are now shown as \t in -bP output.
5690 19. If the log selector return_path_on_delivery was set, Exim crashed when
5691 bouncing a message because it had too many Received: header lines.
5693 20. If two routers both had headers_remove settings, and the first one included
5694 a superfluous trailing colon, the final name in the first list and the
5695 first name in the second list were incorrectly joined into one item (with a
5696 colon in the middle).
5702 1. Added -C and -D options to the exinext utility, mainly to make it easier
5703 to include in the automated testing, but these could be helpful when
5704 multiple configurations are in use.
5706 2. The exinext utility was not formatting the output nicely when there was
5707 an alternate port involved in the retry record key, nor when there was a
5708 message id as well (for retries that were specific to a specific message
5709 and a specific host). It was also confused by IPv6 addresses, because of
5710 the additional colons they contain. I have fixed the IPv4 problem, and
5711 patched it up to do a reasonable job for IPv6.
5713 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
5714 delivery, the log line now contains "pipelined" if PIPELINING was used.
5716 4. An SMTP transport process used to panic and die if the bind() call to set
5717 an explicit outgoing interface failed. This has been changed; it is now
5718 treated in the same way as a connect() failure.
5720 5. A reference to $sender_host_name in the part of a conditional expansion
5721 that was being skipped was still causing a DNS lookup. This no longer
5724 6. The def: expansion condition was not recognizing references to header lines
5725 that used bh_ and bheader_.
5727 7. Added the _cache feature to named lists.
5729 8. The code for checking quota_filecount in the appendfile transport was
5730 allowing one more file than it should have been.
5732 9. For compatibility with Sendmail, the command line option
5740 and sets the incoming protocol and host name (for trusted callers). The
5741 host name and its colon can be omitted when only the protocol is to be set.
5742 Note the Exim already has two private options, -pd and -ps, that refer to
5743 embedded Perl. It is therefore impossible to set a protocol value of "d" or
5744 "s", but I don't think that's a major issue.
5746 10. A number of refactoring changes to the code, none of which should affect
5749 (a) The number of logging options was getting close to filling up the
5750 32-bit word that was used as a bit map. I have split them into two classes:
5751 those that are passed in the argument to log_write(), and those that are
5752 only ever tested independently outside of that function. These are now in
5753 separate 32-bit words, so there is plenty of room for expansion again.
5754 There is no change in the user interface or the logging behaviour.
5756 (b) When building, for example, log lines, the code previously used a
5757 macro that called string_cat() twice, in order to add two strings. This is
5758 not really sufficiently general. Furthermore, there was one instance where
5759 it was actually wrong because one of the argument was used twice, and in
5760 one call a function was used. (As it happened, calling the function twice
5761 did not affect the overall behaviour.) The macro has been replaced by a
5762 function that can join an arbitrary number of extra strings onto a growing
5765 (c) The code for expansion conditions now uses a table and a binary chop
5766 instead of a serial search (which was left over from when there were very
5767 few conditions). Also, it now recognizes conditions like "pam" even when
5768 the relevant support is not compiled in: a suitably worded error message is
5769 given if an attempt is made to use such a condition.
5771 11. Added ${time_interval:xxxxx}.
5773 12. A bug was causing one of the ddress fields not to be passed back correctly
5774 from remote delivery subprocesses. The field in question was not being
5775 subsequently used, so this caused to problems in practice.
5777 13. Added new log selectors queue_time and deliver_time.
5779 14. Might have fixed a bug in maildirsizefile handling that threw up
5780 "unexpected character" debug warnings, and recalculated the data
5781 unnecessarily. In any case, I expanded the warning message to give more
5784 15. Added the message "Restricted characters in address" to the statements in
5785 the default ACL that block characters like @ and % in local parts.
5787 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
5788 Three changes have been made:
5790 (a) There was a serious bug; a negative response to MAIL caused the whole
5791 recipient domain to be cached as invalid, thereby blocking all messages
5792 to all local parts at the same domain, from all senders. This bug has
5793 been fixed. The domain is no longer cached after a negative response to
5794 MAIL if the sender used is not empty.
5796 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
5799 (c) A new callout option, "use_sender" has been added for people who want
5800 the modified behaviour.
5806 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
5809 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
5810 indeed breaks things for older releases.
5812 3. Added additional logging to the case where there is a problem reading data
5813 from a filter that is running in a subprocess using a pipe, in order to
5814 try to track down a specific problem.
5816 4. Testing facility fudge: when running in the test harness and attempting
5817 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
5818 getting "No route to host". Convert this to a timeout.
5820 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
5823 6. Some OS don't have socklen_t but use size_t instead. This affects the
5824 fifth argument of getsockopt() amongst other things. This is now
5825 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
5826 can be set for individual OS. I have set it for SunOS5, OSF1, and
5827 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
5828 some earlier ones do not.
5830 7. Change 4.30/15 was not doing the test caselessly.
5832 8. The standard form for an IPv6 address literal was being rejected by address
5833 parsing in, for example, MAIL and RCPT commands. An example of this kind of
5834 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5835 this, as well as the form without the "IPv6" on the front (but only when
5836 address literals are enabled, of course).
5838 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
5840 10. Exim crashed if a message with an empty sender address specified by -f
5841 encountered a router with an errors_to setting. This could be provoked only
5842 by a command such as
5846 where an empty string was supplied; "<>" did not hit this bug.
5848 11. Installed PCRE release 4.5.
5850 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
5851 remained set. It is now erased.
5853 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
5854 times from message ids (which are base 36 rather than the normal 62).
5856 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
5857 were being counted as actual protocol errors, and logged if the log
5858 selector +smtp_protocol_error was set. One cannot be perfect in this test,
5859 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
5860 and DATA following a set of rejected RCPTs do not count as protocol errors.
5861 In other words, Exim assumes they were pipelined, though this may not
5862 actually be the case. Of course, in all cases the client gets an
5863 appropriate error code.
5865 15. If a lookup fails in an ACL condition, a message about the failure may
5866 be available; it is used if testing the ACL cannot continue, because most
5867 such messages specify what the cause of the deferral is. However, some
5868 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
5869 that caused an old message to be retained and used if a later statement
5870 caused a defer, replacing the real cause of the deferral.
5872 16. If an IP address had so many PTR records that the DNS lookup buffer
5873 was not large enough to hold them, Exim could crash while trying to process
5874 the truncated data. It now detects and logs this case.
5876 17. Further to 4.21/58, another change has been made: if (and only if) the
5877 first line of a message (the first header line) ends with CRLF, a bare LF
5878 in a subsequent header line has a space inserted after it, so as not to
5879 terminate the header.
5881 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
5882 unnecessarily, used atoi() instead of strtol(), and didn't check the
5883 termination when getting file sizes from file names by regex.
5885 19. Completely re-implemented the support for maildirsize files, in the light
5886 of a number of problems with the previous contributed implementation
5887 (4.30/29). In particular:
5889 . If the quota is zero, the maildirsize file is maintained, but no quota is
5892 . If the maildir directory does not exist, it is created before any attempt
5893 to write a maildirsize file.
5895 . The quota value in the file is just a cache; if the quota is changed in
5896 the transport, the new value overrides.
5898 . A regular expression is available for excluding directories from the
5901 20. The autoreply transport checks the characters in options that define the
5902 message's headers; it allows continued headers, but it was checking with
5903 isspace() after an embedded newline instead of explicitly looking for a
5906 21. If all the "regular" hosts to which an address was routed had passed their
5907 expiry times, and had not reached their retry times, the address was
5908 bounced, even if fallback hosts were defined. Now Exim should go on to try
5911 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
5912 equivalent code in the SMTP transport. Some hosts send humungous responses
5913 to HELO/EHLO, more than 1024 it seems.
5915 23. Refactoring: code in filter.c used (void *) for "any old type" but this
5916 gives compiler warnings in some environments. I've now done it "properly",
5919 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
5920 (because of problems with the built-in one) was declared to return uschar *
5921 instead of char *, causing compiler failure.
5923 25. Fixed a file descriptor leak when processing alias/forward files.
5925 26. Fixed a minor format string issue in dbfn.c.
5927 27. Typo in exim.c: ("dmbnz" for "dbmnz").
5929 28. If a filter file refered to $h_xxx or $message_headers, and the headers
5930 contained RFC 2047 "words", Exim's memory could, under certain conditions,
5933 29. When a sender address is verified, it is cached, to save repeating the test
5934 when there is more than one recipient in a message. However, when the
5935 verification involves a callout, it is possible for different callout
5936 options to be set for different recipients. It is too complicated to keep
5937 track of this in the cache, so now Exim always runs a verification when a
5938 callout is required, relying on the callout cache for the optimization.
5939 The overhead is duplication of the address routing, but this should not be
5942 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
5943 to be verified with callout=postmaster, and the main callout worked but the
5944 postmaster check failed, the verification correctly failed. However, if a
5945 subsequent RCPT command asked for sender verification *without* the
5946 postmaster check, incorrect caching caused this verification also to fail,
5949 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
5950 it was not caching the DNS options (qualify_single, search_parents) that
5951 were used when the lookup failed. A subsequent lookup with different
5952 options therefore always gave the same answer, though there were cases
5953 where it should not have. (Example: a "domains = !$mx_any" option on a
5954 dnslookup router: the "domains" option is always processed without any
5955 widening, but the router might have qualify_single set.) Now Exim uses the
5956 cached value only when the same options are set.
5958 32. Added John Jetmore's "exipick" utility to the distribution.
5960 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
5961 than a timeout (e.g. a certificate is required, and is not provided), an
5962 Exim server now closes the connection immediately. Previously it waited for
5963 the client to close - but if the client is SSL, it seems that they each
5964 wait for each other, leading to a delay before one of them times out.
5966 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
5967 maintained 0.8.x compatibility because I don't think many are using it, and
5968 it is clearly obsolete.
5970 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
5973 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
5974 client certificate was expired. A simple patch fixes this, though I don't
5975 understand the full logic of why the verify callback is called multiple
5978 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
5979 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
5980 which causes problems with some clients (such as the Certicom SSL Plus
5981 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
5982 disables the coutermeasure allowing Eudora to connect."
5984 38. Exim was not checking that a write() to a log file succeeded. This could
5985 lead to Bad Things if a log got too big, in particular if it hit a file
5986 size limit. Exim now panics and dies if it cannot write to a log file, just
5987 as it does if it cannot open a log file.
5989 39. Modified OS/Makefile-Linux so that it now contains
5991 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
5993 The two -D definitions ensure that Exim is compiled with large file
5994 support, which makes it possible to handle log files that are bigger than
5997 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
5998 instance) a domain was checked against a named list that involved a lookup,
5999 causing $domain_data to be set, then another domain was checked against the
6000 same list, then the first domain was re-checked, the value of $domain_data
6001 after the final check could be wrong. In particular, if the second check
6002 failed, it could be set empty. This bug probably also applied to
6005 41. The strip_trailing_dot option was not being applied to the address given
6006 with the -f command-line option.
6008 42. The code for reading a message's header from the spool was incrementing
6009 $received_count, but never initializing it. This meant that the value was
6010 incorrect (doubled) while delivering a message in the same process in which
6011 it was received. In the most common configuration of Exim, this never
6012 happens - a fresh exec is done - but it can happen when
6013 deliver_drop_privilege is set.
6015 43. When Exim logs an SMTP synchronization error - client data sent too soon -
6016 it now includes up to 150 characters of the unexpected data in the log
6019 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
6020 and building data strings. The size of both of these buffers was 10 000
6021 bytes - far larger than anybody would *ever* want, thought I. Needless to
6022 say, somebody hit the limit. I have increased the maximum line length to
6023 20 000 and the maximum data length of concatenated lines to 100 000. I have
6024 also fixed two bugs, because there was no checking on these buffers. Tsk,
6025 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
6026 buffer is too small.
6028 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
6029 lsearch lookups. Now it does.
6031 46. When parsing a route_list item in a manualroute router, a fixed-length
6032 buffer was used for the list of hosts. I made this 1024 bytes long,
6033 thinking that nobody would ever have a list of hosts that long. Wrong.
6034 Somebody had a whole pile of complicated expansion conditions, and the
6035 string was silently truncated, leading to an expansion error. It turns out
6036 that it is easier to change to an unlimited length (owing to other changes
6037 that have happened since this code was originally written) than to build
6038 structure for giving a limitation error. The length of the item that
6039 expands into the list of hosts is now unlimited.
6041 47. The lsearch lookup could not handle data where the length of text line was
6042 more than 4095 characters. Such lines were truncated, leading to shortened
6043 data being returned. It should now handle lines of any length.
6045 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
6046 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
6049 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
6050 replacing strings such as BIN_DIRECTORY in a source file: the replacement
6051 no longer happens in comment lines. A list of replacements is now placed
6052 at the head of all of the source files, except those whose only change is
6053 to replace PERL_COMMAND in the very first #! line.
6055 50. Replaced the slow insertion sort in queue.c, for sorting the list of
6056 messages on the queue, with a bottom-up merge sort, using code contributed
6057 by Michael Haardt. This should make operations like -bp somewhat faster on
6058 large queues. It won't affect queue runners, except when queue_run_in_order
6061 51. Installed eximstats 1.31 in the distribution.
6063 52. Added support for SRV lookups to the dnslookup router.
6065 53. If an ACL referred to $message_body or $message_body_end, the value was not
6066 reset for any messages that followed in the same SMTP session.
6068 54. The store-handling optimization for building very long strings was not
6069 differentiating between the different store pools. I don't think this
6070 actually made any difference in practice, but I've tidied it.
6072 55. While running the routers to verify a sender address, $sender_address
6073 was still set to the sender address. This is wrong, because when routing to
6074 send a bounce to the sender, it would be empty. Therefore, I have changed
6075 it so that, while verifying a sender address, $sender_address is set to <>.
6076 (There is no change to what happens when verifying a recipient address.)
6078 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
6079 target A or AAAA records (if not already returned) without resetting the
6080 qualify_single or search_parents options of the DNS resolver. These are
6081 inappropriate in this case because the targets of MX and SRV records must
6082 be FQDNs. A broken DNS record could cause trouble if it happened to have a
6083 target that, when qualified, matched something in the local domain. These
6084 two options are now turned off when doing these lookups.
6086 57. It seems that at least some releases of Reiserfs (which does not have the
6087 concept of a fixed number of inodes) returns zero and not -1 for the
6088 number of available inodes. This interacted badly with check_spool_inodes,
6089 which assumed that -1 was the "no such thing" setting. What I have done is
6090 to check that the total number of inodes is greater than zero before doing
6091 the test of how many are available.
6093 58. When a "warn" ACL statement has a log_message modifier, the message is
6094 remembered, and not repeated. This is to avoid a lot of repetition when a
6095 message has many recipients that cause the same warning to be written.
6096 However, Exim was preserving the list of already written lines for an
6097 entire SMTP session, which doesn't seem right. The memory is now reset if a
6098 new message is started.
6100 59. The "rewrite" debugging flag was not showing the result of rewriting in the
6101 debugging output unless log_rewrite was also set.
6103 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
6104 of (int)(handle) when we know that handle contains (void *)(-1).
6106 61. The Exim daemon panic-logs an error return when it closes the incoming
6107 connection. However "connection reset by peer" seems to be common, and
6108 isn't really an error worthy of noting specially, so that particular error
6111 62. When Exim is trying to find all the local interfaces, it used to panic and
6112 die if the ioctl to get the interface flags failed. However, it seems that
6113 on at least one OS (Solaris 9) it is possible to have an interface that is
6114 included in the list of interfaces, but for which you get a failure error
6115 for this call. This happens when the interface is not "plumbed" into a
6116 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
6117 failure of the "get flags" call assumes that the interface is down.
6119 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
6120 makes life easier for people who are doing arithmetic on fields extracted
6121 from dates, where you often get leading zeros that should not be
6122 interpreted as octal.
6124 64. Added qualify_domain to the redirect router, to override the global
6127 65. If a pathologically long header line contained very many addresses (the
6128 report of this problem mentioned 10 000) and each of them was rewritten,
6129 Exim could use up a very large amount of memory. (It kept on making new
6130 copies of the header line as it rewrote, and never released the old ones.)
6131 At the expense of a bit more processing, the header rewriting function has
6132 been changed so that it no longer eats memory in this way.
6134 66. The generation of the Received: header has been moved from the time that a
6135 message starts to be received, to the time that it finishes. The timestamp
6136 in the Received: header should now be very close to that of the <= log
6137 line. There are two side-effects of this change:
6139 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
6140 logged header lines no longer include the local Received: line, because
6141 it has not yet been created. The same applies to a copy of the message
6142 that is returned to a non-SMTP sender when a message is rejected.
6144 (b) When a filter file is tested using -bf, no additional Received: header
6145 is added to the test message. After some thought, I decided that this
6148 This change does not affect the value of $received_for. It is still set
6149 after address rewriting, but before local_scan() is called.
6151 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
6153 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
6154 gave an unhelpful panic error message, and a defer error. I have managed to
6155 change this behaviour so that it now rejects any supplied certificate,
6156 which seems right, as the list of acceptable certificates is empty.
6158 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
6159 gave an unhelpful defer error. I have not managed to make this reject any
6160 supplied certificates, but the error message it gives is "no certificate
6161 supplied", which is not helpful.
6163 70. exigrep's output now also includes lines that are not associated with any
6164 message, but which match the given pattern. Implemented by a patch from
6165 Martin Sluka, which also tidied up the Perl a bit.
6167 71. Recipient callout verification, like sender verification, was using <> in
6168 the MAIL FROM command. This isn't really the right thing, since the actual
6169 sender may affect whether the remote host accepts the recipient or not. I
6170 have changed it to use the actual sender in the callout; this means that
6171 the cache record is now keyed on a recipient/sender pair, not just the
6172 recipient address. There doesn't seem to be a real danger of callout loops,
6173 since a callout by the remote host to check the sender would use <>.
6174 [SEE ABOVE: changed after hitting problems.]
6176 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
6177 temporary errors. However, in the case of such a code being given after
6178 the end of a data transmission (i.e. after ".") Exim was failing to write
6179 a retry record for the message. (Yes, there was some broken host that was
6180 actually sending 8xx at this point.)
6182 73. An unknown lookup type in a host list could cause Exim to panic-die when
6183 the list was checked. (An example that provoked this was putting <; in the
6184 middle of a list instead of at the start.) If this happened during a DATA
6185 ACL check, a -D file could be left lying around. This kind of configuration
6186 error no longer causes Exim to die; instead it causes a defer error. The
6187 incident is still logged to the main and panic logs.
6189 74. Buglet left over from Exim 3 conversion. The message "too many messages
6190 in one connection" was written to the rejectlog but not the mainlog, except
6191 when address rewriting (yes!) was being logged.
6193 75. Added write_rejectlog option.
6195 76. When a system filter was run not as root (that is, when system_filter_user
6196 was set), the values of the $n variables were not being returned to the
6197 main process; thus, they were not subsequently available in the $sn
6200 77. Added +return_path_on_delivery log selector.
6202 78. A connection timeout was being treated differently from recipients deferred
6203 when testing hosts_max_try with a message that was older than the host's
6204 retry timeout. (The host should not be counted, thus allowing all hosts to
6205 be tried at least once before bouncing.) This may have been the cause of an
6206 occasionally reported bug whereby a message would remain on the queue
6207 longer than the retry timeout, but would be bounced if a delivery was
6208 forced. I say "may" because I never totally pinned down the problem;
6209 setting up timeout/retry tests is difficult. See also the next item.
6211 79. The ultimate address timeout was not being applied to errors that involved
6212 a combination of host plus message (for example, a timeout on a MAIL
6213 command). When an address resolved to a number of possible hosts, and they
6214 were not all tried for each delivery (e.g. because of hosts_max_try), a
6215 message could remain on the queue longer than the retry timeout.
6217 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
6220 81. Fixed an obscure SMTP outgoing bug which required at least the following
6221 conditions: (a) there was another message waiting for the same server;
6222 (b) the server returned 5xx to all RCPT commands in the first message so
6223 that the message was not completed; (c) the server dropped the connection
6224 or gave a negative response to the RSET that Exim sends to abort the
6225 transaction. The observed case was a dropped connection after DATA that had
6226 been sent in pipelining mode. That is, the server had advertised PIPELINING
6227 but was not implementing it correctly. The effect of the bug was incorrect
6228 behaviour, such as trying another host, and this could lead to a crash.
6234 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
6235 and daemon.c were passed as pointers to ints; they should have been
6236 pointers to socklen_t variables (which are typically unsigned ints).
6238 2. Some signed/unsigned type warnings in the os.c file for Linux have been
6241 3. Fixed a really odd bug that affected only the testing scheme; patching a
6242 certain fixed string in the binary changed the value of another string that
6243 happened to be identical to the end of the original first string.
6245 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
6246 name", it returns that address as the IP address. On some operating
6247 systems (e.g. Solaris), it also passes back the IP address string as the
6248 "host name". However, on others (e.g. Linux), it passes back an empty
6249 string. Exim wasn't checking for this, and was changing the host name to an
6250 empty string, assuming it had been canonicalized.
6252 5. Although rare, it is permitted to have more than one PTR record for a given
6253 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
6254 all the names associated with an address, because they do in Solaris.
6255 However, it seems that they do not in Linux for data that comes from the
6256 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
6257 I found this out when I moved to a new Linux workstation and tried to run
6258 the Exim test suite.
6260 To get round this problem I have changed the code so that it now does its
6261 own call to the DNS to look up PTR records when searching for a host name.
6262 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
6263 addresses that are only in /etc/hosts are still found.
6265 This behaviour is, however, controlled by an option called host_lookup_
6266 order, which defaults to "bydns:byaddr". If people want to use the other
6267 order, or indeed, just use one or the other means of lookup, they can
6268 specify it in this variable.
6270 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
6271 some operating systems, this comes back from gethostbyaddr() as an empty
6272 string, and this is what Exim used to test for. However, it seems that in
6273 other systems, "." is yielded. Exim now tests for this case too.
6275 7. The values of check_spool_space and check_log_space are now held internally
6276 as a number of kilobytes instead of an absolute number of bytes. If a
6277 numbers is specified without 'K' or 'M', it is rounded up to the nearest
6278 kilobyte. This means that much larger values can be stored.
6280 8. Exim monitor: an attempt to get the action menu when not actually pointing
6281 at a message produces an empty menu entitled "No message selected". This
6282 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
6283 no entries in it ("Shell widget menu has zero width and/or height"). So I
6284 have added a single, blank menu entry in this case.
6286 9. Added ${quote_local_part.
6288 10. MIME decoding is now applied to the contents of Subject: header lines when
6291 11. Now that a reference to $sender_host_address automatically causes a reverse
6292 lookup to occur if necessary (4.13/18), there is no need to arrange for a
6293 host lookup before query-style lookups in lists that might use this
6294 variable. This has therefore been abolished, and the "net-" prefix is no
6295 longer necessary for query-style lookups.
6297 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
6298 have been a typo for LFLAGS, so it has been changed.
6300 13. The install script calls Exim with "-C /dev/null" in order to find the
6301 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
6302 to be output. However, since Exim outputs its version number before the
6303 error, it didn't break the script. It just looked ugly. I fixed this by
6304 always allowing "-C /dev/null" if the caller is root.
6306 14. Ignore overlarge ACL variable number when reading spool file - insurance
6307 against a later release with more variables having written the file.
6309 15. The standard form for an IPv6 address literal was being rejected by EHLO.
6310 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
6311 this, as well as the form without the "IPv6" on the front.
6313 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
6314 OS/Makefile-Darwin file.
6316 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
6317 only to LDAP libraries that do not have LDAP_OPT_DEREF.
6319 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
6320 variable. However, some OS use ints for inodes. I've added cast to long int
6321 to get rid of the compiler warning.
6323 19. I had forgotten to lock out "/../" in configuration file names when
6324 ALT_CONFIG_PREFIX was set.
6326 20. Routers used for verification do not need to specify transports. However,
6327 if such a router generated a host list, and callout was configured, Exim
6328 crashed, because it could not find a port number from the (non-existent)
6329 transport. It now assumes port 25 in this circumstance.
6331 21. Added the -t option to exigrep.
6333 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
6334 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
6335 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
6336 them is set, LOOKUP_LSEARCH is forced.
6338 23. "exim -bV" now outputs a list of lookups that are included in the binary.
6340 24. Added sender and host information to the "rejected by local_scan()" log
6341 line; previously there was no indication of these.
6343 25. Added .include_if_exists.
6345 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
6346 Linux. It turns out that not all file systems support this. Apparently some
6347 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
6348 do it.) To cope with this, the error EINVAL, which means that sync-ing is
6349 not supported on the file descriptor, is now ignored when Exim is trying to
6350 sync a directory. This applies only to Linux.
6352 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
6354 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
6355 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
6358 29. Added support for maildirsize files from supplied patch (modified a bit).
6360 30. The use of :fail: followed by an empty string could lead Exim to respond to
6361 sender verification failures with (e.g.):
6363 550 Verification failed for <xxx>
6364 550 Sender verify failed
6366 where the first response line was missing the '-' that indicates it is not
6367 the final line of the response.
6369 31. The loop for finding the name of the user that called Exim had a hardwired
6370 limit of 10; it now uses the value of finduser_retries, which is used for
6371 all other user lookups.
6373 32. Added $received_count variable, available in data and not_smtp ACLs, and at
6376 33. Exim was neglecting to zero errno before one call of strtol() when
6377 expanding a string and expecting an integer value. On some systems this
6378 resulted in spurious "integer overflow" errors. Also, it was casting the
6379 result into an int without checking.
6381 34. Testing for a connection timeout using "timeout_connect" in the retry rules
6382 did not work. The code looks as if it has *never* worked, though it appears
6383 to have been documented since at least release 1.62. I have made it work.
6385 35. The "timeout_DNS" error in retry rules, also documented since at least
6386 1.62, also never worked. As it isn't clear exactly what this means, and
6387 clearly it isn't a major issue, I have abolished the feature by treating it
6388 as "timeout", and writing a warning to the main and panic logs.
6390 36. The display of retry rules for -brt wasn't always showing the error code
6393 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
6394 timeout_connect_A, timeout_connect_MX.
6396 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
6397 to the empty sender.
6399 39. The daemon was not analysing the content of -oX till after it had closed
6400 stderr and disconnected from the controlling terminal. This meant that any
6401 syntax errors were only noted on the panic log, and the return code from
6402 the command was 0. By re-arranging the code a little, I've made the
6403 decoding happen first, so such errors now appear on stderr, and the return
6404 code is 1. However, the actual setting up of the sockets still happens in
6405 the disconnected process, so errors there are still only recorded on the
6408 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
6409 connections (as happens on some IP stacks) was logged at start up time as
6410 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
6411 it from "IPv6 and IPv4", which means that two separate sockets are being
6414 41. The debug output for gethostbyname2() or getipnodebyname() failures now
6415 says whether AF_INET or AF_INET6 was passed as an argument.
6417 42. Exiwhat output was messed up when time zones were included in log
6420 43. Exiwhat now gives more information about the daemon's listening ports,
6421 and whether -tls-on-connect was used.
6423 44. The "port" option of the smtp transport is now expanded.
6425 45. A "message" modifier in a "warn" statement in a non-message ACL was being
6426 silently ignored. Now an error message is written to the main and panic
6429 46. There's a new ACL modifier called "logwrite" which writes to a log file
6430 as soon as it is encountered.
6432 47. Added $local_user_uid and $local_user_gid at routing time.
6434 48. Exim crashed when trying to verify a sender address that was being
6437 49. Exim was recognizing only a space character after ".include". It now also
6438 recognizes a tab character.
6440 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
6441 extracting the relevant information from the specification. The man page no
6442 longer contains scrambled data for the -d option, and I've added a section
6443 at the front about calling Exim under different names.
6445 51. Added "extra_headers" argument to the "mail" command in filter files.
6447 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
6450 53. Installed eximstats 1.29.
6452 54. Added transport_filter_timeout as a generic transport option.
6454 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
6455 Cc: header lines. This was required by RFC 822, but it not required by RFC
6458 56. Exim used to add From:, Date:, and Message-Id: header lines to any
6459 incoming messages that did not have them. Now it does so only if the
6460 message originates locally, that is, if there is no associated remote host
6461 address. When Resent- header lines are present, this applies to the Resent-
6462 lines rather than the non-Resent- lines.
6464 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
6465 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
6467 58. Messages for configuration errors now include the name of the main
6468 configuration file - useful now that there may be more than one file in a
6469 list (.included file names were always shown).
6471 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
6472 for those rare installations that do not start the daemon as root or run it
6473 setuid root. I've cut out the call to initgroups() if the daemon is not
6476 60. The Exim user and group can now be bound into the binary as text strings
6477 that are looked up at the start of Exim's processing.
6479 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
6481 62. Added $mailstore_basename variable.
6483 63. Installed patch to sieve.c from Michael Haardt.
6485 64. When Exim failed to open the panic log after failing to open the main log,
6486 the original message it was trying to log was written to stderr and debug
6487 output, but if they were not available (the usual case in production), it
6488 was lost. Now it is written to syslog before the two lines that record the
6489 failures to open the logs.
6491 65. Users' Exim filters run in subprocesses under the user's uid. It is
6492 possible for a "deliver" command or an alias in a "personal" command to
6493 provoke an address rewrite. If logging of address rewriting is configured,
6494 this fails because the process is not running as root or exim. There may be
6495 a better way of dealing with this, but for the moment (because 4.30 needs
6496 to be released), I have disabled address rewrite logging when running a
6497 filter in a non-root, non-exim process.
6503 1. The buildconfig auxiliary program wasn't quoting the value set for
6504 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
6505 not defined. This bug was masked in 4.22 by the effect that was fixed in
6508 2. Some messages that were rejected after a message id was allocated were
6509 shown as "incomplete" by exigrep. It no longer does this for messages that
6510 are rejected by local_scan() or the DATA or non-SMTP ACLs.
6512 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
6513 have allow_domain_literals set, the ID did not get logged in the <= line.
6514 Domain literals are now always recognized in Message-ID: header lines.
6516 4. The first argument for a ${extract expansion item is the key name or field
6517 number. Leading and trailing spaces in this item were not being ignored,
6518 causing some misleading effects.
6520 5. When deliver_drop_privilege was set, single queue runner processes started
6521 manually (i.e. by the command "exim -q") or by the daemon (which uses the
6522 same command in the process it spins off) were not dropping privilege.
6524 6. When the daemon running as "exim" started a queue runner, it always
6525 re-executed Exim in the spun-off process. This is a waste of effort when
6526 deliver_drop_privilege is set. The new process now just calls the
6527 queue-runner function directly.
6533 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
6536 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
6537 ignored. Though the use of -oP was forcing the writing of a pid file, it
6538 was always written to the default place.
6540 3. If the message "no IP address found for host xxxx" is generated during
6541 incoming verification, it is now followed by identification of the incoming
6542 connection (so you can more easily find what provoked it).
6544 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
6546 5. Added some features to "harden" Exim a bit more against certain attacks:
6548 (a) There is now a build-time option called FIXED_NEVER_USERS that can
6549 be put in Local/Makefile. This is like the never_users runtime option,
6550 but it cannot be overridden. The default setting is "root".
6552 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
6553 prefix string with which any file named in a -C command line option
6556 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
6557 is retained for -C and -D only if the caller of Exim is root. Without
6558 it, the exim user may also use -C and -D and retain privilege.
6560 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
6561 command line option is disabled.
6563 6. Macro names set by the -D option must start with an upper case letter, just
6564 like macro names defined in the configuration file.
6566 7. Added "dereference=" facility to LDAP.
6568 8. Two instances of the typo "uknown" in the source files are fixed.
6570 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
6571 the Configure-Makefile script screwed up while processing it.
6573 10. Incorporated PCRE 4.4.
6575 11. The SMTP synchronization check was not operating right at the start of an
6576 SMTP session. For example, it could not catch a HELO sent before the client
6577 waited for the greeting. There is now a check for outstanding input at the
6578 point when the greeting is written. Because of the duplex, asynchronous
6579 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
6580 way, but not yet received, when the check is performed.
6582 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
6583 on TCP/IP sockets, because this apparently causes some broken clients to
6586 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
6587 unchanged) from the Cygwin maintainer.
6589 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
6590 was supported instead of testing for mbx. Effectively a typo.
6592 15. The spa authenticator server code was not checking that the input it
6593 received was valid base64.
6595 16. The debug output line for the "set" modifier in ACLs was not showing the
6596 name of the variable that was being set.
6598 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
6600 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
6601 lookup on its own. Something else was needed to trigger the lookup. For
6602 example, a match in host_lookup or the need for a host name in a host list.
6603 Now, if $sender_host_name is referenced and the host name has not yet been
6604 looked up, a lookup is performed. If the lookup fails, the variable remains
6605 empty, and $host_lookup_failed is set to "1".
6607 19. Added "eqi" as a case-independent comparison operator.
6609 20. The saslauthd authentication condition could segfault if neither service
6610 nor realm was specified.
6612 21. If an overflowing value such as "2048M" was set for message_size_limit, the
6613 error message that was logged was misleading, and incoming SMTP
6614 connections were dropped. The message is now more accurate, and temporary
6615 errors are given to SMTP connections.
6617 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
6618 (except RSET) with a 421 error, until QUIT is received. However, it was
6619 failing to send a response to QUIT.
6621 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
6622 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
6623 ACL is now run after the helo_try_verify_hosts code.
6625 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
6626 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
6627 case-independent, so other case variants are also recognized). Apparently
6628 some systems use these upper case variants.
6630 25. If more than two messages were waiting for the same host, and a transport
6631 filter was specified for the transport, Exim sent two messages over the
6632 same TCP/IP connection, and then failed with "socket operation on non-
6633 socket" when it tried to send the third.
6635 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
6637 27. The extern definition of crypt16() in expand.c was not being excluded when
6638 the OS had its own crypt16() function.
6640 28. Added bounce_return_body as a new option, and bounce_return_size_limit
6641 as a preferred synonym for return_size_limit, both as an option and as an
6644 29. Added LIBS=-liconv to OS/Makefile-OSF1.
6646 30. Changed the default configuration ACL to relax the local part checking rule
6647 for addresses that are not in any local domains. For these addresses,
6648 slashes and pipe symbols are allowed within local parts, but the sequence
6649 /../ is explicitly forbidden.
6651 31. SPA server authentication was not clearing the challenge buffer before
6654 32. log_message in a "warn" ACL statement was writing to the reject log as
6655 well as to the main log, which contradicts the documentation and doesn't
6656 seem right (because no rejection is happening). So I have stopped it.
6658 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
6659 However, I am unable to do any testing of this.
6661 34. Fixed an infelicity in the appendfile transport. When checking directories
6662 for a mailbox, to see if any needed to be created, it was accidentally
6663 using path names with one or more superfluous leading slashes; tracing
6664 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
6666 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
6667 contents are added to the log line that is written for every discarded
6668 recipient. (Previously a log_message setting was ignored.)
6670 36. The ${quote: operator now quotes the string if it is empty.
6672 37. The install script runs exim in order to find its version number. If for
6673 some reason other than non-existence or emptiness, which it checks, it
6674 could not run './exim', it was installing it with an empty version number,
6675 i.e. as "exim-". This error state is now caught, and the installation is
6678 38. An argument was missing from the function that creates an error message
6679 when Exim fails to connect to the socket for saslauthd authentication.
6680 This could cause Exim to crash, or give a corrupted message.
6682 39. Added isip, isip4, and isip6 to ${if conditions.
6684 40. The ACL variables $acl_xx are now saved with the message, and can be
6685 accessed later in routers, transports, and filters.
6687 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
6688 strings in the file are not string-expanded.
6690 42. If a MAIL command specified a SIZE value that was too large to fit into an
6691 int variable, the check against message_size_limit failed. Such values are
6692 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
6693 day this will have to be increased, but I don't think I want to be around
6694 when emails are that large.
6701 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
6702 iconv() is not standard in FreeBSD.
6704 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
6705 IPv6 enabled. The observed symptom was a segmentation fault on return from
6706 the function os_common_find_running_interfaces() in src/os.c.
6708 3. In the check_special_case() function in daemon.c I had used "errno" as an
6709 argument name, which causes warnings on some systems. This was basically a
6710 typo, since it was named "eno" in the comments!
6712 4. The code that waits for the clock to tick (at a resolution of some fraction
6713 of a second) so as to ensure message-id uniqueness was always waiting for
6714 at least one whole tick, when it could have waited for less. [This is
6715 almost certainly not relevant at current processor speeds, where it is
6716 unlikely to ever wait at all. But we try to future-proof.]
6718 5. The function that sleeps for a time interval that includes fractions of a
6719 second contained a race. It did not block SIGALRM between setting the
6720 timer, and suspending (a couple of lines later). If the interval was short
6721 and the sigsuspend() was delayed until after it had expired, the suspension
6722 never ended. On busy systems this could lead to processes getting stuck for
6725 6. Some uncommon configurations may cause a lookup to happen in a queue runner
6726 process, before it forks any delivery processes. The open lookup caching
6727 mechanism meant that the open file or database connection was passed into
6728 the delivery process. The problem was that delivery processes always tidy
6729 up cached lookup data. This could cause a problem for the next delivery
6730 process started by the queue runner, because the external queue runner
6731 process does not know about the closure. So the next delivery process
6732 still has data in the lookup cache. In the case of a file lookup, there was
6733 no problem because closing a file descriptor in a subprocess doesn't affect
6734 the parent. However, if the lookup was caching a connection to a database,
6735 the connection was closed, and the second delivery process was likely to
6736 see errors such as "PGSQL: query failed: server closed the connection
6737 unexpectedly". The problem has been fixed by closing all cached lookups
6738 in a queue runner before running a delivery process.
6740 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
6741 seem to have the "const" qualifier which it has on other OS. I've
6744 8. Change 4.21/2 was too strict. It is only if there are two authenticators
6745 *of the same type* (client or server) with the same public name that an
6746 error should be diagnosed.
6748 9. When Exim looked up a host name for an IP address, but failed to find the
6749 original IP address when looking up the host name (a safety check), it
6750 output the message "<ip address> does not match any IP for NULL", which was
6751 confusing, to say the least. The bug was that the host name should have
6752 appeared instead of "NULL".
6754 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
6755 user that is built into the binary, and the -C or -D options is used, root
6756 privilege is dropped before the configuration file is read. In addition,
6757 logging is switched to stderr instead of the normal log files. If the
6758 configuration then re-defines the Exim user, the unprivileged environment
6759 is probably not what is expected, so Exim logs a panic warning message (but
6762 However, if deliver_drop_privilege is set, the unprivileged state may well
6763 be exactly what is intended, so the warning has been cut out in that case,
6764 and Exim is allowed to try to write to its normal log files.
6770 1. smtp_return_error_details was not giving details for temporary sender
6771 or receiver verification errors.
6773 2. Diagnose a configuration error if two authenticators have the same public
6776 3. Exim used not to create the message log file for a message until the first
6777 delivery attempt. This could be confusing when incoming messages were held
6778 for policy or load reasons. The message log file is now created at the time
6779 the message is received, and an initial "Received" line is written to it.
6781 4. The automatically generated man page for command line options had a minor
6782 bug that caused no ill effects; however, a more serious problem was that
6783 the procedure for building the man page automatically didn't always
6784 operate. Consequently, release 4.20 contains an out-of-date version. This
6785 shouldn't happen again.
6787 5. When building Exim with embedded Perl support, the script that builds the
6788 Makefile was calling 'perl' to find its compile-time parameters, ignoring
6789 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
6791 6. The freeze_tell option was not being used for messages that were frozen on
6792 arrival, either by an ACL or by local_scan().
6794 7. Added the smtp_incomplete_transaction log selector.
6796 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
6797 was accepting AUTH without a new EHLO.
6799 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
6800 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
6803 10. Logging of TCP/IP connections (when configured) now happens in the main
6804 daemon process instead of the child process, so that the TCP/IP connection
6805 count is more accurate (but it can never be perfect).
6807 11. The use of "drop" in a nested ACL was not being handled correctly in the
6808 outer ACL. Now, if condition failure induced by the nested "drop" causes
6809 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
6810 or "require"), the connection is dropped.
6812 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
6813 that yield "discard" can now be used with an "accept" or a "discard" verb,
6814 but an error is generated for any others (because I can't see a useful way
6815 to define what should happen).
6817 13. When an ACL is read dynamically from a file (or anywhere else), the lines
6818 are now processed in the same way as lines in the Exim configuration file.
6819 In particular, continuation lines are supported.
6821 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
6823 15. Added -ti meaning -t -i.
6825 16. Check for letters, digits, hyphens, and dots in the names of dnslist
6826 domains, and warn by logging if others are found.
6828 17. At least on BSD, alignment is not guaranteed for the array of ifreq's
6829 returned from GIFCONF when Exim is trying to find the list of interfaces on
6830 a host. The code in os.c has been modified to copy each ifreq to an aligned
6831 structure in all cases.
6833 Also, in some cases, the returned ifreq's were being copied to a 'struct
6834 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
6835 means the last couple of bytes of an IPv6 address could be chopped if the
6836 ifreq contained only a normal sockaddr (14 bytes storage).
6838 18. Named domain lists were not supported in the hosts_treat_as_local option.
6839 An entry such as +xxxx was not recognized, and was treated as a literal
6842 19. Ensure that header lines added by a DATA ACL are included in the reject log
6843 if the ACL subsequently rejects the message.
6845 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
6846 MD5 (which is deprecated).
6848 21. When testing a filter file using -bf, Exim was writing a message when it
6849 took the sender from a "From " line in the message, but it was not doing so
6850 when it took $return_path from a Return-Path: header line. It now does.
6852 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
6853 with a valid header line field name (a series of printing characters
6854 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
6856 23. Changed "disc" in the source to "disk" to conform to the documentation and
6857 the book and for uniformity.
6859 24. Ignore Sendmail's -Ooption=value command line item.
6861 25. When execve() failed while trying to run a command in a pipe transport,
6862 Exim was returning EX_UNAVAILABLE (69) from the subprocess. However, this
6863 could be confused with a return value of 69 from the command itself. This
6864 has been changed to 127, the value the shell returns if it is asked to run
6865 a non-existent command. The wording for the related log line suggests a
6866 non-existent command as the problem.
6868 26. If received_header_text expands to an empty string, do not add a Received:
6869 header line to the message. (Well, it adds a token one on the spool, but
6870 marks it "old" so that it doesn't get used or transmitted.)
6872 27. Installed eximstats 1.28 (addition of -nt option).
6874 28. There was no check for failure on the call to getsockname() in the daemon
6875 code. This can fail if there is a shortage of resources on the system, with
6876 ENOMEM, for example. A temporary error is now given on failure.
6878 29. Contrary to the C standard, it seems that in some environments, the
6879 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
6880 program. Exim now does this explicitly; it affects the formatting of
6881 timestamps using strftime().
6883 30. If exiqsumm was given junk data, it threw up some uninitialized variable
6884 complaints. I've now initialized all the variables, to avoid this.
6886 32. Header lines added by a system filter were not being "seen" during
6887 transport-time rewrites.
6889 33. The info_callback() function passed to OpenSSL is set up with type void
6890 (*)(SSL *, int, int), as described somewhere. However, when calling the
6891 function (actually a macro) that sets it up, the type void(*)() is
6892 expected. I've put in a cast to prevent warnings from picky compilers.
6894 34. If a DNS black list lookup found a CNAME record, but there were no A
6895 records associated with the domain it pointed at, Exim crashed.
6897 35. If a DNS black list lookup returned more than one A record, Exim ignored
6898 all but the first. It now scans all returned addresses if a particular IP
6899 value is being sought. In this situation, the contents of the
6900 $dnslist_value variable are a list of all the addresses, separated by a
6903 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
6904 to accept a host name and all its aliases if the forward lookup for any of
6905 them yielded the IP address of the incoming connection. Now it accepts only
6906 those names whose forward lookup yields the correct IP address. Any other
6907 names are discarded. This closes a loophole whereby a rogue DNS
6908 administrator could create reverse DNS records to break through a
6909 wildcarded host restriction in an ACL.
6911 37. If a user filter or a system filter that ran in a subprocess used any of
6912 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
6913 the wrong values were passed to the pipe command ($thisaddress had the
6914 value of $0, $0 had the value of $1, etc). This bug was introduced by
6915 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
6917 38. Improved the line breaking for long SMTP error messages from ACLs.
6918 Previously, if there was no break point between 40 and 75 characters, Exim
6919 left the rest of the message alone. Two changes have been made: (a) I've
6920 reduced the minimum length to 35 characters; (b) if it can't find a break
6921 point between 35 and 75 characters, it looks ahead and uses the first one
6922 that it finds. This may give the occasional overlong line, but at least the
6923 remaining text gets split now.
6925 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
6926 file descriptors might be low, and that setting 1000 would always raise it.
6927 It turns out that in some environments, the limit is already over 1000 and
6928 that lowering it causes trouble. So now Exim takes care not to decrease it.
6930 40. When delivering a message, the value of $return_path is set to $sender_
6931 address at the start of routing (routers may change the value). By an
6932 oversight, this default was not being set up when an address was tested by
6933 -bt or -bv, which affected the outcome if any router or filter referred to
6936 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
6937 log only when "message" or "log_message" are set. However, if one of the
6938 conditions was an address verification, or a call to a nested ACL, the
6939 messages generated by the underlying test were being passed through. This
6940 no longer happens. The underlying message is available in $acl_verify_
6941 message for both "message" and "log_message" expansions, so it can be
6942 passed through if needed.
6944 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
6945 new expansion $bh_ to give the encoded byte string without charset
6946 translation. Translation happens only if iconv() is available; HAVE_ICONV
6947 indicates this at build time. HEADERS_CHARSET gives the charset to
6948 translate to; headers_charset can change it in the configuration, and
6949 "headers charset" can change it in an individual filter file.
6951 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
6952 that creates RFC 2047 encoded "words" labels them as that charset instead
6953 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
6954 expansion operator; (ii) when Exim creates a From: line for a local
6955 message; (iii) when a header line is rewritten to include a "phrase" part.
6957 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
6958 buggy, causing it to skip the first lines of messages whose message ID
6959 ended in 'D'. This would not have bitten before Exim release 4.14, because
6960 message IDs were unlikely to end in 'D' before then. The effect was to have
6961 incorrect size information for certain domains.
6963 45. #include "config.h" was missing at the start of the crypt16.c module. This
6964 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
6967 46. If there was a timeout during a "random" callout check, Exim treated it as
6968 a failure of the random address, and carried on sending RSET and the real
6969 address. If the delay was just some slowness somewhere, the response to the
6970 original RCPT would be taken as a response to RSET and so on, causing
6971 mayhem of various kinds.
6973 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
6974 when I implemented it. It didn't allow for the fact that some option values
6975 may legitimately be negative (e.g. size_addition), and it didn't even do
6976 the right test for positive values.
6978 48. Domain names in DNS records are case-independent. Exim always looks them up
6979 in lower case. Some resolvers return domain names in exactly the case they
6980 appear in the zone file, that is, they may contain uppercase letters. Not
6981 all resolvers do this - some return always lower case. Exim was treating a
6982 change of case by a resolver as a change of domain, similar to a widening
6983 of a domain abbreviation. This triggered its re-routing code and so it was
6984 trying to route what was effectively the same domain again. This normally
6985 caused routing to fail (because the router wouldn't handle the domain
6986 twice). Now Exim checks for this case specially, and just changes the
6987 casing of the domain that it ultimately uses when it transmits the message
6990 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
6993 50. If a filter generated a file delivery with a non-absolute name (possible if
6994 no home directory exists for the router), the forbid_file option was not
6997 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
6998 the existing equality matching.
7000 52. Exim was using ints instead of ino_t variables in some places where it was
7001 dealing with inode numbers.
7003 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
7004 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
7005 called TMPDIR, and if it finds it is different, it changes its value.
7007 54. The smtp_printf() function is now made available to local_scan() so
7008 additional output lines can be written before returning. There is also an
7009 smtp_fflush() function to enable the detection of a dropped connection.
7010 The variables smtp_input and smtp_batched_input are exported to
7013 55. Changed the default runtime configuration: the message "Unknown user"
7014 has been removed from the ACL, and instead placed on the localuser router,
7015 using the cannot_route_message feature. This means that any verification
7016 failures that generate their own messages won't get overridden. Similarly,
7017 the "Unrouteable address" message that was in the ACL for unverifiable
7018 relay addresses has also been removed.
7020 56. Added hosts_avoid_esmtp to the smtp transport.
7022 57. The exicyclog script was not checking for the esoteric option
7023 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
7024 will work only if exicyclog is run under the appropriate euid.
7026 58. Following a discussion on the list, the rules by which Exim recognises line
7027 endings on incoming messages have been changed. The -dropcr and drop_cr
7028 options are now no-ops, retained only for backwards compatibility. The
7029 following line terminators are recognized: LF CRLF CR. However, special
7030 processing applies to CR:
7032 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
7033 nor a local message in the state where . is a terminator.
7035 (ii) If a bare CR is encountered in a header line, an extra space is added
7036 after the line terminator so as not to end the header. The reasoning
7037 behind this is that bare CRs in header lines are most likely either
7038 to be mistakes, or people trying to play silly games.
7040 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
7041 was being incorrectly given as 18 bytes larger than it should have been.
7042 This is a VOB (very old bug).
7044 60. This may never have affected anything current, but just in case it has:
7045 When the local host is found other than at the start of a list of hosts,
7046 the local host, those with the same MX, and any that follow, are discarded.
7047 When the list in question was part of a longer list of hosts, the following
7048 hosts (not currently being processed) were also being discarded. This no
7049 longer happens. I'm not sure if this situation could ever has previously
7052 61. Added the "/MX" feature to lists of hosts in the manualroute and query
7055 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
7056 header. This is something that is recommended in a new Internet Draft, and
7057 is something that is documented as being done by Sendmail. There are two
7058 possible values. For messages generated by the autoreply transport, Exim
7061 Auto-Submitted: auto-replied
7063 whereas for all other generated messages (e.g. bounces) it adds
7065 Auto-Submitted: auto-generated
7067 63. The "personal" condition in filters now includes a test for the
7068 Auto-Submitted: header. If it contains the string "auto-" the message it
7069 not considered personal.
7071 64. Added rcpt_include_affixes as a generic transport option.
7073 65. Added queue_only_override (default true).
7075 66. Added the syslog_duplication option.
7077 67. If what should have been the first header line of a message consisted of
7078 a space followed by a colon, Exim was mis-interpreting it as a header line.
7079 It isn't of course - it is syntactically invalid and should therefore be
7080 treated as the start of the message body. The misbehaviour could have
7081 caused a number of strange effects, including loss of data in subsequent
7082 header lines, and spool format errors.
7084 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
7085 client host had authenticated. This control can now be exercised by an ACL
7086 for more flexibility.
7088 69. By default, callouts do not happen when testing with -bh. There is now a
7089 variant, -bhc, which does actually run the callout code, including
7090 consulting and updating the callout cache.
7092 70. Added support for saslauthd authentication, courtesy of Alexander
7095 71. If statvfs() failed on the spool or log directories while checking their
7096 size for availability, Exim confusingly gave the error "space shortage".
7097 Furthermore, in debugging mode it crashed with a floating point exception.
7098 These checks are done if check_{spool,log}_{space,inodes} are set, and when
7099 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
7100 serious problem, Exim now writes to the main and panic logs when this
7101 happens, with details of the failure. It then refuses to accept the
7102 incoming message, giving the message "spool directory problem" or "log
7103 directory problem" with a 421 code for SMTP messages.
7105 72. When Exim is about to re-exec itself, it ensures that the file descriptors
7106 0, 1, and 2 exist, because some OS complain for execs without them (see
7107 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
7108 descriptors. However, the code omitted to check that the open succeeded,
7109 causing mysterious errors if for some reason the permissions on /dev/null
7110 got screwed. Now Exim writes a message to the main and panic logs, and
7111 bombs out if it can't open /dev/null.
7113 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
7114 interact so that it is all more flexible. It is supposed to remain
7115 backwards compatible. Also added extra_local_interfaces.
7117 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
7118 to bomb out with an assertion failure - to the client this appears as a
7119 connection drop. This problem occurs in the part of the code that was taken
7120 from the Samba project. Fortunately, the assertion is in a very simple
7121 function, so I have fixed this by reproducing the function inline in the
7122 one place where it is called, and arranging for authentication to fail
7123 instead of killing the process with assert().
7125 75. The SPA client code was not working when the server requested OEM rather
7126 than Unicode encoding.
7128 76. Added code to make require_files with a specific uid setting more usable in
7129 the case where statting the file as root fails - usually a non-root-mounted
7130 NFS file system. When this happens and the failure is EACCES, Exim now
7131 forks a subprocess and does the per-uid checking as the relevant uid.
7133 77. Added process_log_path.
7135 78. If log_file_path was not explicitly set, a setting of check_log_space or
7136 check_log_inodes was ignored.
7138 79. If a space check for the spool or log partitions fails, the incident is now
7139 logged. Of course, in the latter case the data may get lost...
7141 80. Added the %p formatting code to string_format() so that it can be used to
7142 print addresses in debug_print(). Adjusted all the address printing in the
7143 debugging in store.c to use %p rather than %d.
7145 81. There was a concern that a line of code in smtp_in.c could overflow a
7146 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
7147 initially expressed, the concern was not well-founded, because trailing
7148 spaces are removed early. However, if the trailing spaces were followed by
7149 a NULL, they did not get removed, so the overflow was possible. Two fixes
7152 (a) I re-wrote the offending code in a cleaner fashion.
7153 (b) If an incoming SMTP command contains a NULL character, it is rejected
7156 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
7157 runs initgroups(), so that if the Exim user is in any additional groups,
7158 they will be used during message reception.
7164 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / users / jgh / exim.git / blob