1 Change log file for Exim from version 4.21
2 ------------------------------------------
3 This document describes *changes* to previous versions, that might
4 affect Exim's operation, with an unchanged configuration file. For new
5 options, and new features, see the NewStuff file next to this ChangeLog.
8 Since Exim version 4.90
9 -----------------------
11 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.
190 JH/01 Rework error string handling in TLS interface so that the caller in
191 more cases is responsible for logging. This permits library-sourced
192 string to be attached to addresses during delivery, and collapses
193 pairs of long lines into single ones.
195 PP/01 Allow PKG_CONFIG_PATH to be set in Local/Makefile and use it correctly
196 during configuration. Wildcards are allowed and expanded.
198 JH/02 Rework error string handling in DKIM to pass more info back to callers.
199 This permits better logging.
201 JH/03 Rework the transport continued-connection mechanism: when TLS is active,
202 do not close it down and have the child transport start it up again on
203 the passed-on TCP connection. Instead, proxy the child (and any
204 subsequent ones) for TLS via a unix-domain socket channel. Logging is
205 affected: the continued delivery log lines do not have any DNSSEC, TLS
206 Certificate or OCSP information. TLS cipher information is still logged.
208 JH/04 Shorten the log line for daemon startup by collapsing adjacent sets of
209 identical IP addresses on different listening ports. Will also affect
212 PP/02 Bug 2070: uClibc defines __GLIBC__ without providing glibc headers;
213 add noisy ifdef guards to special-case this sillyness.
214 Patch from Bernd Kuhls.
216 JH/05 Tighten up the checking in isip4 (et al): dotted-quad components larger
217 than 255 are no longer allowed.
219 JH/06 Default openssl_options to include +no_ticket, to reduce load on peers.
220 Disable the session-cache too, which might reduce our load. Since we
221 currrectly use a new context for every connection, both as server and
222 client, there is no benefit for these.
223 GnuTLS appears to not support tickets server-side by default (we don't
224 call gnutls_session_ticket_enable_server()) but client side is enabled
225 by default on recent versions (3.1.3 +) unless the PFS priority string
228 PP/03 Add $SOURCE_DATE_EPOCH support for reproducible builds, per spec at
229 <https://reproducible-builds.org/specs/source-date-epoch/>.
231 JH/07 Fix smtp transport use of limited max_rcpt under mua_wrapper. Previously
232 the check for any unsuccessful recipients did not notice the limit, and
233 erroneously found still-pending ones.
235 JH/08 Pipeline CHUNKING command and data together, on kernels that support
236 MSG_MORE. Only in-clear (not on TLS connections).
238 JH/09 Avoid using a temporary file during transport using dkim. Unless a
239 transport-filter is involved we can buffer the headers in memory for
240 creating the signature, and read the spool data file once for the
241 signature and again for transmission.
243 JH/10 Enable use of sendfile in Linux builds as default. It was disabled in
244 4.77 as the kernel support then wasn't solid, having issues in 64bit
245 mode. Now, it's been long enough. Add support for FreeBSD also.
247 JH/11 Bug 2104: Fix continued use of a transport connection with TLS. In the
248 case where the routing stage had gathered several addresses to send to
249 a host before calling the transport for the first, we previously failed
250 to close down TLS in the old transport process before passing the TCP
251 connection to the new process. The new one sent a STARTTLS command
252 which naturally failed, giving a failed delivery and bloating the retry
253 database. Investigation and fix prototype from Wolfgang Breyha.
255 JH/12 Fix check on SMTP command input synchronisation. Previously there were
256 false-negatives in the check that the sender had not preempted a response
257 or prompt from Exim (running as a server), due to that code's lack of
258 awareness of the SMTP input buffering.
260 PP/04 Add commandline_checks_require_admin option.
261 Exim drops privileges sanely, various checks such as -be aren't a
262 security problem, as long as you trust local users with access to their
263 own account. When invoked by services which pass untrusted data to
264 Exim, this might be an issue. Set this option in main configuration
265 AND make fixes to the calling application, such as using `--` to stop
268 JH/13 Do pipelining under TLS. Previously, although safe, no advantage was
269 taken. Now take care to pack both (client) MAIL,RCPT,DATA, and (server)
270 responses to those, into a single TLS record each way (this usually means
271 a single packet). As a side issue, smtp_enforce_sync now works on TLS
274 PP/05 OpenSSL/1.1: use DH_bits() for more accurate DH param sizes. This
275 affects you only if you're dancing at the edge of the param size limits.
276 If you are, and this message makes sense to you, then: raise the
277 configured limit or use OpenSSL 1.1. Nothing we can do for older
280 JH/14 For the "sock" variant of the malware scanner interface, accept an empty
281 cmdline element to get the documented default one. Previously it was
284 JH/15 Fix a crash in the smtp transport caused when two hosts in succession
285 are unsuable for non-message-specific reasons - eg. connection timeout,
286 banner-time rejection.
288 JH/16 Fix logging of delivery remote port, when specified by router, under
291 PP/06 Repair manualroute's ability to take options in any order, even if one
292 is the name of a transport.
295 HS/01 Cleanup, prevent repeated use of -p/-oMr (CVE-2017-1000369)
297 JH/17 Change the list-building routines interface to use the expanding-string
298 triplet model, for better allocation and copying behaviour.
300 JH/18 Prebuild the data-structure for "builtin" macros, for faster startup.
301 Previously it was constructed the first time a possibly-matching string
302 was met in the configuration file input during startup; now it is done
305 JH/19 Bug 2141: Use the full-complex API for Berkeley DB rather than the legacy-
306 compatible one, to avoid the (poorly documented) possibility of a config
307 file in the working directory redirecting the DB files, possibly correpting
308 some existing file. CVE-2017-10140 assigned for BDB.
310 JH/20 Bug 2147: Do not defer for a verify-with-callout-and-random which is not
311 cache-hot. Previously, although the result was properly cached, the
312 initial verify call returned a defer.
314 JH/21 Bug 2151: Avoid using SIZE on the MAIL for a callout verify, on any but
315 the main verify for receipient in uncached-mode.
317 JH/22 Retire historical build files to an "unsupported" subdir. These are
318 defined as "ones for which we have no current evidence of testing".
320 JH/23 DKIM: enforce the DNS pubkey record "h" permitted-hashes optional field,
321 if present. Previously it was ignored.
323 JH/24 Start using specified-initialisers in C structure init coding. This is
324 a C99 feature (it's 2017, so now considered safe).
326 JH/25 Use one-bit bitfields for flags in the "addr" data structure. Previously
327 if was a fixed-sized field and bitmask ops via macros; it is now more
330 PP/07 GitHub PR 56: Apply MariaDB build fix.
331 Patch provided by Jaroslav Škarvada.
333 PP/08 Bug 2161: Fix regression in sieve quoted-printable handling introduced
334 during Coverity cleanups [4.87 JH/47]
335 Diagnosis and fix provided by Michael Fischer v. Mollard.
337 JH/26 Fix DKIM bug: when the pseudoheader generated for signing was exactly
338 the right size to place the terminating semicolon on its own folded
339 line, the header hash was calculated to an incorrect value thanks to
340 the (relaxed) space the fold became.
342 HS/02 Fix Bug 2130: large writes from the transport subprocess where chunked
343 and confused the parent.
345 JH/27 Fix SOCKS bug: an unitialized pointer was deref'd by the transport process
346 which could crash as a result. This could lead to undeliverable messages.
348 JH/28 Logging: "next input sent too soon" now shows where input was truncated
351 JH/29 Fix queue_run_in_order to ignore the PID portion of the message ID. This
352 matters on fast-turnover and PID-randomising systems, which were getting
353 out-of-order delivery.
355 JH/30 Fix a logging bug on aarch64: an unsafe routine was previously used for
356 a possibly-overlapping copy. The symptom was that "Remote host closed
357 connection in response to HELO" was logged instead of the actual 4xx
360 JH/31 Fix CHUNKING code to properly flush the unwanted chunk after an error.
361 Previously only that bufferd was discarded, resulting in SYMTP command
364 JH/32 DKIM: when a message has multiple signatures matching an identity given
365 in dkim_verify_signers, run the dkim acl once for each. Previously only
366 one run was done. Bug 2189.
368 JH/33 Downgrade an unfound-list name (usually a typo in the config file) from
369 "panic the current process" to "deliberately defer". The panic log is
370 still written with the problem list name; the mail and reject logs now
371 get a temp-reject line for the message that was being handled, saying
372 something like "domains check lookup or other defer". The SMTP 451
373 message is still "Temporary local problem".
375 JH/34 Bug 2199: Fix a use-after-free while reading smtp input for header lines.
376 A crafted sequence of BDAT commands could result in in-use memory beeing
377 freed. CVE-2017-16943.
379 HS/03 Bug 2201: Fix checking for leading-dot on a line during headers reading
380 from SMTP input. Previously it was always done; now only done for DATA
381 and not BDAT commands. CVE-2017-16944.
383 JH/35 Bug 2201: Flush received data in BDAT mode after detecting an error fatal
384 to the message (such as an overlong header line). Previously this was
385 not done and we did not exit BDAT mode. Followon from the previous item
386 though a different problem.
392 JH/01 Bug 1922: Support IDNA2008. This has slightly different conversion rules
393 than -2003 did; needs libidn2 in addition to libidn.
395 JH/02 The path option on a pipe transport is now expanded before use.
397 PP/01 GitHub PR 50: Do not call ldap_start_tls_s on ldapi:// connections.
398 Patch provided by "Björn", documentation fix added too.
400 JH/03 Bug 2003: fix Proxy Protocol v2 handling: the address size field was
401 missing a wire-to-host endian conversion.
403 JH/04 Bug 2004: fix CHUNKING in non-PIPELINEING mode. Chunk data following
404 close after a BDAT command line could be taken as a following command,
405 giving a synch failure. Fix by only checking for synch immediately
406 before acknowledging the chunk.
408 PP/02 GitHub PR 52: many spelling fixes, which include fixing parsing of
409 no_require_dnssec option and creation of _HAVE_TRANSPORT_APPEND_MAILDIR
410 macro. Patches provided by Josh Soref.
412 JH/05 Have the EHLO response advertise VRFY, if there is a vrfy ACL defined.
413 Previously we did not; the RFC seems ambiguous and VRFY is not listed
414 by IANA as a service extension. However, John Klensin suggests that we
417 JH/06 Bug 2017: Fix DKIM verification in -bh test mode. The data feed into
418 the dkim code may be unix-mode line endings rather than smtp wire-format
419 CRLF, so prepend a CR to any bare LF.
421 JH/07 Rationalise the coding for callout smtp conversations and transport ones.
422 As a side-benfit, callouts can now use PIPELINING hence fewer round-trips.
424 JH/08 Bug 2016: Fix DKIM verification vs. CHUNKING. Any BDAT commands after
425 the first were themselves being wrongly included in the feed into dkim
426 processing; with most chunk sizes in use this resulted in an incorrect
427 body hash calculated value.
429 JH/09 Bug 2014: permit inclusion of a DKIM-Signature header in a received
430 DKIM signature block, for verification. Although advised against by
431 standards it is specifically not ruled illegal.
433 JH/10 Bug 2025: Fix reception of (quoted) local-parts with embedded spaces.
435 JH/11 Bug 2029: Fix crash in DKIM verification when a message signature block is
436 missing a body hash (the bh= tag).
438 JH/12 Bug 2018: Re-order Proxy Protocol startup versus TLS-on-connect startup.
439 It seems that HAProxy sends the Proxy Protocol information in clear and
440 only then does a TLS startup, so do the same.
442 JH/13 Bug 2027: Avoid attempting to use TCP Fast Open for non-transport client
443 TCP connections (such as for Spamd) unless the daemon successfully set
444 Fast Open mode on its listening sockets. This fixes breakage seen on
445 too-old kernels or those not configured for Fast Open, at the cost of
446 requiring both directions being enabled for TFO, and TFO never being used
447 by non-daemon-related Exim processes.
449 JH/14 Bug 2000: Reject messages recieved with CHUNKING but with malformed line
450 endings, at least on the first header line. Try to canonify any that get
451 past that check, despite the cost.
453 JH/15 Angle-bracket nesting (an error inserted by broken sendmails) levels are
454 now limited to an arbitrary five deep, while parsing addresses with the
455 strip_excess_angle_brackets option enabled.
457 PP/03 Bug 2018: For Proxy Protocol and TLS-on-connect, do not over-read and
458 instead leave the unprompted TLS handshake in socket buffer for the
459 TLS library to consume.
461 PP/04 Bug 2018: Also handle Proxy Protocol v2 safely.
463 PP/05 FreeBSD compat: handle that Ports no longer create /usr/bin/perl
465 JH/16 Drop variables when they go out of scope. Memory management drops a whole
466 region in one operation, for speed, and this leaves assigned pointers
467 dangling. Add checks run only under the testsuite which checks all
468 variables at a store-reset and panics on a dangling pointer; add code
469 explicitly nulling out all the variables discovered. Fixes one known
470 bug: a transport crash, where a dangling pointer for $sending_ip_address
471 originally assigned in a verify callout, is re-used.
473 PP/06 Drop '.' from @INC in various Perl scripts.
475 PP/07 Switch FreeBSD iconv to always use the base-system libc functions.
477 PP/08 Reduce a number of compilation warnings under clang; building with
478 CC=clang CFLAGS+=-Wno-dangling-else -Wno-logical-op-parentheses
479 should be warning-free.
481 JH/17 Fix inbound CHUNKING when DKIM disabled at runtime.
483 HS/01 Fix portability problems introduced by PP/08 for platforms where
484 realloc(NULL) is not equivalent to malloc() [SunOS et al].
486 HS/02 Bug 1974: Fix missing line terminator on the last received BDAT
487 chunk. This allows us to accept broken chunked messages. We need a more
488 general solution here.
490 PP/09 Wrote util/chunking_fixqueue_finalnewlines.pl to help recover
491 already-broken messages in the queue.
493 JH/18 Bug 2061: Fix ${extract } corrupting an enclosing ${reduce } $value.
495 JH/19 Fix reference counting bug in routing-generated-address tracking.
501 JH/01 Use SIZE on MAIL FROM in a cutthrough connection, if the destination
502 supports it and a size is available (ie. the sending peer gave us one).
504 JH/02 The obsolete acl condition "demime" is removed (finally, after ten
505 years of being deprecated). The replacements are the ACLs
506 acl_smtp_mime and acl_not_smtp_mime.
508 JH/03 Upgrade security requirements imposed for hosts_try_dane: previously
509 a downgraded non-dane trust-anchor for the TLS connection (CA-style)
510 or even an in-clear connection were permitted. Now, if the host lookup
511 was dnssec and dane was requested then the host is only used if the
512 TLSA lookup succeeds and is dnssec. Further hosts (eg. lower priority
513 MXs) will be tried (for hosts_try_dane though not for hosts_require_dane)
514 if one fails this test.
515 This means that a poorly-configured remote DNS will make it incommunicado;
516 but it protects against a DNS-interception attack on it.
518 JH/04 Bug 1810: make continued-use of an open smtp transport connection
519 non-noisy when a race steals the message being considered.
521 JH/05 If main configuration option tls_certificate is unset, generate a
522 self-signed certificate for inbound TLS connections.
524 JH/06 Bug 165: hide more cases of password exposure - this time in expansions
525 in rewrites and routers.
527 JH/07 Retire gnutls_require_mac et.al. These were nonfunctional since 4.80
528 and logged a warning sing 4.83; now they are a configuration file error.
530 JH/08 Bug 1836: Fix crash in VRFY handling when handed an unqualified name
531 (lacking @domain). Apply the same qualification processing as RCPT.
533 JH/09 Bug 1804: Avoid writing msglog files when in -bh or -bhc mode.
535 JH/10 Support ${sha256:} applied to a string (as well as the previous
538 JH/11 Cutthrough: avoid using the callout hints db on a verify callout when
539 a cutthrough deliver is pending, as we always want to make a connection.
540 This also avoids re-routing the message when later placing the cutthrough
541 connection after a verify cache hit.
542 Do not update it with the verify result either.
544 JH/12 Cutthrough: disable when verify option success_on_redirect is used, and
545 when routing results in more than one destination address.
547 JH/13 Cutthrough: expand transport dkim_domain option when testing for dkim
548 signing (which inhibits the cutthrough capability). Previously only
549 the presence of an option was tested; now an expansion evaluating as
550 empty is permissible (obviously it should depend only on data available
551 when the cutthrough connection is made).
553 JH/14 Fix logging of errors under PIPELINING. Previously the log line giving
554 the relevant preceding SMTP command did not note the pipelining mode.
556 JH/15 Fix counting of empty lines in $body_linecount and $message_linecount.
557 Previously they were not counted.
559 JH/16 DANE: treat a TLSA lookup response having all non-TLSA RRs, the same
560 as one having no matching records. Previously we deferred the message
561 that needed the lookup.
563 JH/17 Fakereject: previously logged as a normal message arrival "<="; now
564 distinguished as "(=".
566 JH/18 Bug 1867: make the fail_defer_domains option on a dnslookup router work
567 for missing MX records. Previously it only worked for missing A records.
569 JH/19 Bug 1850: support Radius libraries that return REJECT_RC.
571 JH/20 Bug 1872: Ensure that acl_smtp_notquit is run when the connection drops
572 after the data-go-ahead and data-ack. Patch from Jason Betts.
574 JH/21 Bug 1846: Send DMARC forensic reports for reject and quarantine results,
575 even for a "none" policy. Patch from Tony Meyer.
577 JH/22 Fix continued use of a connection for further deliveries. If a port was
578 specified by a router, it must also match for the delivery to be
581 JH/23 Bug 1874: fix continued use of a connection for further deliveries.
582 When one of the recipients of a message was unsuitable for the connection
583 (has no matching addresses), we lost track of needing to mark it
584 deferred. As a result mail would be lost.
586 JH/24 Bug 1832: Log EHLO response on getting conn-close response for HELO.
588 JH/25 Decoding ACL controls is now done using a binary search; the source code
589 takes up less space and should be simpler to maintain. Merge the ACL
590 condition decode tables also, with similar effect.
592 JH/26 Fix problem with one_time used on a redirect router which returned the
593 parent address unchanged. A retry would see the parent address marked as
594 delivered, so not attempt the (identical) child. As a result mail would
597 JH/27 Fix a possible security hole, wherein a process operating with the Exim
598 UID can gain a root shell. Credit to http://www.halfdog.net/ for
599 discovery and writeup. Ubuntu bug 1580454; no bug raised against Exim
602 JH/28 Enable {spool,log} filesystem space and inode checks as default.
603 Main config options check_{log,spool}_{inodes,space} are now
604 100 inodes, 10MB unless set otherwise in the configuration.
606 JH/29 Fix the connection_reject log selector to apply to the connect ACL.
607 Previously it only applied to the main-section connection policy
610 JH/30 Bug 1897: fix callouts connection fallback from TLS to cleartext.
612 PP/01 Changed default Diffie-Hellman parameters to be Exim-specific, created
613 by me. Added RFC7919 DH primes as an alternative.
615 PP/02 Unbreak build via pkg-config with new hash support when crypto headers
616 are not in the system include path.
618 JH/31 Fix longstanding bug with aborted TLS server connection handling. Under
619 GnuTLS, when a session startup failed (eg because the client disconnected)
620 Exim did stdio operations after fclose. This was exposed by a recent
621 change which nulled out the file handle after the fclose.
623 JH/32 Bug 1909: Fix OCSP proof verification for cases where the proof is
624 signed directly by the cert-signing cert, rather than an intermediate
625 OCSP-signing cert. This is the model used by LetsEncrypt.
627 JH/33 Bug 1914: Ensure socket is nonblocking before draining after SMTP QUIT.
629 HS/01 Fix leak in verify callout under GnuTLS, about 3MB per recipient on
630 an incoming connection.
632 HS/02 Bug 1802: Do not half-close the connection after sending a request
635 HS/03 Use "auto" as the default EC curve parameter. For OpenSSL < 1.0.2
636 fallback to "prime256v1".
638 JH/34 SECURITY: Use proper copy of DATA command in error message.
639 Could leak key material. Remotely exploitable. CVE-2016-9963.
645 JH/01 Bug 1664: Disable OCSP for GnuTLS library versions at/before 3.3.16
646 and 3.4.4 - once the server is enabled to respond to an OCSP request
647 it does even when not requested, resulting in a stapling non-aware
648 client dropping the TLS connection.
650 TF/01 Code cleanup: Overhaul the debug_selector and log_selector machinery to
651 support variable-length bit vectors. No functional change.
653 TF/02 Improve the consistency of logging incoming and outgoing interfaces.
654 The I= interface field on outgoing lines is now after the H= remote
655 host field, same as incoming lines. There is a separate
656 outgoing_interface log selector which allows you to disable the
659 JH/02 Bug 728: Close logfiles after a daemon-process "exceptional" log write.
660 If not running log_selector +smtp_connection the mainlog would be held
661 open indefinitely after a "too many connections" event, including to a
662 deleted file after a log rotate. Leave the per net connection logging
663 leaving it open for efficiency as that will be quickly detected by the
664 check on the next write.
666 HS/01 Bug 1671: Fix post transport crash.
667 Processing the wait-<transport> messages could crash the delivery
668 process if the message IDs didn't exist for some reason. When
669 using 'split_spool_directory=yes' the construction of the spool
670 file name failed already, exposing the same netto behaviour.
672 JH/03 Bug 425: Capture substrings in $regex1, $regex2 etc from regex &
673 mime_regex ACL conditions.
675 JH/04 Bug 1686: When compiled with EXPERIMENTAL_DSN_INFO: Add extra information
676 to DSN fail messages (bounces): remote IP, remote greeting, remote response
677 to HELO, local diagnostic string.
679 JH/05 Downgrade message for a TLS-certificate-based authentication fail from
680 log line to debug. Even when configured with a tls authenticator many
681 client connections are expected to not authenticate in this way, so
682 an authenticate fail is not an error.
684 HS/02 Add the Exim version string to the process info. This way exiwhat
685 gives some more detail about the running daemon.
687 JH/06 Bug 1395: time-limit caching of DNS lookups, to the TTL value. This may
688 matter for fast-change records such as DNSBLs.
690 JH/07 Bug 1678: Always record an interface option value, if set, as part of a
691 retry record, even if constant. There may be multiple transports with
692 different interface settings and the retry behaviour needs to be kept
695 JH/08 Bug 1586: exiqgrep now refuses to run if there are unexpected arguments.
697 JH/09 Bug 1700: ignore space & tab embedded in base64 during decode.
699 JH/10 Bug 840: fix log_defer_output option of pipe transport
701 JH/11 Bug 830: use same host for all RCPTS of a message, even under
702 hosts_randomize. This matters a lot when combined with mua_wrapper.
704 JH/12 Bug 1706: percent and underbar characters are no longer escaped by the
705 ${quote_pgsql:<string>} operator.
707 JH/13 Bug 1708: avoid misaligned access in cached lookup.
709 JH/14 Change header file name for freeradius-client. Relevant if compiling
710 with Radius support; from the Gentoo tree and checked under Fedora.
712 JH/15 Bug 1712: Introduce $prdr_requested flag variable
714 JH/16 Bug 1714: Permit an empty string as expansion result for transport
715 option transport_filter, meaning no filtering.
717 JH/17 Bug 1713: Fix non-PDKIM_DEBUG build. Patch from Jasen Betts.
719 JH/18 Bug 1709: When built with TLS support, the tls_advertise_hosts option now
720 defaults to "*" (all hosts). The variable is now available when not built
721 with TLS, default unset, mainly to enable keeping the testsuite sane.
722 If a server certificate is not supplied (via tls_certificate) an error is
723 logged, and clients will find TLS connections fail on startup. Presumably
724 they will retry in-clear.
725 Packagers of Exim are strongly encouraged to create a server certificate
726 at installation time.
728 HS/03 Add -bP config_file as a synonym for -bP configure_file, for consistency
729 with the $config_file variable.
731 JH/19 Two additional event types: msg:rcpt:defer and msg:rcpt:host:defer. Both
732 in transport context, after the attempt, and per-recipient. The latter type
733 is per host attempted. The event data is the error message, and the errno
734 information encodes the lookup type (A vs. MX) used for the (first) host,
735 and the trailing two digits of the smtp 4xx response.
737 GF/01 Bug 1715: Fix for race condition in exicyclog, where exim could attempt
738 to write to mainlog (or rejectlog, paniclog) in the window between file
739 creation and permissions/ownership being changed. Particularly affects
740 installations where exicyclog is run as root, rather than exim user;
741 result is that the running daemon panics and dies.
743 JH/20 Bug 1701: For MySQL lookups, support MySQL config file option group names.
745 JH/21 Bug 1720: Add support for priority groups and weighted-random proxy
746 selection for the EXPERIMENTAL_SOCKS feature, via new per-proxy options
747 "pri" and "weight". Note that the previous implicit priority given by the
748 list order is no longer honoured.
750 JH/22 Bugs 963, 1721: Fix some corner cases in message body canonicalization
753 JH/23 Move SOCKS5 support from Experimental to mainline, enabled for a build
754 by defining SUPPORT_SOCKS.
756 JH/26 Move PROXY support from Experimental to mainline, enabled for a build
757 by defining SUPPORT_PROXY. Note that the proxy_required_hosts option
758 is renamed to hosts_proxy, and the proxy_{host,target}_{address,port}.
759 variables are renamed to proxy_{local,external}_{address,port}.
761 JH/27 Move Internationalisation support from Experimental to mainline, enabled
762 for a build by defining SUPPORT_I18N
764 JH/28 Bug 1745: Fix redis lookups to handle (quoted) spaces embedded in parts
765 of the query string, and make ${quote_redis:} do that quoting.
767 JH/29 Move Events support from Experimental to mainline, enabled by default
768 and removable for a build by defining DISABLE_EVENT.
770 JH/30 Updated DANE implementation code to current from Viktor Dukhovni.
772 JH/31 Fix bug with hosts_connection_nolog and named-lists which were wrongly
773 cached by the daemon.
775 JH/32 Move Redis support from Experimental to mainline, enabled for a build
776 by defining LOOKUP_REDIS. The libhiredis library is required.
778 JH/33 Bug 1748: Permit ACL dnslists= condition in non-smtp ACLs if explicit
779 keys are given for lookup.
781 JH/34 Bug 1192: replace the embedded copy of PolarSSL RSA routines in the DKIM
782 support, by using OpenSSL or GnuTLS library ones. This means DKIM is
783 only supported when built with TLS support. The PolarSSL SHA routines
784 are still used when the TLS library is too old for convenient support.
786 JH/35 Require SINGLE_DH_USE by default in OpenSSL (main config option
787 openssl_options), for security. OpenSSL forces this from version 1.1.0
788 server-side so match that on older versions.
790 JH/36 Bug 1778: longstanding bug in memory use by the ${run } expansion: A fresh
791 allocation for $value could be released as the expansion processing
792 concluded, but leaving the global pointer active for it.
794 JH/37 Bug 1769: Permit a VRFY ACL to override the default 252 response,
795 and to use the domains and local_parts ACL conditions.
797 JH/38 Fix cutthrough bug with body lines having a single dot. The dot was
798 incorrectly not doubled on cutthrough transmission, hence seen as a
799 body-termination at the receiving system - resulting in truncated mails.
800 Commonly the sender saw a TCP-level error, and retransmitted the message
801 via the normal store-and-forward channel. This could result in duplicates
802 received - but deduplicating mailstores were liable to retain only the
803 initial truncated version.
805 JH/39 Bug 1781: Fix use of DKIM private-keys having trailing '=' in the base-64.
807 JH/40 Fix crash in queryprogram router when compiled with EXPERIMENTAL_SRS.
809 JH/41 Bug 1792: Fix selection of headers to sign for DKIM: bottom-up. While
810 we're in there, support oversigning also; bug 1309.
812 JH/42 Bug 1796: Fix error logged on a malware scanner connection failure.
814 HS/04 Add support for keep_environment and add_environment options.
816 JH/43 Tidy coding issues detected by gcc --fsanitize=undefined. Some remain;
817 either intentional arithmetic overflow during PRNG, or testing config-
820 JH/44 Bug 1800: The combination of a -bhc commandline option and cutthrough
821 delivery resulted in actual delivery. Cancel cutthrough before DATA
824 JH/45 Fix cutthrough, when connection not opened by verify and target hard-
825 rejects a recipient: pass the reject to the originator.
827 JH/46 Multiple issues raised by Coverity. Some were obvious or plausible bugs.
828 Many were false-positives and ignorable, but it's worth fixing the
831 JH/47 Fix build on HP-UX and older Solaris, which need (un)setenv now also
832 for the new environment-manipulation done at startup. Move the routines
833 from being local to tls.c to being global via the os.c file.
835 JH/48 Bug 1807: Fix ${extract } for the numeric/3-string case. While preparsing
836 an extract embedded as result-arg for a map, the first arg for extract
837 is unavailable so we cannot tell if this is a numbered or keyed
838 extraction. Accept either.
844 JH/01 Bug 1545: The smtp transport option "retry_include_ip_address" is now
847 JH/02 The smtp transport option "multi_domain" is now expanded.
849 JH/03 The smtp transport now requests PRDR by default, if the server offers
852 JH/04 Certificate name checking on server certificates, when exim is a client,
853 is now done by default. The transport option tls_verify_cert_hostnames
854 can be used to disable this per-host. The build option
855 EXPERIMENTAL_CERTNAMES is withdrawn.
857 JH/05 The value of the tls_verify_certificates smtp transport and main options
858 default to the word "system" to access the system default CA bundle.
859 For GnuTLS, only version 3.0.20 or later.
861 JH/06 Verification of the server certificate for a TLS connection is now tried
862 (but not required) by default. The verification status is now logged by
863 default, for both outbound TLS and client-certificate supplying inbound
866 JH/07 Changed the default rfc1413 lookup settings to disable calls. Few
869 JH/08 The EXPERIMENTAL_DSN compile option is no longer needed; all Delivery
870 Status Notification (bounce) messages are now MIME format per RFC 3464.
871 Support for RFC 3461 DSN options NOTIFY,ENVID,RET,ORCPT can be advertised
872 under the control of the dsn_advertise_hosts option, and routers may
873 have a dsn_lasthop option.
875 JH/09 A timeout of 2 minutes is now applied to all malware scanner types by
876 default, modifiable by a malware= option. The list separator for
877 the options can now be changed in the usual way. Bug 68.
879 JH/10 The smtp_receive_timeout main option is now expanded before use.
881 JH/11 The incoming_interface log option now also enables logging of the
882 local interface on delivery outgoing connections.
884 JH/12 The cutthrough-routing facility now supports multi-recipient mails,
885 if the interface and destination host and port all match.
887 JH/13 Bug 344: The verify = reverse_host_lookup ACL condition now accepts a
890 JH/14 Bug 1573: The spam= ACL condition now additionally supports Rspamd.
891 Patch from Andrew Lewis.
893 JH/15 Bug 670: The spamd_address main option (for the spam= ACL condition)
894 now supports optional time-restrictions, weighting, and priority
895 modifiers per server. Patch originally by <rommer@active.by>.
897 JH/16 The spamd_address main option now supports a mixed list of local
898 and remote servers. Remote servers can be IPv6 addresses, and
899 specify a port-range.
901 JH/17 Bug 68: The spamd_address main option now supports an optional
902 timeout value per server.
904 JH/18 Bug 1581: Router and transport options headers_add/remove can
905 now have the list separator specified.
907 JH/19 Bug 392: spamd_address, and clamd av_scanner, now support retry
910 JH/20 Bug 1571: Ensure that $tls_in_peerdn is set, when verification fails
913 JH/21 Support for the A6 type of dns record is withdrawn.
915 JH/22 Bug 608: The result of a QUIT or not-QUIT toplevel ACL now matters
916 rather than the verbs used.
918 JH/23 Bug 1572: Increase limit on SMTP confirmation message copy size
919 from 255 to 1024 chars.
921 JH/24 Verification callouts now attempt to use TLS by default.
923 HS/01 DNSSEC options (dnssec_require_domains, dnssec_request_domains)
924 are generic router options now. The defaults didn't change.
926 JH/25 Bug 466: Add RFC2322 support for MIME attachment filenames.
927 Original patch from Alexander Shikoff, worked over by JH.
929 HS/02 Bug 1575: exigrep falls back to autodetection of compressed
930 files if ZCAT_COMMAND is not executable.
932 JH/26 Bug 1539: Add timeout/retry options on dnsdb lookups.
934 JH/27 Bug 286: Support SOA lookup in dnsdb lookups.
936 JH/28 Bug 1588: Do not use the A lookup following an AAAA for setting the FQDN.
937 Normally benign, it bites when the pair was led to by a CNAME;
938 modern usage is to not canonicalize the domain to a CNAME target
939 (and we were inconsistent anyway for A-only vs AAAA+A).
941 JH/29 Bug 1632: Removed the word "rejected" from line logged for ACL discards.
943 JH/30 Check the forward DNS lookup for DNSSEC, in addition to the reverse,
944 when evaluating $sender_host_dnssec.
946 JH/31 Check the HELO verification lookup for DNSSEC, adding new
947 $sender_helo_dnssec variable.
949 JH/32 Bug 1397: Enable ECDHE on OpenSSL, just the NIST P-256 curve.
951 JH/33 Bug 1346: Note MAIL cmd seen in -bS batch, to avoid smtp_no_mail log.
953 JH/34 Bug 1648: Fix a memory leak seen with "mailq" and large queues.
955 JH/35 Bug 1642: Fix support of $spam_ variables at delivery time. Was
956 documented as working, but never had. Support all but $spam_report.
958 JH/36 Bug 1659: Guard checking of input smtp commands again pseudo-command
959 added for tls authenticator.
961 HS/03 Add perl_taintmode main config option
967 TL/01 When running the test suite, the README says that variables such as
968 no_msglog_check are global and can be placed anywhere in a specific
969 test's script, however it was observed that placement needed to be near
970 the beginning for it to behave that way. Changed the runtest perl
971 script to read through the entire script once to detect and set these
972 variables, reset to the beginning of the script, and then run through
973 the script parsing/test process like normal.
975 TL/02 The BSD's have an arc4random API. One of the functions to induce
976 adding randomness was arc4random_stir(), but it has been removed in
977 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
978 function when detected.
980 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
981 cause callback expansion.
983 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
984 syntax errors in an expansion can be treated as a string instead of
985 logging or causing an error, due to the internal use of bool_lax
986 instead of bool when processing it.
988 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
989 server certificates when making smtp deliveries.
991 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
993 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
995 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
997 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
998 Merged patch from Sebastian Wiedenroth.
1000 JH/05 Fix results-pipe from transport process. Several recipients, combined
1001 with certificate use, exposed issues where response data items split
1002 over buffer boundaries were not parsed properly. This eventually
1003 resulted in duplicates being sent. This issue only became common enough
1004 to notice due to the introduction of connection certificate information,
1005 the item size being so much larger. Found and fixed by Wolfgang Breyha.
1007 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
1008 size buffer was used, resulting in syntax errors when an expansion
1011 JH/07 Add support for directories of certificates when compiled with a GnuTLS
1012 version 3.3.6 or later.
1014 JH/08 Rename the TPDA experimental facility to Event Actions. The #ifdef
1015 is EXPERIMENTAL_EVENT, the main-configuration and transport options
1016 both become "event_action", the variables become $event_name, $event_data
1017 and $event_defer_errno. There is a new variable $verify_mode, usable in
1018 routers, transports and related events. The tls:cert event is now also
1019 raised for inbound connections, if the main configuration event_action
1022 TL/06 In test suite, disable OCSP for old versions of openssl which contained
1023 early OCSP support, but no stapling (appears to be less than 1.0.0).
1025 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
1026 server certificate names available under the smtp transport option
1027 "tls_verify_cert_hostname" now do not permit multi-component wildcard
1030 JH/10 Time-related extraction expansions from certificates now use the main
1031 option "timezone" setting for output formatting, and are consistent
1032 between OpenSSL and GnuTLS compilations. Bug 1541.
1034 JH/11 Fix a crash in mime ACL when meeting a zero-length, quoted or RFC2047-
1035 encoded parameter in the incoming message. Bug 1558.
1037 JH/12 Bug 1527: Autogrow buffer used in reading spool files. Since they now
1038 include certificate info, eximon was claiming there were spoolfile
1041 JH/13 Bug 1521: Fix ldap lookup for single-attr request, multiple-attr return.
1043 JH/14 Log delivery-related information more consistently, using the sequence
1044 "H=<name> [<ip>]" wherever possible.
1046 TL/07 Bug 1547: Omit RFCs from release. Draft and RFCs have licenses which
1047 are problematic for Debian distribution, omit them from the release
1050 JH/15 Updates and fixes to the EXPERIMENTAL_DSN feature.
1052 JH/16 Fix string representation of time values on 64bit time_t architectures.
1055 JH/17 Fix a null-indirection in certextract expansions when a nondefault
1056 output list separator was used.
1061 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
1062 checkers that were complaining about end of non-void function with no
1065 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
1066 This was a regression introduced in 4.83 by another bugfix.
1068 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
1070 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
1071 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
1077 TF/01 Correctly close the server side of TLS when forking for delivery.
1079 When a message was received over SMTP with TLS, Exim failed to clear up
1080 the incoming connection properly after forking off the child process to
1081 deliver the message. In some situations the subsequent outgoing
1082 delivery connection happened to have the same fd number as the incoming
1083 connection previously had. Exim would try to use TLS and fail, logging
1084 a "Bad file descriptor" error.
1086 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
1087 utilities have not been installed.
1089 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
1090 temporary space as the ACL may create new global variables.
1092 TL/01 LDAP support uses per connection or global context settings, depending
1093 upon the detected version of the libraries at build time.
1095 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
1096 to extract and use the src ip:port in logging and expansions as if it
1097 were a direct connection from the outside internet. PPv2 support was
1098 updated based on HAProxy spec change in May 2014.
1100 JH/02 Add ${listextract {number}{list}{success}{fail}}.
1102 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
1103 Properly escape header and check for NULL return.
1105 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
1108 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
1110 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
1111 characters in header names, implemented as a verify condition.
1112 Contributed by Michael Fischer v. Mollard.
1114 TL/05 Rename SPF condition results err_perm and err_temp to standardized
1115 results permerror and temperror. Previous values are deprecated but
1116 still accepted. In a future release, err_perm and err_temp will be
1117 completely removed, which will be a backward incompatibility if the
1118 ACL tests for either of these two old results. Patch contributed by
1119 user bes-internal on the mailing list.
1121 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
1123 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
1124 selectors, in both main and reject logs.
1126 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
1129 JH/07 Add malware type "sock" for talking to simple daemon.
1131 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
1133 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
1134 routers/transports under cutthrough routing.
1136 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
1137 numbers. Touch up "bool" conditional to keep the same definition.
1139 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
1141 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
1143 JH/12 Expand items in router/transport headers_add or headers_remove lists
1144 individually rather than the list as a whole. Bug 1452.
1146 Required for reasonable handling of multiple headers_ options when
1147 they may be empty; requires that headers_remove items with embedded
1148 colons must have them doubled (or the list-separator changed).
1150 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
1151 view the policy declared in the DMARC record. Currently, $dmarc_status
1152 is a combined value of both the record presence and the result of the
1155 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
1157 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
1158 dnslookup router and the smtp transport (applying to the forward
1161 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
1162 of ldap servers used for a specific lookup. Patch provided by Heiko
1165 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
1166 New variable $lookup_dnssec_authenticated for observability.
1168 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
1169 Patch submitted by Lars Timman.
1171 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
1173 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
1174 Requires trusted mode and valid format message id, aborts otherwise.
1175 Patch contributed by Heiko Schlichting.
1177 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
1178 certextract with support for various fields. Bug 1358.
1180 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
1181 is requested by default, modifiable by smtp transport option
1184 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
1185 operate on certificate variables to give certificate fingerprints
1186 Also new ${sha256:cert_variable}.
1188 JH/23 The PRDR feature is moved from being Experimental into the mainline.
1190 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
1191 Christian Aistleitner.
1193 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
1195 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
1196 file. Patch from Wolfgang Breyha.
1198 JH/25 Expand the coverage of the delivery $host and $host_address to
1199 client authenticators run in verify callout. Bug 1476.
1201 JH/26 Port service names are now accepted for tls_on_connect_ports, to
1202 align with daemon_smtp_ports. Bug 72.
1204 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
1205 support and error reporting did not work properly.
1207 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
1208 and is readable. Patch from Andrew Colin Kissa.
1210 TL/14 Enhance documentation of ${run expansion and how it parses the
1211 commandline after expansion, particularly in the case when an
1212 unquoted variable expansion results in an empty value.
1214 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
1216 PP/02 Fix internal collision of T_APL on systems which support RFC3123
1217 by renaming away from it. Addresses GH issue 15, reported by
1220 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
1222 TL/15 SECURITY: prevent double expansion in math comparison functions
1223 (can expand unsanitized data). Not remotely exploitable.
1230 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
1232 PP/02 Make -n do something, by making it not do something.
1233 When combined with -bP, the name of an option is not output.
1235 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
1238 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
1239 $sender_host_name and config options to manage this, and basic check
1242 PP/05 DSCP support for outbound connections and control modifier for inbound.
1244 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
1245 (Only plugin which currently uses this is kerberos4, which nobody should
1246 be using, but we should make it available and other future plugins might
1247 conceivably use it, even though it would break NAT; stuff *should* be
1248 using channel bindings instead).
1250 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
1251 name; added for Sendmail compatibility; requires admin caller.
1252 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
1253 just ignore it); requires trusted caller.
1254 Also parse but ignore: -Ac -Am -X<logfile>
1257 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
1259 TL/02 Add +smtp_confirmation as a default logging option.
1261 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
1262 Patch by Magnus Holmgren from 2007-02-20.
1264 TL/04 Bugzilla 1281 - Spec typo.
1265 Bugzilla 1283 - Spec typo.
1266 Bugzilla 1290 - Spec grammar fixes.
1268 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
1270 TL/06 Add Experimental DMARC support using libopendmarc libraries.
1272 TL/07 Fix an out of order global option causing a segfault. Reported to dev
1273 mailing list by by Dmitry Isaikin.
1275 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
1277 JH/02 Support "G" suffix to numbers in ${if comparisons.
1279 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
1281 NM/01 Bugzilla 1197 - Spec typo
1282 Bugzilla 1196 - Spec examples corrections
1284 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
1286 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
1287 gnutls_enable_pkcs11, but renamed to more accurately indicate its
1290 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
1291 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
1293 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
1294 "acl {{name}{arg}...}", and optional args on acl condition
1297 JH/05 Permit multiple router/transport headers_add/remove lines.
1299 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
1301 JH/07 Avoid using a waiting database for a single-message-only transport.
1302 Performance patch from Paul Fisher. Bugzilla 1262.
1304 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
1307 JH/09 Add $headers_added variable, with content from use of ACL modifier
1308 add_header (but not yet added to the message). Bugzilla 199.
1310 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
1311 Pulled from Bugzilla 817 by Wolfgang Breyha.
1313 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
1315 (nb: this is the same fix as in Exim 4.80.1)
1317 JH/11 Add A= logging on delivery lines, and a client_set_id option on
1320 JH/12 Add optional authenticated_sender logging to A= and a log_selector
1323 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
1325 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
1326 advertise SMTP AUTH mechanism to us, instead of a generic
1327 protocol violation error. Also, make Exim more robust to bad
1328 data from the Dovecot auth socket.
1330 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
1332 When a queue runner is handling a message, Exim first routes the
1333 recipient addresses, during which it prunes them based on the retry
1334 hints database. After that it attempts to deliver the message to
1335 any remaining recipients. It then updates the hints database using
1338 So if a recipient address works intermittently, it can get repeatedly
1339 deferred at routing time. The retry hints record remains fresh so the
1340 address never reaches the final cutoff time.
1342 This is a fairly common occurrence when a user is bumping up against
1343 their storage quota. Exim had some logic in its local delivery code
1344 to deal with this. However it did not apply to per-recipient defers
1345 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
1347 This change adds a proper retry rule check during routing so that the
1348 final cutoff time is checked against the message's age. We only do
1349 this check if there is an address retry record and there is not a
1350 domain retry record; this implies that previous attempts to handle
1351 the address had the retry_use_local_parts option turned on. We use
1352 this as an approximation for the destination being like a local
1353 delivery, as in LMTP.
1355 I suspect this new check makes the old local delivery cutoff check
1356 redundant, but I have not verified this so I left the code in place.
1358 TF/02 Correct gecos expansion when From: is a prefix of the username.
1360 Test 0254 submits a message to Exim with the header
1364 When I ran the test suite under the user fanf2, Exim expanded
1365 the header to contain my full name, whereas it should have added
1366 a Resent-Sender: header. It erroneously treats any prefix of the
1367 username as equal to the username.
1369 This change corrects that bug.
1371 GF/01 DCC debug and logging tidyup
1372 Error conditions log to paniclog rather than rejectlog.
1373 Debug lines prefixed by "DCC: " to remove any ambiguity.
1375 TF/03 Avoid unnecessary rebuilds of lookup-related code.
1377 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
1378 Bug spotted by Jeremy Harris; was flawed since initial commit.
1379 Would have resulted in OCSP responses post-SNI triggering an Exim
1380 NULL dereference and crash.
1382 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
1384 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
1385 Bug detection, analysis and fix by Samuel Thibault.
1386 Bugzilla 1331, Debian bug #698092.
1388 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
1390 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
1391 Server implementation by Todd Lyons, client by JH.
1392 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
1393 config variable "prdr_enable" controls whether the server
1394 advertises the facility. If the client requests PRDR a new
1395 acl_data_smtp_prdr ACL is called once for each recipient, after
1396 the body content is received and before the acl_smtp_data ACL.
1397 The client is controlled by both of: a hosts_try_prdr option
1398 on the smtp transport, and the server advertisement.
1399 Default client logging of deliveries and rejections involving
1400 PRDR are flagged with the string "PRDR".
1402 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
1403 fclose(). Diagnosis by Todd Lyons.
1405 PP/17 Update configure.default to handle IPv6 localhost better.
1406 Patch by Alain Williams (plus minor tweaks).
1409 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
1410 This is now consistent with GnuTLS, and is now documented: the
1411 previous undocumented portable approach to treating the option as
1412 unset was to force an expansion failure. That still works, and
1413 an empty string is now equivalent.
1415 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
1416 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
1417 not performing validation itself.
1419 PP/20 Added force_command boolean option to pipe transport.
1420 Patch from Nick Koston, of cPanel Inc.
1422 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
1425 TF/04 Added udpsend ACL modifier and hexquote expansion operator
1427 PP/21 Fix eximon continuous updating with timestamped log-files.
1428 Broken in a format-string cleanup in 4.80, missed when I repaired the
1429 other false fix of the same issue.
1430 Report and fix from Heiko Schlichting.
1433 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
1434 Report from Prashanth Katuri.
1436 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
1437 It's SecureTransport, so affects any MacOS clients which use the
1438 system-integrated TLS libraries, including email clients.
1440 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
1441 using a MIME ACL for non-SMTP local injection.
1442 Report and assistance in diagnosis by Warren Baker.
1444 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
1446 JH/16 Fix comparisons for 64b. Bugzilla 1385.
1448 TL/09 Add expansion variable $authenticated_fail_id to keep track of
1449 last id that failed so it may be referenced in subsequent ACL's.
1451 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
1454 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
1455 ldap library initialization, allowing self-signed CA's to be
1456 used. Also properly sets require_cert option later in code by
1457 using NULL (global ldap config) instead of ldap handle (per
1458 session). Bug diagnosis and testing by alxgomz.
1460 TL/12 Enhanced documentation in the ratelimit.pl script provided in
1461 the src/util/ subdirectory.
1463 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
1464 renamed to Transport Post Delivery Action by Jeremy Harris, as
1467 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
1468 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
1469 redis_servers = needs to be configured which will be used by the redis
1470 lookup. Patch from Warren Baker, of The Packet Hub.
1472 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
1474 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
1475 hostname or reverse DNS when processing a host list. Used suggestions
1476 from multiple comments on this bug.
1478 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
1480 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
1481 Missed a few lines, added it to make the runtest require no keyboard
1484 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
1485 contains upper case chars. Make router use caseful_local_part.
1487 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
1488 support when GnuTLS has been built with p11-kit.
1494 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
1496 This, or similar/improved, will also be change PP/11 of 4.82.
1502 PP/01 Handle short writes when writing local log-files.
1503 In practice, only affects FreeBSD (8 onwards).
1504 Bugzilla 1053, with thanks to Dmitry Isaikin.
1506 NM/01 Bugzilla 949 - Documentation tweak
1508 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
1511 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
1513 PP/02 Implemented gsasl authenticator.
1515 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
1517 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
1518 `pkg-config foo` for cflags/libs.
1520 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
1521 with rest of GSASL and with heimdal_gssapi.
1523 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
1524 `pkg-config foo` for cflags/libs for the TLS implementation.
1526 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
1527 properties get this fed in as external SSF. A number of robustness
1528 and debugging improvements to the cyrus_sasl authenticator.
1530 PP/08 cyrus_sasl server now expands the server_realm option.
1532 PP/09 Bugzilla 1214 - Log authentication information in reject log.
1533 Patch by Jeremy Harris.
1535 PP/10 Added dbmjz lookup type.
1537 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
1539 PP/12 MAIL args handles TAB as well as SP, for better interop with
1540 non-compliant senders.
1541 Analysis and variant patch by Todd Lyons.
1543 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
1544 Bug report from Lars Müller <lars@samba.org> (via SUSE),
1545 Patch from Dirk Mueller <dmueller@suse.com>
1547 PP/13 tls_peerdn now print-escaped for spool files.
1548 Observed some $tls_peerdn in wild which contained \n, which resulted
1549 in spool file corruption.
1551 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
1552 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
1553 or write after TLS renegotiation, which otherwise led to messages
1556 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
1557 as a tracking header (ie: a signed header comes before the signature).
1558 Patch from Wolfgang Breyha.
1560 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
1561 comma-sep list; embedded commas doubled.
1563 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
1565 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
1567 Report and patch from Dmitry Banschikov.
1569 PP/16 Removed "dont_insert_empty_fragments" from "openssl_options".
1570 Removed SSL_clear() after SSL_new() which led to protocol negotiation
1571 failures. We appear to now support TLS1.1+ with Exim.
1573 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
1574 lets Exim select keys and certificates based upon TLS SNI from client.
1575 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
1576 before an outbound SMTP session. New log_selector, +tls_sni.
1578 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
1579 NULL dereference. Report and patch from Alun Jones.
1581 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
1582 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
1583 Not seeing resolver debug output on NetBSD, but suspect this is a
1584 resolver implementation change.
1586 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
1587 Left warnings. Added "eximon gdb" invocation mode.
1589 PP/21 Defaulting "accept_8bitmime" to true, not false.
1591 PP/22 Added -bw for inetd wait mode support.
1593 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
1594 locate the relevant includes and libraries. Made this the default.
1596 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
1597 Bugzilla 1246, report and most of solution from Tomasz Kusy.
1599 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
1600 This may cause build issues on older platforms.
1602 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
1603 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
1604 gnutls_require_mac & gnutls_require_protocols (no longer supported).
1605 Added SNI support via GnuTLS too.
1606 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
1608 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
1610 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
1611 Applied second patch from Janne, implementing suggestion to default
1612 multiple-strings-in-record handling to match SPF spec.
1614 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
1616 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
1617 read-only, out of scope).
1618 Patch from Wolfgang Breyha, report from Stuart Northfield.
1620 PP/29 Fix three issues highlighted by clang analyser static analysis.
1621 Only crash-plausible issue would require the Cambridge-specific
1622 iplookup router and a misconfiguration.
1623 Report from Marcin Mirosław.
1625 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
1627 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
1628 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
1629 As part of this, removing so much warning spew let me fix some minor
1630 real issues in debug logging.
1632 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
1633 assignment on my part. Fixed.
1635 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
1636 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
1637 Janne Snabb (who went above and beyond: thank you).
1639 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
1640 string otherwise requires a connection and a bunch more work and it's
1641 relatively easy to get wrong. Should also expose TLS library linkage
1644 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
1645 64-bit ${eval} (JH/03).
1647 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
1648 GNU libc to support some of the 64-bit stuff, should not lead to
1649 conflicts. Defined before os.h is pulled in, so if a given platform
1650 needs to override this, it can.
1652 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
1653 protection layer was required, which is not implemented.
1654 Bugzilla 1254, patch from Wolfgang Breyha.
1656 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
1657 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
1658 tls_dhparam take prime identifiers. Also unbreak combination of
1659 OpenSSL+DH_params+TLSSNI.
1661 PP/39 Disable SSLv2 by default in OpenSSL support.
1667 PP/01 Solaris build fix for Oracle's LDAP libraries.
1668 Bugzilla 1109, patch from Stephen Usher.
1670 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
1672 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
1675 TF/02 Fix a couple more cases where we did not log the error message
1676 when unlink() failed. See also change 4.74-TF/03.
1678 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
1679 lock up or crash if it happened to be inside a call to libc when it
1680 got a SIGUSR1 from exiwhat.
1682 The SIGUSR1 handler appends the current process status to the process
1683 log which is later printed by exiwhat. It used to use the general
1684 purpose logging code to do this, but several functions it calls are
1685 not safe for signals.
1687 The new output code in the SIGUSR1 handler is specific to the process
1688 log, and simple enough that it's easy to inspect for signal safety.
1689 Removing some special cases also simplifies the general logging code.
1690 Removing the spurious timestamps from the process log simplifies
1693 TF/04 Improved ratelimit ACL condition.
1695 The /noupdate option has been deprecated in favour of /readonly which
1696 has clearer semantics. The /leaky, /strict, and /readonly update modes
1697 are mutually exclusive. The update mode is no longer included in the
1698 database key; it just determines when the database is updated. (This
1699 means that when you upgrade Exim will forget old rate measurements.)
1701 Exim now checks that the per_* options are used with an update mode that
1702 makes sense for the current ACL. For example, when Exim is processing a
1703 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
1704 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
1705 must specify per_mail/readonly. If you omit the update mode it defaults to
1706 /leaky where that makes sense (as before) or /readonly where required.
1708 The /noupdate option is now undocumented but still supported for
1709 backwards compatibility. It is equivalent to /readonly except that in
1710 ACLs where /readonly is required you may specify /leaky/noupdate or
1711 /strict/noupdate which are treated the same as /readonly.
1713 A useful new feature is the /count= option. This is a generalization
1714 of the per_byte option, so that you can measure the throughput of other
1715 aggregate values. For example, the per_byte option is now equivalent
1716 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
1718 The per_rcpt option has been generalized using the /count= mechanism
1719 (though it's more complicated than the per_byte equivalence). When it is
1720 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
1721 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
1722 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
1723 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
1724 Note that using per_rcpt with a non-readonly update mode in more than
1725 one ACL will cause the recipients to be double-counted. (The per_mail
1726 and per_byte options don't have this problem.)
1728 The handling of very low rates has changed slightly. If the computed rate
1729 is less than the event's count (usually one) then this event is the first
1730 after a long gap. In this case the rate is set to the same as this event's
1731 count, so that the first message of a spam run is counted properly.
1733 The major new feature is a mechanism for counting the rate of unique
1734 events. The new per_addr option counts the number of different
1735 recipients that someone has sent messages to in the last time period. It
1736 behaves like per_rcpt if all the recipient addresses are different, but
1737 duplicate recipient addresses do not increase the measured rate. Like
1738 the /count= option this is a general mechanism, so the per_addr option
1739 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
1740 example, measure the rate that a client uses different sender addresses
1741 with the options per_mail/unique=$sender_address. There are further
1742 details in the main documentation.
1744 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
1746 TF/06 Removed a few PCRE remnants.
1748 TF/07 Automatically extract Exim's version number from tags in the git
1749 repository when doing development or release builds.
1751 PP/02 Raise smtp_cmd_buffer_size to 16kB.
1752 Bugzilla 879. Patch from Paul Fisher.
1754 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
1755 Heavily based on revision 40f9a89a from Simon Arlott's tree.
1758 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
1760 PP/05 Variable $av_failed, true if the AV scanner deferred.
1761 Bugzilla 1078. Patch from John Horne.
1763 PP/06 Stop make process more reliably on build failure.
1764 Bugzilla 1087. Patch from Heiko Schlittermann.
1766 PP/07 Make maildir_use_size_file an _expandable_ boolean.
1767 Bugzilla 1089. Patch from Heiko Schlittermann.
1769 PP/08 Handle ${run} returning more data than OS pipe buffer size.
1770 Bugzilla 1131. Patch from Holger Weiß.
1772 PP/09 Handle IPv6 addresses with SPF.
1773 Bugzilla 860. Patch from Wolfgang Breyha.
1775 PP/10 GnuTLS: support TLS 1.2 & 1.1.
1777 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
1780 PP/11 match_* no longer expand right-hand-side by default.
1781 New compile-time build option, EXPAND_LISTMATCH_RHS.
1782 New expansion conditions, "inlist", "inlisti".
1784 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
1786 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
1788 PP/14 fix log_write() format string regression from TF/03.
1789 Bugzilla 1152. Patch from Dmitry Isaikin.
1795 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
1797 PP/02 Harmonised TLS library version reporting; only show if debugging.
1798 Layout now matches that introduced for other libraries in 4.74 PP/03.
1800 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
1802 PP/04 New "dns_use_edns0" global option.
1804 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
1807 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
1808 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
1810 TK/01 Updated PolarSSL code to 0.14.2.
1811 Bugzilla 1097. Patch from Andreas Metzler.
1813 PP/07 Catch divide-by-zero in ${eval:...}.
1814 Fixes bugzilla 1102.
1816 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
1819 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
1820 format-string attack -- SECURITY: remote arbitrary code execution.
1822 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
1823 time unintentionally subject to list matching rules, letting the header
1824 cause arbitrary Exim lookups (of items which can occur in lists, *not*
1825 arbitrary string expansion). This allowed for information disclosure.
1827 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
1828 INT_MIN/-1 -- value coerced to INT_MAX.
1834 NM/01 Workaround for PCRE version dependency in version reporting
1837 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
1838 This fixes portability to compilers other than gcc, notably
1839 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
1841 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
1842 makefiles for portability to HP-UX and POSIX correctness.
1844 PP/01 Permit LOOKUP_foo enabling on the make command-line.
1845 Also via indented variable definition in the Makefile.
1846 (Debugging by Oliver Heesakkers).
1848 PP/02 Restore caching of spamd results with expanded spamd_address.
1849 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
1851 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
1852 Improves build reliability. Fix from: Frank Elsner
1854 NM/02 Fix wide character breakage in the rfc2047 coding
1855 Fixes bug 1064. Patch from Andrey N. Oktyabrski
1857 NM/03 Allow underscore in dnslist lookups
1858 Fixes bug 1026. Patch from Graeme Fowler
1860 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
1861 Code patches from Adam Ciarcinski of NetBSD.
1863 NM/04 Fixed exiqgrep to cope with mailq missing size issue
1866 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
1867 is logged, to avoid truncation. Patch from John Horne.
1869 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
1870 Patch from Jakob Hirsch.
1872 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
1873 SQL string expansion failure details.
1874 Patch from Andrey Oktyabrski.
1876 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
1877 Patch from Simon Arlott.
1879 PP/09 New lookups functionality failed to compile on old gcc which rejects
1880 extern declarations in function scope.
1881 Patch from Oliver Fleischmann
1883 PP/10 Use sig_atomic_t for flags set from signal handlers.
1884 Check getgroups() return and improve debugging.
1885 Fixed developed for diagnosis in bug 927 (which turned out to be
1888 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
1889 Patch from Mark Zealey.
1891 PP/12 Bugzilla 1056: Improved spamd server selection.
1892 Patch from Mark Zealey.
1894 PP/13 Bugzilla 1086: Deal with maildir quota file races.
1895 Based on patch from Heiko Schlittermann.
1897 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
1898 Patch from Uwe Doering, sign-off by Michael Haardt.
1900 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
1901 variable declaration deep within a block. Bug and patch from
1904 PP/15 lookups-Makefile IRIX compatibility coercion.
1906 PP/16 Make DISABLE_DKIM build knob functional.
1908 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
1909 Patch by Simon Arlott
1911 TF/03 Fix valgrind.h portability to C89 compilers that do not support
1912 variable argument macros. Our copy now differs from upstream.
1918 TF/01 Failure to get a lock on a hints database can have serious
1919 consequences so log it to the panic log.
1921 TF/02 Log LMTP confirmation messages in the same way as SMTP,
1922 controlled using the smtp_confirmation log selector.
1924 TF/03 Include the error message when we fail to unlink a spool file.
1926 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
1927 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
1928 for maintaining out-of-tree patches for some time.
1930 PP/01 Bugzilla 139: Documentation and portability issues.
1931 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
1932 Handle per-OS dynamic-module compilation flags.
1934 PP/02 Let /dev/null have normal permissions.
1935 The 4.73 fixes were a little too stringent and complained about the
1936 permissions on /dev/null. Exempt it from some checks.
1937 Reported by Andreas M. Kirchwitz.
1939 PP/03 Report version information for many libraries, including
1940 Exim version information for dynamically loaded libraries. Created
1941 version.h, now support a version extension string for distributors
1942 who patch heavily. Dynamic module ABI change.
1944 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
1945 privilege escalation vulnerability whereby the Exim run-time user
1946 can cause root to append content of the attacker's choosing to
1949 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
1952 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
1953 If dropping privileges for untrusted macros, we disabled normal logging
1954 on the basis that it would fail; for the Exim run-time user, this is not
1955 the case, and it resulted in successful deliveries going unlogged.
1956 Fixed. Reported by Andreas Metzler.
1962 PP/01 Date: & Message-Id: revert to normally being appended to a message,
1963 only prepend for the Resent-* case. Fixes regression introduced in
1964 Exim 4.70 by NM/22 for Bugzilla 607.
1966 PP/02 Include check_rfc2047_length in configure.default because we're seeing
1967 increasing numbers of administrators be bitten by this.
1969 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
1971 PP/03 Bugzilla 994: added openssl_options main configuration option.
1973 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
1975 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
1977 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
1979 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
1980 without a peer certificate, leading to a segfault because of an
1981 assumption that peers always have certificates. Be a little more
1982 paranoid. Problem reported by Martin Tscholak.
1984 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
1985 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
1986 NB: ClamAV planning to remove STREAM in "middle of 2010".
1987 CL also introduces -bmalware, various -d+acl logging additions and
1988 more caution in buffer sizes.
1990 PP/09 Implemented reverse_ip expansion operator.
1992 PP/10 Bugzilla 937: provide a "debug" ACL control.
1994 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
1996 PP/12 Bugzilla 973: Implement --version.
1998 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
2000 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
2002 PP/15 Bugzilla 816: support multiple condition rules on Routers.
2004 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
2005 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
2006 ignore trailing whitespace.
2008 JJ/02 prevent non-panic DKIM error from being sent to paniclog
2010 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
2013 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
2014 Notification from Dr Andrew Aitchison.
2016 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
2017 ExtendedDetectionInfo response format.
2018 Notification from John Horne.
2020 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
2023 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
2024 XSL and documented dependency on system catalogs, with examples of how
2027 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
2030 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
2031 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
2032 configuration file which is writeable by the Exim user or group.
2034 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
2035 of configuration files to cover files specified with the -C option if
2036 they are going to be used with root privileges, not just the default
2039 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
2040 option (effectively making it always true).
2042 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
2043 files to be used while preserving root privileges.
2045 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
2046 that rogue child processes cannot use them.
2048 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
2049 run-time user, instead of root.
2051 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
2052 Exim run-time user without dropping privileges.
2054 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
2055 result string, instead of calling string_vformat() twice with the same
2058 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
2059 for other users. Others should always drop root privileges if they use
2060 -C on the command line, even for a whitelisted configure file.
2062 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
2064 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
2070 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
2071 $data_path, and $header_path variables; fixed documentation bugs and
2074 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
2075 exipick to access non-standard spools, including the "frozen" queue
2078 NM/01 Bugzilla 965: Support mysql stored procedures.
2079 Patch from Alain Williams
2081 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
2083 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
2084 Patch from Andreas Metzler
2086 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
2087 Patch from Kirill Miazine
2089 NM/05 Bugzilla 671: Added umask to procmail example.
2091 JJ/03 installed exipick 20100323.0, fixing doc bug
2093 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
2094 directory. Notification and patch from Dan Rosenberg.
2096 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
2098 TK/02 Improve log output when DKIM signing operation fails.
2100 MH/01 Treat the transport option dkim_domain as a colon separated
2101 list, not as a single string, and sign the message with each element,
2102 omitting multiple occurences of the same signer.
2104 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
2105 Bugzilla 985, 986. Patch by Simon Arlott
2107 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
2108 Patch by Simon Arlott
2110 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
2111 MBX locking. Notification from Dan Rosenberg.
2117 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
2119 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
2121 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
2123 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
2125 NM/04 Bugzilla 915: Flag broken perl installation during build.
2131 TK/01 Added patch by Johannes Berg that expands the main option
2132 "spamd_address" if it starts with a dollar sign.
2134 TK/02 Write list of recipients to X-Envelope-Sender header when building
2135 the mbox-format spool file for content scanning (suggested by Jakob
2138 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
2139 (http://www.dcc-servers.net/) support via dccifd. Activated by
2140 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
2142 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
2143 by Mark Daniel Reidel <mr@df.eu>.
2145 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
2146 When building exim an external PCRE library is now needed -
2147 PCRE is a system library on the majority of modern systems.
2148 See entry on PCRE_LIBS in EDITME file.
2150 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
2151 conversation. Added nologin parameter to request.
2152 Patch contributed by Kirill Miazine.
2154 TF/01 Do not log submission mode rewrites if they do not change the address.
2156 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
2158 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
2159 log files in place. Contributed by Roberto Lima.
2161 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
2163 TF/03 Bugzilla 615: When checking the local_parts router precondition
2164 after a local_part_suffix or local_part_prefix option, Exim now
2165 does not use the address's named list lookup cache, since this
2166 contains cached lookups for the whole local part.
2168 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
2169 Robert Millan. Documentation is in experimental-spec.txt.
2171 TF/04 Bugzilla 668: Fix parallel build (make -j).
2173 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
2175 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
2176 Patch provided by Jan Srzednicki.
2178 TF/05 Leading white space used to be stripped from $spam_report which
2179 wrecked the formatting. Now it is preserved.
2181 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
2182 that they are available at delivery time.
2184 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
2186 TF/08 TLS error reporting now respects the incoming_interface and
2187 incoming_port log selectors.
2189 TF/09 Produce a more useful error message if an SMTP transport's hosts
2190 setting expands to an empty string.
2192 NM/06 Bugzilla 744: EXPN did not work under TLS.
2193 Patch provided by Phil Pennock.
2195 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
2196 Patch provided by Richard Godbee.
2198 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
2199 acl_smtp_notquit, added index entry.
2201 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
2202 Patch provided by Eugene Bujak.
2204 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
2205 accept(). Patch provided by Maxim Dounin.
2207 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
2208 Patch provided by Phil Pennock.
2210 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
2212 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
2213 Patch provided by Brad "anomie" Jorsch.
2215 NM/14 Bugzilla 622: Added timeout setting to transport filter.
2216 Patch provided by Dean Brooks.
2218 TK/05 Add native DKIM support (does not depend on external libraries).
2220 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
2221 Patch provided by Graeme Fowler.
2223 NM/16 Bugzilla 851: Documentation example syntax fix.
2225 NM/17 Changed NOTICE file to remove references to embedded PCRE.
2227 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
2230 NM/19 Bugzilla 745: TLS version reporting.
2231 Patch provided by Phil Pennock.
2233 NM/20 Bugzilla 167: bool: condition support.
2234 Patch provided by Phil Pennock.
2236 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
2237 clients. Patch provided by Phil Pennock.
2239 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
2240 Patch provided by Brad "anomie" Jorsch.
2242 NM/23 Bugzilla 687: Fix misparses in eximstats.
2243 Patch provided by Heiko Schlittermann.
2245 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
2246 Patch provided by Heiko Schlittermann.
2248 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
2249 plus update to original patch.
2251 NM/26 Bugzilla 799: Documentation correction for ratelimit.
2253 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
2254 Patch provided by David Brownlee.
2256 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
2258 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
2260 NM/30 Bugzilla 888: TLS documentation bugfixes.
2262 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
2264 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
2265 Unlike the original bugzilla I have changed all shell scripts in src tree.
2267 NM/33 Bugzilla 898: Transport filter timeout fix.
2268 Patch by Todd Rinaldo.
2270 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
2271 Patch by Serge Demonchaux.
2273 NM/35 Bugzilla 39: Base64 decode bug fixes.
2274 Patch by Jakob Hirsch.
2276 NM/36 Bugzilla 909: Correct connect() call in dcc code.
2278 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
2280 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
2282 NM/39 Bugzilla 911: Fixed MakeLinks build script.
2288 TK/01 Add preliminary DKIM support. Currently requires a forked version of
2289 ALT-N's libdkim that I have put here:
2290 http://duncanthrax.net/exim-experimental/
2292 Note to Michael Haardt: I had to rename some vars in sieve.c. They
2293 were called 'true' and it seems that C99 defines that as a reserved
2294 keyword to be used with 'bool' variable types. That means you could
2295 not include C99-style headers which use bools without triggering
2296 build errors in sieve.c.
2298 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
2299 as mailq or other aliases. Changed the --help handling significantly
2300 to do whats expected. exim_usage() emits usage/help information.
2302 SC/01 Added the -bylocaldomain option to eximstats.
2304 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
2306 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
2308 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
2314 PH/01 Another patch from the Sieve maintainer.
2316 PH/02 When an IPv6 address is converted to a string for single-key lookup
2317 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
2318 dots are used instead of colons so that keys in lsearch files need not
2319 contain colons. This was done some time before quoting was made available
2320 in lsearch files. However, iplsearch files do require colons in IPv6 keys
2321 (notated using the quote facility) so as to distinguish them from IPv4
2322 keys. This meant that lookups for IP addresses in host lists did not work
2323 for iplsearch lookups.
2325 This has been fixed by arranging for IPv6 addresses to be expressed with
2326 colons if the lookup type is iplsearch. This is not incompatible, because
2327 previously such lookups could never work.
2329 The situation is now rather anomalous, since one *can* have colons in
2330 ordinary lsearch keys. However, making the change in all cases is
2331 incompatible and would probably break a number of configurations.
2333 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
2336 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
2337 conversion specification without a maximum field width, thereby enabling
2338 a rogue spamd server to cause a buffer overflow. While nobody in their
2339 right mind would setup Exim to query an untrusted spamd server, an
2340 attacker that gains access to a server running spamd could potentially
2341 exploit this vulnerability to run arbitrary code as the Exim user.
2343 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
2344 $primary_hostname instead of what libspf2 thinks the hosts name is.
2346 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
2347 a directory entry by the name of the lookup key. Previously, if a
2348 symlink pointed to a non-existing file or a file in a directory that
2349 Exim lacked permissions to read, a lookup for a key matching that
2350 symlink would fail. Now it is enough that a matching directory entry
2351 exists, symlink or not. (Bugzilla 503.)
2353 PH/03 The body_linecount and body_zerocount variables are now exported in the
2356 PH/04 Added the $dnslist_matched variable.
2358 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
2359 This means they are set thereafter only if the connection becomes
2362 PH/06 Added the client_condition to authenticators so that some can be skipped
2363 by clients under certain conditions.
2365 PH/07 The error message for a badly-placed control=no_multiline_responses left
2366 "_responses" off the end of the name.
2368 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
2370 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
2371 (without spaces) instead of just copying the configuration text.
2373 PH/10 Added the /noupdate option to the ratelimit ACL condition.
2375 PH/11 Added $max_received_linelength.
2377 PH/12 Added +ignore_defer and +include_defer to host lists.
2379 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
2380 way in which PCRE > 7.0 is built.
2382 PH/14 Implemented queue_only_load_latch.
2384 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
2385 MAIL command. The effect was to mangle the value on 64-bit systems.
2387 PH/16 Another patch from the Sieve maintainer.
2389 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
2391 PH/18 If a system quota error occurred while trying to create the file for
2392 a maildir delivery, the message "Mailbox is full" was not appended to the
2393 bounce if the delivery eventually timed out. Change 4.67/27 below applied
2394 only to a quota excession during the actual writing of the file.
2396 PH/19 It seems that peer DN values may contain newlines (and other non-printing
2397 characters?) which causes problems in log lines. The DN values are now
2398 passed through string_printing() before being added to log lines.
2400 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
2401 and InterBase are left for another time.)
2403 PH/21 Added message_body_newlines option.
2405 PH/22 Guard against possible overflow in moan_check_errorcopy().
2407 PH/23 POSIX allows open() to be a macro; guard against that.
2409 PH/24 If the recipient of an error message contained an @ in the local part
2410 (suitably quoted, of course), incorrect values were put in $domain and
2411 $local_part during the evaluation of errors_copy.
2417 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
2418 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
2421 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
2422 issue a MAIL command.
2424 PH/02 In an ACL statement such as
2426 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
2428 if a client was not listed at all, or was listed with a value other than
2429 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
2430 the condition was not true (as it should be), so access was not denied.
2431 The bug was that the ! inversion was incorrectly passed on to the second
2432 item. This has been fixed.
2434 PH/03 Added additional dnslists conditions == and =& which are different from
2435 = and & when the dns lookup returns more than one IP address.
2437 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
2438 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
2440 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
2441 FSYNC, which compiles an option called disable_fsync that allows for
2442 bypassing fsync(). The documentation is heavily laced with warnings.
2444 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
2446 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
2447 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
2448 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
2449 including adding "make clean"; (3) Added -fPIC when compiling the test
2450 dynamically loaded module, to get rid of a warning.
2452 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
2453 message fails, move_frozen_messages = true and ignore_bounce_errors_after
2454 = 0s. The bug is otherwise harmless.
2456 PH/07 There was a bug in the dovecot authenticator such that the value of
2457 $auth1 could be overwritten, and so not correctly preserved, after a
2458 successful authentication. This usually meant that the value preserved by
2459 the server_setid option was incorrect.
2461 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
2463 PH/09 Installed PCRE release 7.0.
2465 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
2466 run for batched SMTP input. It is now run at the start of every message
2467 in the batch. While fixing this I discovered that the process information
2468 (output by running exiwhat) was not always getting set for -bs and -bS
2469 input. This is fixed, and it now also says "batched" for BSMTP.
2471 PH/11 Added control=no_pipelining.
2473 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
2474 patch, slightly modified), and move the expansion of helo_data till after
2475 the connection is made in the smtp transport (so it can use these
2478 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
2480 PH/14 Added log_selector = +pid.
2482 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
2484 PH/16 Add ${if forany and ${if forall.
2486 PH/17 Added dsn_from option to vary the From: line in DSNs.
2488 PH/18 Flush SMTP output before performing a callout, unless control =
2489 no_callout_flush is set.
2491 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
2492 was true (the default) a successful delivery failed to delete the retry
2493 item, thus causing premature timeout of the address. The bug is now
2496 PH/20 Added hosts_avoid_pipelining to the smtp transport.
2498 PH/21 Long custom messages for fakedefer and fakereject are now split up
2499 into multiline responses in the same way that messages for "deny" and
2500 other ACL rejections are.
2502 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
2503 with slight modification.
2505 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
2506 draft, changing the syntax and factoring some duplicate code".
2508 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
2509 for deliveries of the second and subsequent messages over the same SMTP
2512 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
2513 ${reduce, with only minor "tidies".
2515 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
2517 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
2518 expansion side effects.
2520 PH/27 When a message times out after an over-quota error from an Exim-imposed
2521 quota, the bounce message says "mailbox is full". This message was not
2522 being given when it was a system quota that was exceeded. It now should
2525 MH/03 Made $recipients available in local_scan(). local_scan() already has
2526 better access to the recipient list through recipients_list[], but
2527 $recipients can be useful in postmaster-provided expansion strings.
2529 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
2530 in the case of a MAIL command with additional options following the
2531 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
2532 were accidentally chopped off.
2534 PH/29 SMTP synchronization checks are implemented when a command is read -
2535 there is a check that no more input is waiting when there shouldn't be
2536 any. However, for some commands, a delay in an ACL can mean that it is
2537 some time before the response is written. In this time, more input might
2538 arrive, invalidly. So now there are extra checks after an ACL has run for
2539 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
2540 pipelining has not been advertised.
2542 PH/30 MH's patch to allow iscntrl() characters to be list separators.
2544 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
2545 returned in the SMTP response when smtp_return_error_details was false.
2546 This has been fixed.
2548 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
2549 instead of the C I/O that was originally supplied, because problems were
2550 reported on Solaris.
2552 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
2553 Exim which did not show up earlier: it was assuming that a call to
2554 SSL_CTX_set_info_callback() might give an error value. In fact, there is
2555 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
2556 was a macro that became an assignment, so it seemed to work. This has
2557 changed to a proper function call with a void return, hence the compile
2558 error. Exim's code has been fixed.
2560 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
2563 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
2565 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
2567 PH/37 If a message is not accepted after it has had an id assigned (e.g.
2568 because it turns out to be too big or there is a timeout) there is no
2569 "Completed" line in the log. When some messages of this type were
2570 selected by exigrep, they were listed as "not completed". Others were
2571 picked up by some special patterns. I have improved the selection
2572 criteria to be more general.
2574 PH/38 The host_find_failed option in the manualroute router can now be set
2575 to "ignore", to completely ignore a host whose IP address cannot be
2576 found. If all hosts are ignored, the behaviour is controlled by the new
2577 host_all_ignored option.
2579 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
2580 homing or because of multiple MX records with /mx) generated more than
2581 one IP address, and the following item turned out to be the local host,
2582 all the secondary addresses of the first item were incorrectly removed
2583 from the list, along with the local host and any following hosts (which
2584 is what is supposed to happen).
2586 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
2587 whoever called Exim into the -H file. In the case of the daemon it was
2588 behaving confusingly. When first started, it used values for whoever
2589 started the daemon, but after a SIGHUP it used the Exim user (because it
2590 calls itself on a restart). I have changed the code so that it now always
2593 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
2594 message are rejected with the same error (e.g. no authentication or bad
2595 sender address), and a DATA command is nevertheless sent (as can happen
2596 with PIPELINING or a stupid MUA), the error message that was given to the
2597 RCPT commands is included in the rejection of the DATA command. This is
2598 intended to be helpful for MUAs that show only the final error to their
2601 PH/42 Another patch from the Sieve maintainer.
2603 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
2604 Eximstats - Fixed some broken HTML links and added missing column headers
2606 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
2607 columns for Rejects, Temp Rejects, Ham, and Spam rows.
2609 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
2611 PH/43 Yet another patch from the Sieve maintainer.
2613 PH/44 I found a way to check for a TCP/IP connection going away before sending
2614 the response to the final '.' that terminates a message, but only in the
2615 case where the client has not sent further data following the '.'
2616 (unfortunately, this is allowed). However, in many cases there won't be
2617 any further data because there won't be any more messages to send. A call
2618 to select() can be used: if it shows that the input is "ready", there is
2619 either input waiting, or the socket has been closed. An attempt to read
2620 the next input character can distinguish the two cases. Previously, Exim
2621 would have sent an OK response which the client would never have see.
2622 This could lead to message repetition. This fix should cure that, at
2623 least in a lot of common cases.
2625 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
2626 advertised in response to EHLO.
2632 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
2633 fixed by 4.65/MH/01 (is this a record?) are fixed:
2635 (i) An empty string was always treated as zero by the numeric comparison
2636 operators. This behaviour has been restored.
2638 (ii) It is documented that the numeric comparison operators always treat
2639 their arguments as decimal numbers. This was broken in that numbers
2640 starting with 0 were being interpreted as octal.
2642 While fixing these problems I realized that there was another issue that
2643 hadn't been noticed. Values of message_size_limit (both the global option
2644 and the transport option) were treated as octal if they started with 0.
2645 The documentation was vague. These values are now always treated as
2646 decimal, and I will make that clear in the documentation.
2652 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
2653 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
2656 MH/01 Don't check that the operands of numeric comparison operators are
2657 integers when their expansion is in "skipping" mode (fixes bug
2658 introduced by 4.64-PH/07).
2660 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
2661 child addresses, Exim now panics and dies. Previously, because the count
2662 is held in a short int, deliveries were likely to be lost. As such a
2663 large number of recipients for a single message is ridiculous
2664 (performance will be very, very poor), I have chosen to impose a limit
2665 rather than extend the field.
2671 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
2672 leftover -K file (the existence of which was triggered by #402).
2673 While we were at it, introduced process PID as part of the -K
2674 filename. This should rule out race conditions when creating
2677 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
2678 processing considerably. Previous code took too long for large mails,
2679 triggering a timeout which in turn triggers #401.
2681 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
2682 in the DK code in transports.c. sendfile() is not really portable,
2683 hence the _LINUX specificness.
2685 TF/01 In the add_headers option to the mail command in an Exim filter,
2686 there was a bug that Exim would claim a syntax error in any
2687 header after the first one which had an odd number of characters
2690 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
2691 callout verification, Exim cached a "reject" for the entire domain. This
2692 is correct for most verifications, but it is not correct for a recipient
2693 verification with use_sender or use_postmaster set, because in that case
2694 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
2695 case of MAIL FROM:<> rejection from other early rejections (e.g.
2696 rejection of HELO). When verifying a recipient using a non-null MAIL
2697 address, the cache is ignored if it shows MAIL FROM:<> rejection.
2698 Whatever the result of the callout, the value of the domain cache is
2699 left unchanged (for any other kind of callout, getting as far as trying
2700 RCPT means that the domain itself is ok).
2702 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
2705 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
2706 manifest itself as EPIPE rather than ECONNECT. When tidying away a
2707 session, the daemon ignores ECONNECT errors and logs others; it now
2708 ignores EPIPE as well.
2710 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
2711 (quoted-printable decoding).
2713 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
2714 later the small subsequent patch to fix an introduced bug.
2716 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
2718 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
2720 PH/08 An error is now given if message_size_limit is specified negative.
2722 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
2723 to be given (somewhat) arbitrary names.
2725 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
2728 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
2729 miscellaneous code fixes
2731 PH/10 Added the log_reject_target ACL modifier to specify where to log
2734 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
2735 hostname. This is wrong, because it relates to the incoming message (and
2736 probably the interface on which it is arriving) and not to the outgoing
2737 callout (which could be using a different interface). This has been
2738 changed to use the value of the helo_data option from the smtp transport
2739 instead - this is what is used when a message is actually being sent. If
2740 there is no remote transport (possible with a router that sets up host
2741 addresses), $smtp_active_hostname is used.
2743 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
2744 tweaks were necessary in order to get it to work (see also 21 below):
2745 (a) The code assumed that strncpy() returns a negative number on buffer
2746 overflow, which isn't the case. Replaced with Exim's string_format()
2748 (b) There were several signed/unsigned issues. I just did the minimum
2749 hacking in of casts. There is scope for a larger refactoring.
2750 (c) The code used strcasecmp() which is not a standard C function.
2751 Replaced with Exim's strcmpic() function.
2752 (d) The code set only $1; it now sets $auth1 as well.
2753 (e) A simple test gave the error "authentication client didn't specify
2754 service in request". It would seem that Dovecot has changed its
2755 interface. Fortunately there's a specification; I followed it and
2756 changed what the client sends and it appears to be working now.
2758 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
2761 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
2762 address A is aliased to B and C, where B exists and C does not. Without
2763 -v the output is "A verified" because verification stops after a
2764 successful redirection if more than one address is generated. However,
2765 with -v the child addresses are also verified. Exim was outputting "A
2766 failed to verify" and then showing the successful verification for C,
2767 with its parentage. It now outputs "B failed to verify", showing B's
2768 parentage before showing the successful verification of C.
2770 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
2771 look up a TXT record in a specific list after matching in a combined
2774 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
2775 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
2776 they consult the DNS. I had assumed they would set it the way they
2777 wanted; and indeed my experiments on Linux seem to show that in some
2778 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
2779 To be on the safe side, however, I have now made the interface to
2780 host_find_byname() similar to host_find_bydns(), with an argument
2781 containing the DNS resolver options. The host_find_byname() function now
2782 sets these options at its start, just as host_find_bydns() does. The smtp
2783 transport options dns_qualify_single and dns_search_parents are passed to
2784 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
2785 of host_find_byname() use the default settings of RES_DEFNAMES
2786 (qualify_single) but not RES_DNSRCH (search_parents).
2788 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
2789 spool_read_header() do less string testing, by means of a preliminary
2790 switch on the second character of optional "-foo" lines. (This is
2791 overdue, caused by the large number of possibilities that now exist.
2792 Originally there were few.) While I was there, I also converted the
2793 str(n)cmp tests so they don't re-test the leading "-" and the first
2794 character, in the hope this might squeeze out yet more improvement.
2796 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
2797 flag allowing group syntax was set by the header_syntax check but not
2798 turned off, possible causing trouble later; (2) The flag was not being
2799 set at all for the header_verify test, causing "group"-style headers to
2800 be rejected. I have now set it in this case, and also caused header_
2801 verify to ignore an empty address taken from a group. While doing this, I
2802 came across some other cases where the code for allowing group syntax
2803 while scanning a header line wasn't quite right (mostly, not resetting
2804 the flag correctly in the right place). These bugs could have caused
2805 trouble for malformed header lines. I hope it is now all correct.
2807 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
2808 with the "reply" argument non-NULL. The code, however (which originally
2809 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
2810 but it didn't always do it. This confused somebody who was copying the
2811 code for some other use. I have removed all the tests.
2813 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
2814 feature that was used to support insecure browsers during the U.S. crypto
2815 embargo. It requires special client support, and Exim is probably the
2816 only MTA that supported it -- and would never use it because real RSA is
2817 always available. This code has been removed, because it had the bad
2818 effect of slowing Exim down by computing (never used) parameters for the
2819 RSA_EXPORT functionality.
2821 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
2822 authenticator to fail if there's a tab character in the incoming data
2823 (there should never be unless someone is messing about, as it's supposed
2824 to be base64-encoded). Also added, on Timo's advice, the "secured" option
2825 if the connection is using TLS or if the remote IP is the same as the
2826 local IP, and the "valid-client-cert option" if a client certificate has
2829 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
2830 authenticators. This can be used for authorization after authentication
2831 succeeds. (In the case of plaintext, it servers for both authentication
2834 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
2835 if any retry times were supplied.
2837 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
2838 connection, where there is no client IP address to check. In this
2839 situation, the verify now always succeeds.
2841 PH/25 Applied John Jetmore's -Mset patch.
2843 PH/26 Added -bem to be like -Mset, but loading a message from a file.
2845 PH/27 In a string expansion for a processed (not raw) header when multiple
2846 headers of the same name were present, leading whitespace was being
2847 removed from all of them, but trailing whitespace was being removed only
2848 from the last one. Now trailing whitespace is removed from each header
2849 before concatenation. Completely empty headers in a concatenation (as
2850 before) are ignored.
2852 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
2853 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
2855 PH/29 [Removed. This was a change that I later backed out, and forgot to
2856 correct the ChangeLog entry (that I had efficiently created) before
2857 committing the later change.]
2859 PH/30 Exim was sometimes attempting to deliver messages that had suffered
2860 address errors (4xx response to RCPT) over the same connection as other
2861 messages routed to the same hosts. Such deliveries are always "forced",
2862 so retry times are not inspected. This resulted in far too many retries
2863 for the affected addresses. The effect occurred only when there were more
2864 hosts than the hosts_max_try setting in the smtp transport when it had
2865 the 4xx errors. Those hosts that it had tried were not added to the list
2866 of hosts for which the message was waiting, so if all were tried, there
2867 was no problem. Two fixes have been applied:
2869 (i) If there are any address or message errors in an SMTP delivery, none
2870 of the hosts (tried or untried) are now added to the list of hosts
2871 for which the message is waiting, so the message should not be a
2872 candidate for sending over the same connection that was used for a
2873 successful delivery of some other message. This seems entirely
2874 reasonable: after all the message is NOT "waiting for some host".
2875 This is so "obvious" that I'm not sure why it wasn't done
2876 previously. Hope I haven't missed anything, but it can't do any
2877 harm, as the worst effect is to miss an optimization.
2879 (ii) If, despite (i), such a delivery is accidentally attempted, the
2880 routing retry time is respected, so at least it doesn't keep
2881 hammering the server.
2883 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
2884 in ${readsocket because some servers need this prod.
2886 PH/32 Added some extra debug output when updating a wait-xxx database.
2888 PH/33 The hint "could be header name not terminated by colon", which has been
2889 given for certain expansion errors for a long time, was not being given
2890 for the ${if def:h_colon_omitted{... case.
2892 PH/34 The spec says: "With one important exception, whenever a domain list is
2893 being scanned, $domain contains the subject domain." There was at least
2894 one case where this was not true.
2896 PH/35 The error "getsockname() failed: connection reset by peer" was being
2897 written to the panic log as well as the main log, but it isn't really
2898 panic-worthy as it just means the connection died rather early on. I have
2899 removed the panic log writing for the ECONNRESET error when getsockname()
2902 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
2903 runs only) independently of the message's sender address. This meant
2904 that, if the 4xx error was in fact related to the sender, a different
2905 message to the same recipient with a different sender could confuse
2906 things. In particular, this can happen when sending to a greylisting
2907 server, but other circumstances could also provoke similar problems.
2908 I have changed the default so that the retry time for these errors is now
2909 based a combination of the sender and recipient addresses. This change
2910 can be overridden by setting address_retry_include_sender=false in the
2913 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
2914 remote server are returned as part of bounce messages. This was not
2915 happening for LMTP over a pipe (the lmtp transport), but now it is the
2916 same for both kinds of LMTP.
2918 PH/38 Despite being documented as not happening, Exim was rewriting addresses
2919 in header lines that were in fact CNAMEs. This is no longer the case.
2921 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
2922 and queue runs started by the daemon processed all messages. This has
2923 been fixed so that -R and -S can now usefully be given with -q<time>.
2925 PH/40 Import PCRE release 6.7 (fixes some bugs).
2927 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
2929 PH/42 Give an error if -q is specified more than once.
2931 PH/43 Renamed the variables $interface_address and $interface_port as
2932 $received_ip_address and $received_port, to make it clear that these
2933 values apply to message reception, and not to the outgoing interface when
2934 a message is delivered. (The old names remain recognized, of course.)
2936 PH/44 There was no timeout on the connect() call when using a Unix domain
2937 socket in the ${readsocket expansion. There now is.
2939 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
2940 be meaningful with "accept".
2942 SC/01 Eximstats V1.43
2943 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
2945 SC/02 Eximstats V1.44
2946 Use a glob alias rather than an array ref in the generated
2947 parser. This improves both readability and performance.
2949 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
2950 Collect SpamAssassin and rejection statistics.
2951 Don't display local sender or destination tables unless
2952 there is data to show.
2953 Added average volumes into the top table text output.
2955 SC/04 Eximstats V1.46
2956 Collect data on the number of addresses (recipients)
2957 as well as the number of messages.
2959 SC/05 Eximstats V1.47
2960 Added 'Message too big' to the list of mail rejection
2961 reasons (thanks to Marco Gaiarin).
2963 SC/06 Eximstats V1.48
2964 Mainlog lines which have GMT offsets and are too short to
2965 have a flag are now skipped.
2967 SC/07 Eximstats V1.49 (Alain Williams)
2968 Added the -emptyok flag.
2970 SC/08 Eximstats V1.50
2971 Fixes for obtaining the IP address from reject messages.
2973 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
2974 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
2975 whitespace changes from 4.64-PH/27
2977 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
2980 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
2981 are found, allow negative numbers in numeric criteria)
2983 JJ/06 exipick.20061117.2, added new $message_body_missing variable
2985 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
2986 to match changes made in 4.64-PH/43
2988 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
2990 PH/47 Put in an explicit test for a DNS lookup of an address record where the
2991 "domain" is actually an IP address, and force a failure. This locks out
2992 those revolvers/nameservers that support "A-for-A" lookups, in
2993 contravention of the specifications.
2995 PH/48 When a host name was looked up from an IP address, and the subsequent
2996 forward lookup of the name timed out, the host name was left in
2997 $sender_host_name, contrary to the specification.
2999 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
3000 restricted to single-key lookups, Exim was not diagnosing an error if
3001 * or *@ was used with a query-style lookup.
3003 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
3005 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
3006 long ago, but noone interested enough thought of it. Let's just say that
3007 the "1.1" means that there are some new functions that weren't there at
3008 some point in the past.
3010 PH/51 Error processing for expansion failure of helo_data from an smtp
3011 transport during callout processing was broken.
3013 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
3014 tested/used via the -bh/-bhc/-bs options.
3016 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
3017 bug, fixed in subsequent PCRE releases).
3019 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
3020 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
3022 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
3028 SC/01 Use a glob alias rather than an array ref in eximstats generated
3029 parser. This improves both readability and performance.
3031 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
3032 Don't display local sender or destination tables in eximstats unless
3033 there is data to show.
3034 Added average volumes into the eximstats top table text output.
3036 SC/03 Collect data on the number of addresses (recipients) as well
3037 as the number of messages in eximstats.
3039 TF/01 Correct an error in the documentation for the redirect router. Exim
3040 does (usually) call initgroups() when daemonizing.
3042 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
3043 with consistent privilege compared to when running as a daemon.
3045 TF/03 Note in the spec that $authenticated_id is not set for local
3046 submissions from trusted users.
3048 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
3049 Thanks to Dean Brooks <dean@iglou.com> for the patch.
3051 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
3052 by adding some example configuration directives to the default
3053 configuration file. A little bit of work is required to uncomment the
3054 directives and define how usernames and passwords are checked, but
3055 there is now a framework to start from.
3057 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
3058 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
3059 without this. I don't know how relevant this is to other LDAP libraries.
3061 PH/02 Add the verb name to the "unknown ACL verb" error.
3063 PH/03 Magnus Holmgren's patch for filter_prepend_home.
3065 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
3067 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
3068 directory not expanded when it should be if an expanded home directory
3069 was set for the address (which is overridden by the transport).
3071 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
3074 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
3075 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
3076 because it is too late at that time, and has no effect.
3078 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
3079 security issue with \' (bugzilla #107). I could not use the
3080 PQescapeStringConn() function, because it needs a PGconn value as one of
3083 PH/08 When testing addresses using -bt, indicate those final addresses that
3084 are duplicates that would not cause an additional delivery. At least one
3085 person was confused, thinking that -bt output corresponded to deliveries.
3086 (Suppressing duplicates isn't a good idea as you lose the information
3087 about possibly different redirections that led to the duplicates.)
3089 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
3090 systems where poll() doesn't work, in particular OS X.
3092 PH/10 Added more information to debugging output for retry time not reached.
3094 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
3095 operations in malware.c.
3097 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
3100 PH/13 If write_rejectlog was set false when logging was sent to syslog with
3101 syslog_duplication set false, log lines that would normally be written
3102 both the the main log and to the reject log were not written to syslog at
3105 PH/14 In the default configuration, change the use of "message" in ACL warn
3106 statements to "add_header".
3108 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
3109 not followed by a command (e.g. "seen endif").
3111 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
3112 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
3115 PH/17 Added extra conditions to the default value of delay_warning_condition
3119 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
3120 { match{$h_precedence:}{(?i)bulk|list|junk} } \
3121 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
3124 The Auto-Submitted: and various List- headers are standardised, whereas I
3125 don't think Precedence: ever was.
3127 PH/18 Refactored debugging code in route_finduser() to show more information,
3128 in particular, the error code if getpwnam() issues one.
3130 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
3131 This is apparently needed in addition to the PH/07 change above to avoid
3132 any possible encoding problems.
3134 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
3135 but not after initializing Perl.
3137 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
3138 output them only if debugging. By default they are written stderr,
3139 apparently, which is not desirable.
3141 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
3144 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
3147 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
3149 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
3150 authenticated or an ident call has been made. Suppress the default
3151 values for $authenticated_id and $authenticated_sender (but permit -oMai
3152 and -oMas) when testing with -bh.
3154 PH/24 Re-jigged the order of the tests in the default configuration so that the
3155 tests for valid domains and recipients precede the DNS black list and CSA
3156 tests, on the grounds that those ones are more expensive.
3158 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
3159 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
3160 This bug exists in every version of Exim that I still have, right back to
3163 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
3164 However, an attempt to turn on case-sensitivity in a regex key by
3165 including (?-i) didn't work because the subject string was already
3166 lowercased, and the effects were non-intuitive. It turns out that a
3167 one-line patch can be used to allow (?-i) to work as expected.
3173 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
3174 other effects) broke the use of negated acl sub-conditions.
3176 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
3179 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
3180 "Deny" causes Exim to reject the incoming connection with a 554 error.
3181 Unfortunately, if there is a major crisis, such as a disk failure,
3182 tcp-wrappers gives "deny", whereas what one would like would be some
3183 kind of temporary error. A kludge has been added to help with this.
3184 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
3185 554 error is used if errno is still zero or contains ENOENT (which occurs
3186 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
3189 PH/03 Add -lutil to the default FreeBSD LIBS setting.
3191 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
3192 errors. Otherwise a message that provokes a temporary error (when other
3193 messages do not) can cause a whole host to time out.
3195 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
3196 addresses were routed directly to files or pipes from a redirect router.
3197 File deliveries just didn't batch; pipe deliveries might have suffered
3200 PH/06 A failure to get a lock for a hints database would erroneously always say
3201 "Failed to get write lock", even when it was really a read lock.
3203 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
3204 of 0600. This has been changed to use the value of the lockfile_mode
3205 option (which defaults to 0600).
3207 PH/08 Applied small patch from the Sieve maintainer.
3209 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
3210 folder from quota calculations, a direct delivery into this folder messed
3211 up the contents of the maildirsize file. This was because the regex was
3212 used only to exclude .Trash (or whatever) when the size of the mailbox
3213 was calculated. There was no check that a delivery was happening into an
3214 excluded directory. This bug has been fixed by ignoring all quota
3215 processing for deliveries into excluded directories.
3217 PH/10 Added the maildirfolder_create_regex option to appendfile.
3223 PH/01 The code for finding all the local interface addresses on a FreeBSD
3224 system running IPv6 was broken. This may well have applied to all BSD
3225 systems, as well as to others that have similar system calls. The broken
3226 code found IPv4 interfaces correctly, but gave incorrect values for the
3227 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
3228 that it would not match correctly against @[] and not recognize the IPv6
3231 PH/02 The ipliteral router was not recognizing addresses of the form user@
3232 [ipv6:....] because it didn't know about the "ipv6:" prefix.
3234 PH/03 Added disable_ipv6.
3236 PH/04 Changed $reply_address to use the raw form of the headers instead of the
3237 decoded form, because it is most often used to construct To: headers
3238 lines in autoreplies, and the decoded form may well be syntactically
3239 invalid. However, $reply_address has leading white space removed, and all
3240 newlines turned into spaces so that the autoreply transport does not
3243 PH/05 If group was specified without a user on a router, and no group or user
3244 was specified on a transport, the group from the router was ignored.
3246 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
3247 for visible compile-time settings that can be used to change these
3248 numbers, for those that want even more. Backwards compatibility with old
3249 spool files has been maintained. However, going back to a previous Exim
3250 release will lost any variables that are in spool files.
3252 PH/07 Two small changes when running in the test harness: increase delay when
3253 passing a TCP/IP connection to a new process, in case the original
3254 process has to generate a bounce, and remove special handling of
3255 127.0.0.2 (sic), which is no longer necessary.
3257 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
3258 be the same on different OS.
3260 PH/09 Moved a debug statement in filter processing to avoid a race problem when
3263 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
3264 whether --show-vars was specified or not
3266 JJ/02 exipick: Added support for new ACL variable spool format introduced
3269 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
3270 syntactically invalid From: or Reply-to: line, and a filter used this to
3271 generate an autoreply, and therefore failed to obtain an address for the
3272 autoreply, Exim could try to deliver to a non-existent relative file
3273 name, causing unrelated and misleading errors. What now happens is that
3274 it logs this as a hard delivery error, but does not attempt to create a
3277 PH/11 The exinext utility has a -C option for testing purposes, but although
3278 the given file was scanned by exinext itself; it wasn't being passed on
3279 when Exim was called.
3281 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
3282 an end-of-file indication when reading a command response.
3284 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
3285 compiled. In many other places in Exim, IPv6 addresses are always
3286 recognized, so I have changed this. It also means that IPv4 domain
3287 literals of the form [IPV4:n.n.n.n] are now always recognized.
3289 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
3290 used if the router is not running as root, for example, when verifying at
3291 ACL time, or when using -bh. The debugging output from this situation was
3292 non-existent - all you got was a failure to exec. I have made two
3295 (a) Failures to set uid/gid, the current directory, or a process leader
3296 in a subprocess such as that created by queryprogram now generate
3297 suitable debugging output when -d is set.
3299 (b) The queryprogram router detects when it is not running as root,
3300 outputs suitable debugging information if -d is set, and then runs
3301 the subprocess without attempting to change uid/gid.
3303 PH/15 Minor change to Makefile for building test_host (undocumented testing
3306 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
3307 additional section of a DNS packet that returns MX or SRV records.
3308 Instead, it always explicitly searches for A/AAAA records. This avoids
3309 major problems that occur when a DNS server includes only records of one
3310 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
3311 fixed another bug: if SRV records were looked up and the corresponding
3312 address records were *not* found in the additional section, the port
3313 values from the SRV records were lost.
3315 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
3316 using the correct key (the original address) when searching the retry
3317 rules in order to find which one to use for generating the retry hint.
3319 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
3320 adding the default one. Similarly, if it contains a Reply-To: header, the
3321 errors_reply_to option, if set, is not used.
3323 PH/19 When calculating a retry time, Exim used to measure the "time since
3324 failure" by looking at the "first failed" field in the retry record. Now
3325 it does not use this if it is later than than the arrival time of the
3326 message. Instead it uses the arrival time. This makes for better
3327 behaviour in cases where some deliveries succeed, thus re-setting the
3328 "first failed" field. An example is a quota failure for a huge message
3329 when small messages continue to be delivered. Without this change, the
3330 "time since failure" will always be short, possible causing more frequent
3331 delivery attempts for the huge message than are intended.
3332 [Note: This change was subsequently modified - see PH/04 for 4.62.]
3334 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
3335 $1, $2, $3) because the numerical variables can be reset during some
3336 expansion items (e.g. "match"), thereby losing the authentication data.
3338 PH/21 Make -bV show the size of off_t variables so that the test suite can
3339 decide whether to run tests for quotas > 2G.
3341 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
3342 mailbox_size, and mailbox_filecount in the appendfile transport. If a
3343 filecount value is greater than 2G or if a quota value is greater than 2G
3344 on a system where the size of off_t is not greater than 4, a panic error
3347 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
3348 never match. The debug and -bh output now contains an explicit error
3349 message indicating a malformed IPv4 address or mask.
3351 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
3352 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
3353 PH/23 above applies.
3355 PH/25 Do not write to syslog when running in the test harness. The only
3356 occasion when this arises is a failure to open the main or panic logs
3357 (for which there is an explicit test).
3359 PH/26 Added the /no_tell option to "control=freeze".
3361 PH/27 If a host name lookup failed very early in a connection, for example, if
3362 the IP address matched host_lookup and the reverse lookup yielded a name
3363 that did not have a forward lookup, an error message of the form "no IP
3364 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
3365 could be logged. Now it outputs the IP address instead of "NULL".
3367 PH/28 An enabling patch from MH: add new function child_open_exim2() which
3368 allows the sender and the authenticated sender to be set when
3369 submitting a message from within Exim. Since child_open_exim() is
3370 documented for local_scan(), the new function should be too.
3372 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
3373 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
3374 results in an empty string is now treated as unset.
3376 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
3378 PH/31 Added sender_verify_fail logging option.
3380 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
3381 needed by RFC 822 but not by RFC 2822 was commented out. I have now
3382 tidied the source and removed it altogether.
3384 PH/33 When a queue run was abandoned because the load average was too high, a
3385 log line was always written; now it is written only if the queue_run log
3386 selector is set. In addition, the log line for abandonment now contains
3387 information about the queue run such as the pid. This is always present
3388 in "start" and "stop" lines but was omitted from the "abandon" line.
3390 PH/34 Omit spaces between a header name and the colon in the error message that
3391 is given when verify = headers_syntax fails (if there are lots of them,
3392 the message gets confusing).
3394 PH/35 Change the default for dns_check_names_pattern to allow slashes within
3395 names, as there are now some PTR records that contain slashes. This check
3396 is only to protect against broken name servers that fall over on strange
3397 characters, so the fact that it applies to all lookups doesn't matter.
3399 PH/36 Now that the new test suite is complete, we can remove some of the
3400 special code in Exim that was needed for the old test suite. For example,
3401 sorting DNS records because real resolvers return them in an arbitrary
3402 order. The new test suite's fake resolver always returns records in the
3405 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
3406 bounces) except when queue_only is set, to avoid logging races between
3407 the different processes.
3409 PH/38 Panic-die if .include specifies a non-absolute path.
3411 PH/39 A tweak to the "H" retry rule from its user.
3413 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
3414 a label. They prevented compilation on older perls.
3416 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
3417 a warning to be raised on newish perls.
3419 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
3420 on queue. Changes to match documented behaviour of showing count of
3421 messages matching specified criteria.
3423 PH/40 Changed the default ident timeout from 30s to 5s.
3425 PH/41 Added support for the use of login_cap features, on those BSD systems
3426 that have them, for controlling the resources used by pipe deliveries.
3428 PH/42 The content-scanning code uses fopen() to create files in which to put
3429 message data. Previously it was not paying any attention to the mode of
3430 the files. Exim runs with umask(0) because the rest of the code creates
3431 files with open(), and sets the required mode explicitly. Thus, these
3432 files were ending up world-writeable. This was not a big issue, because,
3433 being within the spool directory, they were not world-accessible. I have
3434 created a function called modefopen, which takes an additional mode
3435 argument. It sets umask(777), creates the file, chmods it to the required
3436 mode, then resets the umask. All the relevant calls to fopen() in the
3437 content scanning code have been changed to use this function.
3439 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
3440 to 24 hours. This avoids potential overflow problems when processing G
3441 and H retry rules. I suspect nobody ever tinkers with this value.
3443 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
3445 PH/45 When the plaintext authenticator is running as a client, the server's
3446 challenges are checked to ensure they are valid base64 strings. By
3447 default, the authentication attempt is cancelled if an invalid string is
3448 received. Setting client_ignore_invalid_base64 true ignores these errors.
3449 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
3450 they are received. Thus, the responses can be made to depend on the
3451 challenges. If an invalid string is ignored, an empty string is placed in
3454 PH/46 Messages that are created by the autoreply transport now contains a
3455 References: header, in accordance with RFCs 2822 and 3834.
3457 PH/47 Added authenticated_sender_force to the smtp transport.
3459 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
3461 PH/49 Installed latest patch from the Sieve maintainer.
3463 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
3464 was also set, the appendfile transport was unnecessarily scanning a
3465 directory of message files (e.g. for maildir delivery) to find the count
3466 of files (along with the size), even though it did not need this
3467 information. It now does the scan only if it needs to find either the
3468 size of the count of files.
3470 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
3472 PH/52 Two bugs concerned with error handling when the smtp transport is
3475 (i) Exim was not creating retry information for temporary errors given
3476 for individual recipients after the DATA command when the smtp transport
3477 was used in LMTP mode. This meant that they could be retried too
3478 frequently, and not timed out correctly.
3480 (ii) Exim was setting the flag that allows error details to be returned
3481 for LMTP errors on RCPT commands, but not for LMTP errors for individual
3482 recipients that were returned after the DATA command.
3484 PH/53 This is related to PH/52, but is more general: for any failing address,
3485 when detailed error information was permitted to be returned to the
3486 sender, but the error was temporary, then after the final timeout, only
3487 "retry timeout exceeded" was returned. Now it returns the full error as
3488 well as "retry timeout exceeded".
3490 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
3491 do this, and (what is worse) MTAs that accept it.
3493 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
3494 will now be deprecated.
3496 PH/56 New os.c-cygwin from the Cygwin maintainer.
3498 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
3499 formats (previously only available in exim formats via -bpr, -bpru,
3500 and -bpra. Now also available in native and exiqgrep formats)
3502 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
3503 with very large, slow to parse queues
3505 JJ/08 exipick: added ! as generic prefix to negate any criteria format
3507 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
3509 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
3510 responses to authentication challenges, though it was showing the
3511 challenges; (ii) I've removed the CR characters from the debug output for
3514 PH/58 Allow for the insertion of a newline as well as a space when a string
3515 is turned into more than one encoded-word during RFC 2047 encoding. The
3516 Sieve code now uses this.
3518 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
3519 data_4xx, lost_connection, tls_required.
3521 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
3522 message was being sent as an SMTP response.
3524 PH/61 Add -l and -k options to exicyclog.
3526 PH/62 When verifying, if an address was redirected to one new address, so that
3527 verification continued, and the new address failed or deferred after
3528 having set something in $address_data, the value of $address_data was not
3529 passed back to the ACL. This was different to the case when no
3530 redirection occurred. The value is now passed back in both cases.
3532 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
3533 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
3534 use login_cap.h, so on its own it isn't the distinguishing feature. The
3535 new name refers directly to the setclassresources() function.
3537 PH/65 Added configuration files for NetBSD3.
3539 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
3541 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
3542 is preferred over IPv4.
3544 PH/68 The bounce_return_message and bounce_return_body options were not being
3545 honoured for bounces generated during the reception of non-SMTP messages.
3546 In particular, this applied to messages rejected by the ACL. This bug has
3547 been fixed. However, if bounce_return_message is true and bounce_return_
3548 body is false, the headers that are returned for a non-SMTP message
3549 include only those that have been read before the error was detected.
3550 (In the case of an ACL rejection, they have all been read.)
3552 PH/69 The HTML version of the specification is now built in a directory called
3553 spec_html instead of spec.html, because the latter looks like a path with
3554 a MIME-type, and this confuses some software.
3556 PH/70 Catch two compiler warnings in sieve.c.
3558 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
3559 function verify_get_ident() calls ip_connect() to connect a socket, but
3560 if the "connect()" function timed out, ip_connect() used to close the
3561 socket. However, verify_get_ident() also closes the socket later, and in
3562 between Exim writes to the log, which may get opened at this point. When
3563 the socket was closed in ip_connect(), the log could get the same file
3564 descriptor number as the socket. This naturally causes chaos. The fix is
3565 not to close the socket in ip_connect(); the socket should be closed by
3566 the function that creates it. There was only one place in the code where
3567 this was missing, in the iplookup router, which I don't think anybody now
3568 uses, but I've fixed it anyway.
3570 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
3571 well as to direct DNS lookups. Otherwise the handling of names in host
3572 lists is inconsistent and therefore confusing.
3578 PH/01 Two changes to the default runtime configuration:
3580 (1) Move the checks for relay_from_hosts and authenticated clients from
3581 after to before the (commented out) DNS black list checks.
3583 (2) Add control=submission to the relay_from_hosts and authenticated
3584 clients checks, on the grounds that messages accepted by these
3585 statements are most likely to be submissions.
3587 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
3589 (1) Generate an error if the third argument for the ${prvs expansion is
3592 (2) Treat a missing third argument of ${prvscheck as if it were an empty
3595 (3) Reset the variables that are obtained from the first argument of
3596 ${prvscheck and used in the second argument before leaving the code,
3597 because their memory is reclaimed, so using them afterwards may do
3600 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
3601 one (it's much easier than Tom thought :-).
3603 (5) Because of (4), we can now allow for the use of $prvscheck_result
3604 inside the third argument.
3606 PH/03 For some reason, the default setting of PATH when running a command from
3607 a pipe transport was just "/usr/bin". I have changed it to
3610 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
3611 anything to be listed in the output from -bV.
3613 PH/05 When a filter generated an autoreply, the entire To: header line was
3614 quoted in the delivery log line, like this:
3616 => >A.N.Other <ano@some.domain> <original@ddress> ...
3618 This has been changed so that it extracts the operative address. There
3619 may be more than one such address. If so, they are comma-separated, like
3622 => >ano@some.domain,ona@other.domain <original@ddress> ...
3624 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
3625 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
3626 not being looked up in the rDNS to get a host name, Exim was showing the
3627 IP address twice in Received: lines, even though the IP addresses were
3628 identical. For example:
3630 Received: from [1.2.3.4] (helo=[1.2.3.4])
3632 However, if the real host name was known, it was omitting the HELO data
3633 if it matched the actual IP address. This has been tidied up so that it
3634 doesn't show the same IP address twice.
3636 PH/07 When both +timestamp and +memory debugging was on, the value given by
3637 $tod_xxx expansions could be wrong, because the tod_stamp() function was
3638 called by the debug printing, thereby overwriting the timestamp buffer.
3639 Debugging no longer uses the tod_stamp() function when +timestamp is set.
3641 PH/08 When the original message was included in an autoreply transport, it
3642 always said "this is a copy of the message, including all the headers",
3643 even if body_only or headers_only was set. It now gives an appropriate
3646 PH/09 Applied a patch from the Sieve maintainer which:
3648 o fixes some comments
3649 o adds the (disabled) notify extension core
3650 o adds some debug output for the result of if/elsif tests
3651 o points to the current vacation draft in the documentation
3652 and documents the missing references header update
3656 o fixes a bug in processing the envelope test (when testing
3657 multiple envelope elements, the last element determined the
3660 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
3661 Electronic Mail") by including:
3663 Auto-submitted: auto-generated
3665 in the messages that it generates (bounce messages and others, such as
3666 warnings). In the case of bounce messages for non-SMTP messages, there was
3667 also a typo: it was using "Auto_submitted" (underscore instead of
3668 hyphen). Since every message generated by Exim is necessarily in response
3669 to another message, thes have all been changed to:
3671 Auto-Submitted: auto-replied
3673 in accordance with these statements in the RFC:
3675 The auto-replied keyword:
3677 - SHOULD be used on messages sent in direct response to another
3678 message by an automatic process,
3680 - MUST NOT be used on manually-generated messages,
3682 - MAY be used on Delivery Status Notifications (DSNs) and Message
3683 Disposition Notifications (MDNs),
3685 - MUST NOT be used on messages generated by automatic or periodic
3686 processes, except for messages which are automatic responses to
3689 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
3690 to the default Received: header definition.
3692 PH/12 Added log selector acl_warn_skipped (default on).
3694 PH/13 After a successful wildlsearch lookup, discard the values of numeric
3695 variables because (a) they are in the wrong storage pool and (b) even if
3696 they were copied, it wouldn't work properly because of the caching.
3698 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
3699 checking when decoding. Apparently there are clients that generate
3700 overlong encoded strings. Why am I not surprised?
3702 PH/15 If the first argument of "${if match_address" was not empty, but did not
3703 contain an "@" character, Exim crashed. Now it writes a panic log message
3704 and treats the condition as false.
3706 PH/16 In autoreply, treat an empty string for "once" the same as unset.
3708 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
3709 extension "envelope-auth". The code is finished and in agreement with
3710 other implementations, but there is no documentation so far and in fact,
3711 nobody wrote the draft yet. This extension is currently #undef'ed, thus
3712 not changing the active code.
3714 Print executed "if" and "elsif" statements when debugging is used. This
3715 helps a great deal to understand what a filter does.
3717 Document more things not specified clearly in RFC3028. I had all this
3718 sorted out, when out of a sudden new issues came to my mind. Oops."
3720 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
3723 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
3724 canonical form (as documented). However, after a host name lookup from
3725 the IP address, check_host() was doing a simple string comparison with
3726 addresses acquired from the DNS when checking that the found name did
3727 have the original IP as one of its addresses. Since any found IPv6
3728 addresses are likely to be in abbreviated form, the comparison could
3729 fail. Luckily, there already exists a function for doing the comparison
3730 by converting both addresses to binary, so now that is used instead of
3731 the text comparison.
3733 PH/20 There was another similar case to PH/19, when a complete host name was
3734 given in a host list; looking up its IP address could give an abbreviated
3735 form, whereas the current host's name might or might not be abbreviated.
3736 The same fix has been applied.
3742 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
3743 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
3746 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
3747 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
3749 PH/03 Typo: missing ".o" in src/pcre/Makefile.
3751 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
3752 header line, restrict the check to what is listed in RFCs 2369 and 2929.
3753 Also, for "Auto-Submitted", treat anything other than "no" as
3754 non-personal, in accordance with RFC 3834. (Previously it treated
3755 anything starting "auto-" as non-personal.)
3757 TF/01 The control=submission/name=... option had a problem with syntax
3758 errors if the name included a slash character. The /name= option
3759 now slurps the rest of the string, so it can include any characters
3760 but it must come last in the list of options (after /sender_retain
3763 PH/05 Some modifications to the interface to the fake nameserver for the new
3771 TK/01 Added the "success_on_redirect" address verification option. See
3772 NewStuff for rationale and an example.
3774 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
3776 PH/02 Patch to exigrep to allow it to work on syslog lines.
3778 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
3779 fread() to skip over the body file's header line, because in Cygwin the
3780 header line is locked and is inaccessible.
3782 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
3783 co-exist for some time) to make it clear that it is the Exim ID that is
3784 referenced, not the Message-ID: header line.
3786 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
3787 string_format() function, because snprintf() does not exist on all
3790 PH/06 The use of forbid_filter_existstest now also locks out the use of the
3791 ${stat: expansion item.
3793 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
3794 protocol synchronization error", to keep the pedants happy.
3796 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
3797 well as for IRIX systems, when gcc is being used. See the host.c source
3800 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
3802 PH/10 Named domain lists were not working if used in a queue_smtp_domains
3805 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
3806 transport and to the smtp transport in LMTP mode.
3808 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
3810 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
3811 run a filter in a subprocess. This could lead to confusion in subsequent
3812 lookups in the parent process. There should also be a search_tidyup() at
3813 the end of the subprocess.
3815 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
3816 only if the host matched helo_try_verify_hosts, which caused the
3817 verification to occur when the EHLO/HELO command was issued. The ACL just
3818 tested the remembered result. Now, if a previous verification attempt has
3819 not happened, "verify = helo" does it there and then.
3821 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
3823 TK/03 Fix log output including CR from clamd.
3825 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
3826 exist provoked a memory error which could cause a segfault.
3828 PH/15 Installed PCRE 6.2
3830 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
3832 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
3833 of the problem. Specifically, suggested +O2 rather than +O1 for the
3836 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
3838 PH/20 If a delivery was routed to a non-standard port by means of an SRV
3839 record, the port was not correctly logged when the outgoing_port log
3840 selector was set (it logged the transort's default port).
3842 PH/21 Added support for host-specific ports to manualroute, queryprogram,
3843 fallback_hosts, and "hosts" in the smtp transport.
3845 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
3846 host errors such as "Connection refused".
3848 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
3849 authentication with radiusclient 0.4.9:
3851 - Error returned from rc_read_config was caught wrongly
3852 - Username/password not passed on to radius server due to wrong length.
3854 The presumption is that some radiusclient API changes for 4.51/PH/17
3855 were not taken care of correctly. The code is still untested by me (my
3856 Linux distribution still has 0.3.2 of radiusclient), but it was
3857 contributed by a Radius user.
3859 PH/24 When doing a callout, the value of $domain wasn't set correctly when
3860 expanding the "port" option of the smtp transport.
3862 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
3863 while reading a MIME header. Thanks to Tom Hughes for a patch.
3865 PH/24 Include config.h inside local_scan.h so that configuration settings are
3868 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
3869 that in an ACL for RCPT (for example), you can examine exactly what was
3872 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
3873 commands, but it was not correctly comparing the address with the actual
3874 client host address. Thus, it would show the EHLO address in Received:
3875 header lines when this was not necessary.
3877 PH/27 Added the % operator to ${eval:}.
3879 PH/28 Exim tries to create and chdir to its spool directory when it starts;
3880 it should be ignoring failures (because with -C, for example, it has lost
3881 privilege). It wasn't ignoring creation failures other than "already
3884 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
3887 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
3888 because an input line was too long, either on its own, or by virtue of
3889 too many continuations, the temporary file was not being removed, and the
3890 return code was incorrect.
3892 PH/31 Missing "BOOL" in function definition in filtertest.c.
3894 PH/32 Applied Sieve patches from the maintainer.
3896 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
3898 PH/33 Added "verify = not_blind".
3900 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
3901 Local/Makefile (with some defaults set). These are used in built scripts
3902 such as exicyclog, but they have never been used in the exim_install
3903 script (though there are many overriding facilities there). I have
3904 arranged that the exim_install script now takes note of these two
3907 PH/35 Installed configuration files for Dragonfly.
3909 PH/36 When a locally submitted message by a trusted user did not contain a
3910 From: header, and the sender address was obtained from -f or from an SMTP
3911 MAIL command, and the trusted user did not use -F to supply a sender
3912 name, $originator_name was incorrectly used when constructing a From:
3913 header. Furthermore, $originator_name was used for submission mode
3914 messages from external hosts without From: headers in a similar way,
3915 which is clearly wrong.
3917 PH/37 Added control=suppress_local_fixups.
3919 PH/38 When log_selector = +received_sender was set, and the addition of the
3920 sender made the log line's construction buffer exactly full, or one byte
3921 less than full, an overflow happened when the terminating "\n" was
3924 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
3925 when the result of a list match is failure because a DNS lookup failed.
3927 PH/40 RM_COMMAND is now used in the building process.
3929 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
3930 the "build-* directories that it finds.
3932 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
3933 address in a domain literal was a prefix of an interface address.
3935 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
3936 when verifying a sender address, unless rewrite_headers is false.
3938 PH/44 Wrote a long comment about why errors_to addresses are verified as
3939 recipients, not senders.
3941 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
3942 the ratelimit ACL was added.
3944 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
3946 PH/46 Added extra information about PostgreSQL errors to the error string.
3948 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
3949 avoiding the need to install special zones in a real server. This is
3950 backwards compatible; if it can't find the fake resolver, it drops back.
3951 Thus, both old and new test suites can be run.
3953 TF/02 Added util/ratelimit.pl
3955 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
3956 clock is set back in time.
3958 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
3959 Candler <B.Candler@pobox.com>.
3961 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
3962 OK for addresses that are the result of redirections.
3964 PH/48 A number of further additions for the benefit of the new test suite,
3965 including a fake gethostbyname() that interfaces to the fake DNS resolver
3968 TF/06 The fix for widen_domains has also been applied to qualify_single and
3969 search_parents which are the other dnslookup options that can cause
3972 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
3975 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
3977 TF/07 Exim produced the error message "an SRV record indicated no SMTP
3978 service" if it encountered an MX record with an empty target hostname.
3979 The message is now "an MX or SRV record indicated no SMTP service".
3981 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
3982 if the DNS of the sending site is misconfigured. This is quite a
3983 common situation. This change restores the behaviour of treating a
3984 helo verification defer as a failure.
3986 PH/51 If self=fail was set on a router, the bounce message did not include the
3987 actual error message.
3993 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
3995 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
3996 command itself ended in error, the underlying message about the transport
3997 filter timeout was being overwritten with the pipe command error. Now the
3998 underlying error message should be appended to the second error message.
4000 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
4002 PH/02 Reduce the amount of output that "make" produces by default. Full output
4003 can still be requested.
4005 PH/03 The warning log line about a condition test deferring for a "warn" verb
4006 was being output only once per connection, rather than after each
4007 occurrence (because it was using the same function as for successful
4008 "warn" verbs). This seems wrong, so I have changed it.
4010 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
4011 it should not have, which might have caused a crash in the right
4012 circumstances, but probably never did.
4014 PH/04 Installed a modified version of Tony Finch's patch to make submission
4015 mode fix the return path as well as the Sender: header line, and to
4016 add a /name= option so that you can make the user's friendly name appear
4019 TF/03 Added the control = fakedefer ACL modifier.
4021 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
4022 Mark Lowes for thorough testing.
4024 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
4026 TK/03 Merged latest SRS patch from Miles Wilton.
4028 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
4029 with the definition in sysexits.h (which is #included earlier).
4030 Fortunately, Exim does not actually use EX_OK. The code used to try to
4031 preserve the sysexits.h value, by assuming that macro definitions were
4032 scanned for macro replacements. I have been disabused of this notion,
4033 so now the code just undefines EX_OK before #including unistd.h.
4035 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
4036 in the smtp transport. When a block could not be written in a single
4037 write() function, the timeout was being re-applied to each part-write.
4038 This seems wrong - if the receiver was accepting one byte at a time it
4039 would take for ever. The timeout is now adjusted when this happens. It
4040 doesn't have to be particularly precise.
4042 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
4043 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
4045 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
4046 without a domain if the check to <postmaster@domain> fails.
4048 SC/01 Eximstats: added -xls and the ability to specify output files
4049 (patch written by Frank Heydlauf).
4051 SC/02 Eximstats: use FileHandles for outputting results.
4053 SC/03 Eximstats: allow any combination of xls, txt, and html output.
4055 SC/04 Eximstats: fixed display of large numbers with -nvr option
4057 SC/05 Eximstats: fixed merging of reports with empty tables.
4059 SC/06 Eximstats: added the -include_original_destination flag
4061 SC/07 Eximstats: removed tabs and trailing whitespace.
4063 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
4065 TK/06 MBOX spool code: Add real "From " MBOX separator line
4066 so the .eml file is really in mbox format (even though
4067 most programs do not really care). Patch from Alex Miller.
4069 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
4070 The latter is generated from $received_to and is only set if the
4071 message has one envelope recipient. SA can use these headers,
4072 obviously out-of-the-box. Patch from Alex Miller.
4074 PH/08 The ${def test on a variable was returning false if the variable's
4075 value was "0", contrary to what the specification has always said!
4076 The result should be true unless the variable is empty.
4078 PH/09 The syntax error of a character other than { following "${if
4079 def:variable_name" (after optional whitespace) was not being diagnosed.
4080 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
4081 accidental colon was present, for example, could give incorrect results.
4083 PH/10 Tidied the code in a number of places where the st_size field of a stat()
4084 result is used (not including appendfile, where other changes are about
4087 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
4088 This involved changing a lot of size variables from int to off_t. It
4089 should work with maildirs and everything.
4091 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
4092 spamd dying while we are connected to it.
4094 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
4097 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
4098 function for BDB 4.3.
4100 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
4102 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
4103 the sources for PCRE have been re-arranged and more files are now
4106 PH/15 The code I had for printing potentially long long variables in PH/11
4107 above was not the best (it lost precision). The length of off_t variables
4108 is now inspected at build time, and an appropriate printing format (%ld
4109 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
4110 to be "long long int" or "long int". This is needed for the internal
4111 formatting function string_vformat().
4113 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
4114 the configuration file to be ":syslog", then the script "guesses" where
4115 the logs files are, rather than using the compiled in default. In our
4116 case the guess is not the same as the compiled default, so the script
4117 suddenly stopped working when I started to use syslog. The patch checks
4118 to see if log_file_path is "". If so, it attempts to read it from exim
4119 with no configuration file to get the compiled in version, before it
4120 falls back to the previous guessing code."
4122 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
4123 implementing BATV in an Exim configuration. See NewStuff for the gory
4126 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
4127 Makefile that are specific to HP-UX.
4129 PH/18 If the "use_postmaster" option was set for a recipient callout together
4130 with the "random" option, the postmaster address was used as the MAIL
4131 FROM address for the random test, but not for the subsequent recipient
4132 test. It is now used for both.
4134 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
4135 patch removes a few documentation additions to RFC 3028, because the
4136 latest draft now contains them. It adds the new en;ascii-case comparator
4137 and a new error check for 8bit text in MIME parts. Comparator and
4138 require names are now matched exactly. I enabled the subaddress
4139 extension, but it is not well tested yet (read: it works for me)."
4141 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
4142 rework some of the code of TK/09 above to avoid the hardwired use of
4143 "%lld" and "long long". Replaced the call to snprintf() with a call to
4146 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
4147 records point to non-existent hosts", "retry timeout exceeded", and
4148 "retry time not reached for any host after a long failure period".
4150 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
4151 experimental DomainKeys support:
4153 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
4154 (2) On an error such as an illegally used "control", the wrong name for
4155 the control was given.
4157 These problems did NOT occur unless DomainKeys support was compiled.
4159 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
4161 PH/24 Added ${if match_ip condition.
4163 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
4164 databases so that it will be absolutely obvious if a crash occurs in the
4165 DB library. This is a regular occurrence (often caused by mis-matched
4168 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
4169 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
4170 picked up on a user's system that detects such things. There doesn't seem
4171 to be a gcc warning option for this - only an attribute that has to be
4172 put on the function's prototype. It seems that in Fedora Core 4 they have
4173 set this on a number of new functions. No doubt there will be more in due
4176 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
4177 specify a transport. However, if an address that was verified by such a
4178 router was the subject of a callout, Exim crashed because it tried to
4179 read the rcpt_include_affixes from the non-existent transport. Now it
4180 just assumes that the setting of that option is false. This bug was
4181 introduced by 4.51/PH/31.
4183 PH/28 Changed -d+all to exclude +memory, because that information is very
4184 rarely of interest, but it makes the output a lot bigger. People tend to
4185 do -d+all out of habit.
4187 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
4188 code in os-type was giving problems when libc.so lives in lib64, like on
4191 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
4192 aren't the modern standard, and it seems that some systems' include files
4193 don't always have them. Exim was already checking for some of the newer
4194 ones like T_AAAA, and defining it itself. I've added checks for all the
4195 record types that Exim uses.
4197 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
4198 not automatically generating a new one, as it is supposed to. This
4199 prevented TLS from working. If the file did exist, but contained invalid
4200 data, a new version was generated, as expected. It was only the case of a
4201 non-existent file that was broken.
4203 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
4204 with a change in libdomainkeys > 0.64.
4206 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
4207 from DNS. If the selector record carries the flag, it now has
4208 precedence over the domain-wide flag.
4210 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
4212 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
4213 the use of a port name that isn't defined in /etc/services) occurred, the
4214 message was deferred as in a normal delivery, and thus remained on the
4215 spool, instead of being failed because of the mua_wrapper setting. This
4216 is now fixed, and I tidied up some of the mua_wrapper messages at the
4219 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
4220 the messages in a hash of arrays rather than using individual hashes.
4221 This is a bit cleaner and results in dramatic memory savings, albeit
4222 at a slight CPU cost.
4224 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
4225 as requested by Marc Sherman.
4227 SC/10 Eximstats: added histograms for user specified patterns as requested
4230 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
4232 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
4233 fopen() in the content-scanning modules that did not already have it.
4239 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
4240 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
4242 TK/02 Fix ACL "control" statement not being available in MIME ACL.
4244 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
4246 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
4247 to test Sieve filters that use "vacation".
4249 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
4250 that changes the way the GnuTLS parameters are stored in the cache file.
4251 The new format can be generated externally. For backward compatibility,
4252 if the data in the cache doesn't make sense, Exim assumes it has read an
4253 old-format file, and it generates new data and writes a new file. This
4254 means that you can't go back to an older release without removing the
4257 PH/03 A redirect router that has both "unseen" and "one_time" set does not
4258 work if there are any delivery delays because "one_time" forces the
4259 parent to be marked "delivered", so its unseen clone is never tried
4260 again. For this reason, Exim now forbids the simultaneous setting of
4263 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
4264 redirected to themselves ("homonym" addresses). Read the long ChangeLog
4265 entry if you want to know the details. The fix, however, neglected to
4266 consider the case when local delivery batching is involved. The test for
4267 "previously delivered" was not happening when checking to see if an
4268 address could be batched with a previous (undelivered) one; under
4269 certain circumstances this could lead to multiple deliveries to the same
4272 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
4273 in its include files, and this causes problems building Exim.
4275 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
4276 header_syntax) but Exim was just ignoring anything given after a slash.
4277 In particular, this caused confusion with an attempt to use "verify =
4278 reverse_host_lookup/defer_ok". An error is now given when options are
4279 supplied for verify items that do not have them. (Maybe reverse_host_
4280 lookup should have a defer_ok option, but that's a different point.)
4282 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
4283 defined by RFC 821) to 2048, because there were problems with some AUTH
4284 commands, and RFC 1869 says the size should be increased for extended
4285 SMTP commands that take arguments.
4287 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
4290 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
4291 "unknown" error; now it says that the functionality isn't in the binary.
4293 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
4294 an address' error message when a string expansion fails (syntax or
4295 whatever). Otherwise the password may appear in the log. Following change
4296 PH/42 below, there is no longer a chance of it appearing in a bounce
4299 PH/11 Installed exipick version 20050225.0 from John Jetmore.
4301 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
4302 of its addresses was ever tried. (Bugzilla bug #2.)
4304 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
4305 the result incorrectly in the debug output. (It correctly added a newline
4306 to what was transported.)
4308 TF/01 Added $received_time.
4310 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
4311 commented out examples of how to interface to a virus scanner and to
4312 SpamAssassin. Also added commented examples of av_scanner and
4313 spamd_address settings.
4315 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
4316 and controls are allowed in which ACLs. There were a couple of minor
4317 errors. Some of the entries in the conditions table (which is a table of
4318 where they are NOT allowed) were getting very unwieldy; rewrote them as a
4319 negation of where the condition IS allowed.
4321 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
4323 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
4324 header file does not have a version number, so I've had to invent a new
4325 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
4326 API. The code is untested by me (my Linux distribution still has 0.3.2 of
4327 radiusclient), but it was contributed by a Radius user.
4329 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
4330 files or directories, for OpenSSL.
4332 PH/19 When an Exim process that is running as root has to create an Exim log
4333 file, it does so in a subprocess that runs as exim:exim so as to get the
4334 ownership right at creation (otherwise, other Exim processes might see
4335 the file with the wrong ownership). There was no test for failure of this
4336 fork() call, which would lead to the process getting stuck as it waited
4337 for a non-existent subprocess. Forks do occasionally fail when resources
4338 run out. I reviewed all the other calls to fork(); they all seem to check
4341 PH/20 When checking for unexpected SMTP input at connect time (before writing
4342 the banner), Exim was not dealing correctly with a non-positive return
4343 from the read() function. If the client had disconnected by this time,
4344 the result was a log entry for a synchronization error with an empty
4345 string after "input=" when read() returned zero. If read() returned -1
4346 (an event I could not check), uninitialized data bytes were printed.
4347 There were reports of junk text (parts of files, etc) appearing after
4350 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
4352 PH/22 Added support for macro redefinition, and (re)definition in between
4353 driver and ACL definitions.
4355 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
4356 forgetting to use the resulting value; it was using the unexpanded value.
4358 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
4359 hadn't been configured. The fix is from Juergen Kreileder, who
4360 understands it better than I do:
4362 "Here's what I see happening with three configured cyrus_sasl
4363 authenticators configured (plain, login, cram-md5):
4365 On startup auth_cyrus_sasl_init() gets called for each of these.
4366 This means three calls to sasl_listmech() without a specified mech_list.
4367 => SASL tests which mechs of all available mechs actually work
4368 => three warnings about OTP not working
4369 => the returned list contains: plain, login, cram-md5, digest-md5, ...
4371 With the patch, sasl_listmech() also gets called three times. But now
4372 SASL's mech_list option is set to the server_mech specified in the the
4373 authenticator. Or in other words, the answer from sasl_listmech()
4374 gets limited to just the mech you're testing for (which is different
4376 => the return list contains just 'plain' or 'login', 'cram-md5' or
4377 nothing depending on the value of ob->server_mech.
4379 I've just tested the patch: Authentication still works fine,
4380 unavailable mechs specified in the exim configuration are still
4381 caught, and the auth.log warnings about OTP are gone."
4383 PH/25 When debugging is enabled, the contents of the command line are added
4384 to the debugging output, even when log_selector=+arguments is not
4387 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
4388 answer is "GNU", and only if the return is "GNU/something" is the answer
4391 PH/27 $acl_verify_message is now set immediately after the failure of a
4392 verification in an ACL, and so is available in subsequent modifiers. In
4393 particular, the message can be preserved by coding like this:
4395 warn !verify = sender
4396 set acl_m0 = $acl_verify_message
4398 Previously, $acl_verify_message was set only while expanding "message"
4399 and "log_message" when a very denied access.
4401 PH/28 Modified OS/os.c-Linux with
4403 -#ifndef OS_LOAD_AVERAGE
4404 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
4406 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
4407 nomenclature these days.)
4409 PH/29 Installed patch from the Sieve maintainer that adds the options
4410 sieve_useraddress and sieve_subaddress to the redirect router.
4412 PH/30 In these circumstances:
4413 . Two addresses routed to the same list of hosts;
4414 . First host does not offer TLS;
4415 . First host accepts first address;
4416 . First host gives temporary error to second address;
4417 . Second host offers TLS and a TLS session is established;
4418 . Second host accepts second address.
4419 Exim incorrectly logged both deliveries with the TLS parameters (cipher
4420 and peerdn, if requested) that were in fact used only for the second
4423 PH/31 When doing a callout as part of verifying an address, Exim was not paying
4424 attention to any local part prefix or suffix that was matched by the
4425 router that accepted the address. It now behaves in the same way as it
4426 does for delivery: the affixes are removed from the local part unless
4427 rcpt_include_affixes is set on the transport.
4429 PH/32 Add the sender address, as F=<...>, to the log line when logging a
4430 timeout during the DATA phase of an incoming message.
4432 PH/33 Sieve envelope tests were broken for match types other than :is. I have
4433 applied a patch sanctioned by the Sieve maintainer.
4435 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
4436 the uid or gid is negative. A case of a negative gid caused this to be
4437 noticed. The fix allows for either to be negative.
4439 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
4440 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
4443 PH/36 The change PH/12 above was broken. Fixed it.
4445 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
4446 the grounds that routing the same address twice would always produce the
4447 same answer. This might have been true once, but it is certainly no
4448 longer true now. Routing a child address may depend on the previous
4449 routing that produced that child. Some complicated redirection strategies
4450 went wrong when messages had multiple recipients, and made Exim's
4451 behaviour dependent on the order in which the addresses were given.
4453 I have moved the duplicate checking until after the routing is complete.
4454 Exim scans the addresses that are assigned to local and remote
4455 transports, and removes any duplicates. This means that more work will be
4456 done, as duplicates will always all be routed, but duplicates are
4457 presumably rare, so I don't expect this is of any significance.
4459 For deliveries to pipes, files, and autoreplies, the duplicate checking
4460 still happens during the routing process, since they are not going to be
4463 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
4464 It corrects a timeout issue with spamd. This is Ian's comment: "The
4465 background is that sometimes spamd either never reads data from a
4466 connection it has accepted, or it never writes response data. The exiscan
4467 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
4468 blindly assumes that writes won't block so it may never time out."
4470 PH/39 Allow G after quota size as well as K and M.
4472 PH/40 The value set for $authenticated_id in an authenticator may not contain
4473 binary zeroes or newlines because the value is written to log lines and
4474 to spool files. There was no check on this. Now the value is run through
4475 the string_printing() function so that such characters are converted to
4476 printable escape sequences.
4478 PH/41 $message_linecount is a new variable that contains the total number of
4479 lines in the message. Compare $body_linecount, which is the count for the
4482 PH/42 Exim no longer gives details of delivery errors for specific addresses in
4483 bounce and delay warning messages, except in certain special cases, which
4486 (a) An SMTP error message from a remote host;
4487 (b) A message specified in a :fail: redirection;
4488 (c) A message specified in a "fail" command in a system filter;
4489 (d) A message specified in a FAIL return from the queryprogram router;
4490 (e) A message specified by the cannot_route_message router option.
4492 In these cases only, Exim does include the error details in bounce and
4493 warning messages. There are also a few cases where bland messages such
4494 as "unrouteable address" or "local delivery error" are given.
4496 PH/43 $value is now also set for the "else" part of a ${run expansion.
4498 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
4499 being worked on, but at least Exim now implements the latest version to
4502 PH/45 In a pipe transport, although a timeout while waiting for the pipe
4503 process to complete was treated as a delivery failure, a timeout while
4504 writing the message to the pipe was logged, but erroneously treated as a
4505 successful delivery. Such timeouts include transport filter timeouts. For
4506 consistency with the overall process timeout, these timeouts are now
4507 treated as errors, giving rise to delivery failures by default. However,
4508 there is now a new Boolean option for the pipe transport called
4509 timeout_defer, which, if set TRUE, converts the failures into defers for
4510 both kinds of timeout. A transport filter timeout is now identified in
4513 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
4514 systems where "make" and "gmake" are different, calling "gmake" at top
4515 level broke things. I've arranged for the value of $(MAKE) to be passed
4516 from the Makefile to this script so that it can call the same version of
4520 A note about Exim versions 4.44 and 4.50
4521 ----------------------------------------
4523 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
4524 changes of various kinds. As a consequence, a big documentation update was
4525 needed. This delayed the release for rather longer than seemed good, especially
4526 in the light of a couple of (minor) security issues. Therefore, the changes
4527 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
4528 release. So 4.44 and 4.50 are in effect two different branches that both start
4531 I have left the 4.50 change log unchanged; it contains all the changes since
4532 4.43. The change log for 4.44 is below; many of its items are identical to
4533 those for 4.50. This seems to be the most sensible way to preserve the
4534 historical information.
4540 1. Minor wording change to the doc/README.SIEVE file.
4542 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
4543 computation of the current number of files was incorrect.
4545 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4546 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4549 4. Give more explanation in the error message when the command for a transport
4550 filter fails to execute.
4552 5. There are several places where Exim runs a non-Exim command in a
4553 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4554 was being done only for the command run by the queryprogram router. It is
4555 now done for all such subprocesses. The other cases are: ${run, transport
4556 filters, and the commands run by the lmtp and pipe transports.
4558 6. Added CONFIGURE_GROUP build-time option.
4560 7. Some older OS have a limit of 256 on the maximum number of file
4561 descriptors. Exim was using setrlimit() to set 1000 as a large value
4562 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4563 systems. I've change it so that if it can't get 1000, it tries for 256.
4565 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4566 was an oversight, and furthermore, ever since the addition of extra
4567 controls (e.g. 4.43/32), the checks on when to allow different forms of
4568 "control" were broken. There should now be diagnostics for all cases when a
4569 control that does not make sense is encountered.
4571 9. Added the /retain_sender option to "control=submission".
4573 10. $recipients is now available in the predata ACL (oversight).
4575 11. Tidy the search cache before the fork to do a delivery from a message
4576 received from the command line. Otherwise the child will trigger a lookup
4577 failure and thereby defer the delivery if it tries to use (for example) a
4578 cached ldap connection that the parent has called unbind on.
4580 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4581 of $address_data from the recipient verification was clobbered by the
4582 sender verification.
4584 13. The value of address_data from a sender verification is now available in
4585 $sender_address_data in subsequent conditions in the ACL statement.
4587 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
4589 15. Added a new option "connect=<time>" to callout options, to set a different
4592 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4593 was its contents. (It was OK if the option was not defined at all.)
4595 17. A "Completed" log line is now written for messages that are removed from
4596 the spool by the -Mrm option.
4598 18. New variables $sender_verify_failure and $recipient_verify_failure contain
4599 information about exactly what failed.
4601 19. Added -dd to debug only the daemon process.
4603 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
4604 handles timeouts, both on the server side and network timeouts. Renamed the
4605 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
4607 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
4608 It is now set to "smtps".
4610 22. $host_address is now set to the target address during the checking of
4611 ignore_target_hosts.
4613 23. When checking ignore_target_hosts for an ipliteral router, no host name was
4614 being passed; this would have caused $sender_host_name to have been used if
4615 matching the list had actually called for a host name (not very likely,
4616 since this list is usually IP addresses). A host name is now passed as
4619 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4620 code that specifies a non-restarting handler (typically sigaction() in
4621 modern systems) in an attempt to fix a rare and obscure crash bug.
4623 25. Narrowed the window for a race in the daemon that could cause it to ignore
4624 SIGCHLD signals. This is not a major problem, because they are used only to
4625 wake it up if nothing else does.
4627 26. A malformed maildirsize file could cause Exim to calculate negative values
4628 for the mailbox size or file count. Odd effects could occur as a result.
4629 The maildirsize information is now recalculated if the size or filecount
4632 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4633 support for a long time. Removed HAVE_SYS_VFS_H.
4635 28. Installed the latest version of exipick from John Jetmore.
4637 29. In an address list, if the pattern was not a regular expression, an empty
4638 subject address (from a bounce message) matched only if the pattern was an
4639 empty string. Non-empty patterns were not even tested. This was the wrong
4640 because it is perfectly reasonable to use an empty address as part of a
4641 database query. An empty address is now tested by patterns that are
4642 lookups. However, all the other forms of pattern expect the subject to
4643 contain a local part and a domain, and therefore, for them, an empty
4644 address still always fails if the pattern is not itself empty.
4646 30. Exim went into a mad DNS loop when attempting to do a callout where the
4647 host was specified on an smtp transport, and looking it up yielded more
4648 than one IP address.
4650 31. Re-factored the code for checking spool and log partition space into a
4651 function that finds that data and another that does the check. The former
4652 is then used to implement four new variables: $spool_space, $log_space,
4653 $spool_inodes, and $log_inodes.
4655 32. The RFC2047 encoding function was originally intended for short strings
4656 such as real names; it was not keeping to the 75-character limit for
4657 encoded words that the RFC imposes. It now respects the limit, and
4658 generates multiple encoded words if necessary. To be on the safe side, I
4659 have increased the buffer size for the ${rfc2047: expansion operator from
4662 33. It is now permitted to omit both strings after an "if" condition; if the
4663 condition is true, the result is "true". As before, when the second string
4664 is omitted, a false condition yields an empty string. This makes it less
4665 cumbersome to write custom ACL and router conditions.
4667 34. Failure to deliver a bounce message always caused it to be frozen, even if
4668 there was an errors_to setting on the router. The errors_to setting is now
4671 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4672 canonical form (fully expanded) before being placed in
4673 $sender_host_address.
4675 36. The table in the code that translates DNS record types into text (T_A to
4676 "A" for instance) was missing entries for NS and CNAME. It is just possible
4677 that this could have caused confusion if both these types were looked up
4678 for the same domain, because the text type is used as part of Exim's
4679 per-process caching. But the chance of anyone hitting this buglet seems
4682 37. The dnsdb lookup has been extended in a number of ways.
4684 (1) There is a new type, "zns", which walks up the domain tree until it
4685 finds some nameserver records. It should be used with care.
4687 (2) There is a new type, "mxh", which is like "mx" except that it returns
4688 just the host names, not the priorities.
4690 (3) It is now possible to give a list of domains (or IP addresses) to be
4691 looked up. The behaviour when one of the lookups defers can be
4692 controlled by a keyword.
4694 (4) It is now possible to specify the separator character for use when
4695 multiple records are returned.
4697 38. The dnslists ACL condition has been extended: it is now possible to supply
4698 a list of IP addresses and/or domains to be looked up in a particular DNS
4701 39. Added log_selector=+queue_time_overall.
4703 40. When running the queue in the test harness, wait just a tad after forking a
4704 delivery process, to get repeatability of debugging output.
4706 41. Include certificate and key file names in error message when GnuTLS fails
4707 to set them up, because the GnuTLS error message doesn't include the name
4708 of the failing file when there is a problem reading it.
4710 42. Allow both -bf and -bF in the same test run.
4712 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
4714 44. The "Exiscan patch" is now merged into the mainline Exim source.
4716 45. Sometimes the final signoff response after QUIT could fail to get
4717 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4718 < 0 before doing a fflush(). This bug looks as though it goes back to the
4719 introduction of TLS in release 3.20, but "sometimes" must have been rare
4720 because the tests only now provoked it.
4722 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
4723 (this can affect the format of dates).
4725 47. exim_tidydb, when checking for the continued existence of a message for
4726 which it has found a message-specific retry record, was not finding
4727 messages that were in split spool directories. Consequently, it was
4728 deleting retry records that should have stayed in existence.
4730 48. Steve fixed some bugs in eximstats.
4732 49. The SPA authentication driver was not abandoning authentication and moving
4733 on to the next authenticator when an expansion was forced to fail,
4734 contradicting the general specification for all authenticators. Instead it
4735 was generating a temporary error. It now behaves as specified.
4737 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
4738 (the order specifies the preference for clients). The order is now AES256,
4739 AES128, 3DES, ARCFOUR128.
4741 51. Small patch to Sieve code - explicitly set From: when generating an
4744 52. Exim crashed if a remote delivery caused a very long error message to be
4745 recorded - for instance if somebody sent an entire SpamAssassin report back
4746 as a large number of 550 error lines. This bug was coincidentally fixed by
4747 increasing the size of one of Exim's internal buffers (big_buffer) that
4748 happened as part of the Exiscan merge. However, to be on the safe side, I
4749 have made the code more robust (and fixed the comments that describe what
4752 53. Now that there can be additional text after "Completed" in log lines (if
4753 the queue_time_overall log selector is set), a one-byte patch to exigrep
4754 was needed to allow it to recognize "Completed" as not the last thing in
4757 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
4758 patch that reportedly fixes this has been added. I am not expert enough to
4759 create a test for it. This is what the patch creator wrote:
4761 "I found a little strange behaviour of ldap code when working with
4762 Windows 2003 AD Domain, where users was placed in more than one
4763 Organization Units. When I tried to give exim partial DN, the exit code
4764 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
4765 But simultaneously result of request was absolutely normal ldap result,
4766 so I produce this patch..."
4768 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
4769 REFERENCE, so I have modified the code to exclude the patch when that macro
4772 55. Some experimental protocols are using DNS PTR records for new purposes. The
4773 keys for these records are domain names, not reversed IP addresses. The
4774 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4775 leaves it alone. Component reversal etc. now happens only for IP addresses.
4778 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
4780 57. Double the size of the debug message buffer (to 2048) so that more of very
4781 long debug lines gets shown.
4783 58. The exicyclog utility now does better if the number of log files to keep
4784 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
4786 59. Two changes related to the smtp_active_hostname option:
4788 (1) $smtp_active_hostname is now available as a variable.
4789 (2) The default for smtp_banner uses $smtp_active_hostname instead
4790 of $primary_hostname.
4792 60. The host_aton() function is supposed to be passed a string that is known
4793 to be a valid IP address. However, in the case of IPv6 addresses, it was
4794 not checking this. This is a hostage to fortune. Exim now panics and dies
4795 if the condition is not met. A case was found where this could be provoked
4796 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4797 components; fortuitously, this particular loophole had already been fixed
4798 by change 4.50/55 above.
4800 If there are any other similar loopholes, the new check in host_aton()
4801 itself should stop them being exploited. The report I received stated that
4802 data on the command line could provoke the exploit when Exim was running as
4803 exim, but did not say which command line option was involved. All I could
4804 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4805 running as the user.
4808 61. There was a buffer overflow vulnerability in the SPA authentication code
4809 (which came originally from the Samba project). I have added a test to the
4810 spa_base64_to_bits() function which I hope fixes it.
4813 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
4814 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
4816 63. The daemon start-up calls getloadavg() while still root for those OS that
4817 need the first call to be done as root, but it missed one case: when
4818 deliver_queue_load_max is set with deliver_drop_privilege. This is
4819 necessary for the benefit of the queue runner, because there is no re-exec
4820 when deliver_drop_privilege is set.
4822 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
4823 This has been fixed.
4825 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4826 was in use, was not putting the data itself into the right store pool;
4827 consequently, it could be overwritten for a subsequent message in the same
4828 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4831 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
4833 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
4834 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
4835 the function were treating the return as a boolean value, which happened to
4836 work because 0=false and not-0=true, but is not correct code.
4838 68. The host_aton() function was not handling scoped IPv6 addresses (those
4839 with, for example, "%eth0" on the end) correctly.
4841 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
4842 negated items (that is, ~something) in unsigned ints. Some compilers
4843 apparently mutter when there is no cast.
4845 70. If an address verification called from an ACL failed, and did not produce a
4846 user-specific message (i.e. there was only a "system" message), nothing was
4847 put in $acl_verify_message. In this situation, it now puts the system
4850 71. Change 4.23/11 added synchronization checking at the start of an SMTP
4851 session; change 4.31/43 added the unwanted input to the log line - except
4852 that it did not do this in the start of session case. It now does.
4854 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
4855 This is wrong and can cause the other end to generate a synchronization
4856 error if it is another Exim or anything else that does the synchronization
4857 check. A QUIT command is no longer sent after a timeout.
4859 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
4860 during host lookups.
4862 74. The defer_ok option of callout verification was not working if it was used
4863 when verifying addresses in header lines, that is, for this case:
4865 verify = header_sender/callout=defer_ok
4867 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
4868 those file descriptors could be used for SMTP connections. If anything
4869 wrote to stderr (the example that came up was "warn" in embedded Perl), it
4870 could be sent to the SMTP client, causing chaos. The daemon now opens
4871 stdin, stdout, and stderr to /dev/null when it puts itself into the
4874 76. Arrange for output from Perl's "warn" command to be written to Exim's main
4875 log by default. The user can override this with suitable Perl magic.
4877 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
4878 the log message when discard triggers, was not working for the DATA ACL or
4879 for the non-SMTP ACL.
4881 78. Error message wording change in sieve.c.
4883 79. If smtp_accept_max_per_host was set, the number of connections could be
4884 restricted to fewer than expected, because the daemon was trying to set up
4885 a new connection before checking whether the processes handling previous
4886 connections had finished. The check for completed processes is now done
4887 earlier. On busy systems, this bug wouldn't be noticed because something
4888 else would have woken the daemon, and it would have reaped the completed
4891 80. If a message was submitted locally by a user whose login name contained one
4892 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
4893 It caused a spool format error. I have fixed the spool reading code. A
4894 related problem was that the "from" clause in the Received: line became
4895 illegal because of the space(s). It is now covered by ${quote_local_part.
4897 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
4900 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
4902 83. Patch from Sieve maintainer to fix unterminated string problem in
4903 "vacation" handling.
4905 84. Some minor changes to the Linux configuration files to help with other
4906 OS variants using glibc.
4908 85. One more patch for Sieve to update vacation handling to latest spec.
4911 ----------------------------------------------------
4912 See the note above about the 4.44 and 4.50 releases.
4913 ----------------------------------------------------
4919 1. Change 4.43/35 introduced a bug that caused file counts to be
4920 incorrectly computed when quota_filecount was set in an appendfile
4923 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4924 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4927 3. Give more explanation in the error message when the command for a transport
4928 filter fails to execute.
4930 4. There are several places where Exim runs a non-Exim command in a
4931 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4932 was being done only for the command run by the queryprogram router. It is
4933 now done for all such subprocesses. The other cases are: ${run, transport
4934 filters, and the commands run by the lmtp and pipe transports.
4936 5. Some older OS have a limit of 256 on the maximum number of file
4937 descriptors. Exim was using setrlimit() to set 1000 as a large value
4938 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4939 systems. I've change it so that if it can't get 1000, it tries for 256.
4941 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4942 was an oversight, and furthermore, ever since the addition of extra
4943 controls (e.g. 4.43/32), the checks on when to allow different forms of
4944 "control" were broken. There should now be diagnostics for all cases when a
4945 control that does not make sense is encountered.
4947 7. $recipients is now available in the predata ACL (oversight).
4949 8. Tidy the search cache before the fork to do a delivery from a message
4950 received from the command line. Otherwise the child will trigger a lookup
4951 failure and thereby defer the delivery if it tries to use (for example) a
4952 cached ldap connection that the parent has called unbind on.
4954 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4955 of $address_data from the recipient verification was clobbered by the
4956 sender verification.
4958 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4959 was its contents. (It was OK if the option was not defined at all.)
4961 11. A "Completed" log line is now written for messages that are removed from
4962 the spool by the -Mrm option.
4964 12. $host_address is now set to the target address during the checking of
4965 ignore_target_hosts.
4967 13. When checking ignore_target_hosts for an ipliteral router, no host name was
4968 being passed; this would have caused $sender_host_name to have been used if
4969 matching the list had actually called for a host name (not very likely,
4970 since this list is usually IP addresses). A host name is now passed as
4973 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4974 code that specifies a non-restarting handler (typically sigaction() in
4975 modern systems) in an attempt to fix a rare and obscure crash bug.
4977 15. Narrowed the window for a race in the daemon that could cause it to ignore
4978 SIGCHLD signals. This is not a major problem, because they are used only to
4979 wake it up if nothing else does.
4981 16. A malformed maildirsize file could cause Exim to calculate negative values
4982 for the mailbox size or file count. Odd effects could occur as a result.
4983 The maildirsize information is now recalculated if the size or filecount
4986 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4987 support for a long time. Removed HAVE_SYS_VFS_H.
4989 18. Updated exipick to current release from John Jetmore.
4991 19. Allow an empty sender to be matched against a lookup in an address list.
4992 Previously the only cases considered were a regular expression, or an
4995 20. Exim went into a mad DNS lookup loop when doing a callout where the
4996 host was specified on the transport, if the DNS lookup yielded more than
4999 21. The RFC2047 encoding function was originally intended for short strings
5000 such as real names; it was not keeping to the 75-character limit for
5001 encoded words that the RFC imposes. It now respects the limit, and
5002 generates multiple encoded words if necessary. To be on the safe side, I
5003 have increased the buffer size for the ${rfc2047: expansion operator from
5006 22. Failure to deliver a bounce message always caused it to be frozen, even if
5007 there was an errors_to setting on the router. The errors_to setting is now
5010 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
5011 canonical form (fully expanded) before being placed in
5012 $sender_host_address.
5014 24. Updated eximstats to version 1.33
5016 25. Include certificate and key file names in error message when GnuTLS fails
5017 to set them up, because the GnuTLS error message doesn't include the name
5018 of the failing file when there is a problem reading it.
5020 26. Expand error message when OpenSSL has problems setting up cert/key files.
5023 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
5024 (this can affect the format of dates).
5026 28. exim_tidydb, when checking for the continued existence of a message for
5027 which it has found a message-specific retry record, was not finding
5028 messages that were in split spool directories. Consequently, it was
5029 deleting retry records that should have stayed in existence.
5031 29. eximstats updated to version 1.35
5032 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
5033 1.35 - bugfix such that pie charts by volume are generated correctly
5035 30. The SPA authentication driver was not abandoning authentication and moving
5036 on to the next authenticator when an expansion was forced to fail,
5037 contradicting the general specification for all authenticators. Instead it
5038 was generating a temporary error. It now behaves as specified.
5040 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
5041 (the order specifies the preference for clients). The order is now AES256,
5042 AES128, 3DES, ARCFOUR128.
5044 31. Small patch to Sieve code - explicitly set From: when generating an
5047 32. Exim crashed if a remote delivery caused a very long error message to be
5048 recorded - for instance if somebody sent an entire SpamAssassin report back
5049 as a large number of 550 error lines. This bug was coincidentally fixed by
5050 increasing the size of one of Exim's internal buffers (big_buffer) that
5051 happened as part of the Exiscan merge. However, to be on the safe side, I
5052 have made the code more robust (and fixed the comments that describe what
5055 33. Some experimental protocols are using DNS PTR records for new purposes. The
5056 keys for these records are domain names, not reversed IP addresses. The
5057 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
5058 leaves it alone. Component reversal etc. now happens only for IP addresses.
5061 34. The host_aton() function is supposed to be passed a string that is known
5062 to be a valid IP address. However, in the case of IPv6 addresses, it was
5063 not checking this. This is a hostage to fortune. Exim now panics and dies
5064 if the condition is not met. A case was found where this could be provoked
5065 from a dnsdb PTR lookup with an IPv6 address that had more than 8
5066 components; fortuitously, this particular loophole had already been fixed
5067 by change 4.50/55 or 4.44/33 above.
5069 If there are any other similar loopholes, the new check in host_aton()
5070 itself should stop them being exploited. The report I received stated that
5071 data on the command line could provoke the exploit when Exim was running as
5072 exim, but did not say which command line option was involved. All I could
5073 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
5074 running as the user.
5077 35. There was a buffer overflow vulnerability in the SPA authentication code
5078 (which came originally from the Samba project). I have added a test to the
5079 spa_base64_to_bits() function which I hope fixes it.
5082 36. The daemon start-up calls getloadavg() while still root for those OS that
5083 need the first call to be done as root, but it missed one case: when
5084 deliver_queue_load_max is set with deliver_drop_privilege. This is
5085 necessary for the benefit of the queue runner, because there is no re-exec
5086 when deliver_drop_privilege is set.
5088 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
5089 was in use, was not putting the data itself into the right store pool;
5090 consequently, it could be overwritten for a subsequent message in the same
5091 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
5094 38. Sometimes the final signoff response after QUIT could fail to get
5095 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
5096 < 0 before doing a fflush(). This bug looks as though it goes back to the
5097 introduction of TLS in release 3.20, but "sometimes" must have been rare
5098 because the tests only now provoked it.
5104 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
5105 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
5106 Now it returns an integer. A number of calls were still expecting a T/F
5107 return. Fortuitously, in all cases, the tests worked in OK situations,
5108 which is the norm. However, things would have gone wrong on any write
5109 failures on the smtp file descriptor. This function is used when sending
5110 messages over SMTP and also when doing verify callouts.
5112 2. When Exim is called to do synchronous delivery of a locally submitted
5113 message (the -odf or -odi options), it no longer closes stderr before doing
5116 3. Implemented the mua_wrapper option.
5118 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
5120 5. Implemented the functions header_remove(), header_testname(),
5121 header_add_at_position(), and receive_remove_recipient(), and exported them
5124 6. If an ACL "warn" statement specified the addition of headers, Exim already
5125 inserted X-ACL-Warn: at the start if there was no header name. However, it
5126 was not making this test for the second and subsequent header lines if
5127 there were newlines in the string. This meant that an invalid header could
5128 be inserted if Exim was badly configured.
5130 7. Allow an ACL "warn" statement to add header lines at the start or after all
5131 the Received: headers, as well as at the end.
5133 8. Added the rcpt_4xx retry error code.
5135 9. Added postmaster_mailfrom=xxx to callout verification option.
5137 10. Added mailfrom=xxxx to the callout verification option, for verify=
5140 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
5141 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
5143 12. Inserted some casts to stop certain compilers warning when using pointer
5144 differences as field lengths or precisions in printf-type calls (mostly
5145 affecting debugging statements).
5147 13. Added optional readline() support for -be (dynamically loaded).
5149 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
5150 same clock tick as a message's arrival, so that its received time was the
5151 same as the "first fail" time on the retry record, and that message
5152 remained on the queue past the ultimate address timeout, every queue runner
5153 would try a delivery (because it was past the ultimate address timeout) but
5154 after another failure, the ultimate address timeout, which should have then
5155 bounced the address, did not kick in. This was a "< instead of <=" error;
5156 in most cases the first failure would have been in the next clock tick
5157 after the received time, and all would be well.
5159 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
5160 being recognized when the domain list was tested by the match_domain
5161 condition in an expansion string.
5163 16. Added the ${str2b64: operator.
5165 17. Exim was always calling setrlimit() to set a large limit for the number of
5166 processes, without checking whether the existing limit was already
5167 adequate. (It did check for the limit on file descriptors.) Furthermore,
5168 errors from getrlimit() and setrlimit() were being ignored. Now they are
5169 logged to the main and panic logs, but Exim does carry on, to try to do its
5170 job under whatever limits there are.
5172 18. Imported PCRE 5.0.
5174 19. Trivial typo in log message " temporarily refused connection" (the leading
5177 20. If the log selector return_path_on_delivery was set and an address was
5178 redirected to /dev/null, the delivery process crashed because it assumed
5179 that a return path would always be set for a "successful" delivery. In this
5180 case, the whole delivery is bypassed as an optimization, and therefore no
5183 21. Internal re-arrangement: the function for sending a challenge and reading
5184 a response while authentication was assuming a zero-terminated challenge
5185 string. It's now changed to take a pointer and a length, to allow for
5186 binary data in such strings.
5188 22. Added the cyrus_sasl authenticator (code supplied by MBM).
5190 23. Exim was not respecting finduser_retries when seeking the login of the
5191 uid under which it was called; it was always trying 10 times. (The default
5192 setting of finduser_retries is zero.) Also, it was sleeping after the final
5193 failure, which is pointless.
5195 24. Implemented tls_on_connect_ports.
5197 25. Implemented acl_smtp_predata.
5199 26. If the domain in control=submission is set empty, Exim assumes that the
5200 authenticated id is a complete email address when it generates From: or
5201 Sender: header lines.
5203 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
5204 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
5205 chown and chgrp in /bin and hostname in /usr/bin.
5207 28. Exim was keeping the "process log" file open after each use, just as it
5208 does for the main log. This opens the possibility of it remaining open for
5209 long periods when the USR1 signal hits a daemon. Occasional processlog
5210 errors were reported, that could have been caused by this. Anyway, it seems
5211 much more sensible not to leave this file open at all, so that is what now
5214 29. The long-running daemon process does not normally write to the log once it
5215 has entered its main loop, and it closes the log before doing so. This is
5216 so that log files can straightforwardly be renamed and moved. However,
5217 there are a couple of unusual error situations where the daemon does write
5218 log entries, and I had neglected to close the log afterwards.
5220 30. The text of an SMTP error response that was received during a remote
5221 delivery was being truncated at 512 bytes. This is too short for some of
5222 the long messages that one sometimes sees. I've increased the limit to
5225 31. It is now possible to make retry rules that apply only when a message has a
5226 specific sender, in particular, an empty sender.
5228 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
5229 it possible to be selective about when SMTP synchronization is enforced.
5231 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
5233 32. Implemented hosts_connection_nolog.
5235 33. Added an ACL for QUIT.
5237 34. Setting "delay_warning=" to disable warnings was not working; it gave a
5240 35. Added mailbox_size and mailbox_filecount to appendfile.
5242 36. Added control = no_multiline_responses to ACLs.
5244 37. There was a bug in the logic of the code that waits for the clock to tick
5245 in the case where the clock went backwards by a substantial amount such
5246 that the microsecond fraction of "now" was more than the microsecond
5247 fraction of "then" (but the whole seconds number was less).
5249 38. Added support for the libradius Radius client library this is found on
5250 FreeBSD (previously only the radiusclient library was supported).
5256 1. When certain lookups returned multiple values in the form name=value, the
5257 quoting of the values was not always being done properly. Specifically:
5258 (a) If the value started with a double quote, but contained no whitespace,
5260 (b) If the value contained whitespace other than a space character (i.e.
5261 tabs or newlines or carriage returns) it was not quoted.
5262 This fix has been applied to the mysql and pgsql lookups by writing a
5263 separate quoting function and calling it from the lookup code. The fix
5264 should probably also be applied to nisplus, ibase and oracle lookups, but
5265 since I cannot test any of those, I have not disturbed their existing code.
5267 2. A hit in the callout cache for a specific address caused a log line with no
5268 reason for rejecting RCPT. Now it says "Previous (cached) callout
5269 verification failure".
5271 3. There was an off-by-one bug in the queryprogram router. An over-long
5272 return line was truncated at 256 instead of 255 characters, thereby
5273 overflowing its buffer with the terminating zero. As well as fixing this, I
5274 have increased the buffer size to 1024 (and made a note to document this).
5276 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
5277 when Exim is waiting for an SMTP response from a remote server, Exim
5278 restarts its select() call on the socket, thereby resetting its timeout.
5279 This is not a problem when such interrupts are rare. Somebody set up a cron
5280 job to run exiwhat every 2 minutes, which is less than the normal select()
5281 timeout (5 or 10 minutes). This meant that the select() timeout never
5282 kicked in because it was always reset. I have fixed this by comparing the
5283 time when an interrupt arrives with the time at the start of the first call
5284 to select(). If more time than the timeout has elapsed, the interrupt is
5285 treated as a timeout.
5287 5. Some internal re-factoring in preparation for the addition of Sieve
5288 extensions (by MH). In particular, the "personal" test is moved to a
5289 separate function, and given an option for scanning Cc: and Bcc: (which is
5290 not set for Exim filters).
5292 6. When Exim created an email address using the login of the caller as the
5293 local part (e.g. when creating a From: or Sender: header line), it was not
5294 quoting the local part when it contained special characters such as @.
5296 7. Installed new OpenBSD configuration files.
5298 8. Reworded some messages for syntax errors in "and" and "or" conditions to
5299 try to make them clearer.
5301 9. Callout options, other than the timeout value, were being ignored when
5302 verifying sender addresses in header lines. For example, when using
5304 verify = header_sender/callout=no_cache
5306 the cache was (incorrectly) being used.
5308 10. Added a missing instance of ${EXE} to the exim_install script; this affects
5309 only the Cygwin environment.
5311 11. When return_path_on_delivery was set as a log selector, if different remote
5312 addresses in the same message used different return paths and parallel
5313 remote delivery occurred, the wrong values would sometimes be logged.
5314 (Whenever a remote delivery process finished, the return path value from
5315 the most recently started remote delivery process was logged.)
5317 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
5318 lines when AUTH and/or TLS are in use. This is the "received protocol"
5319 field. Exim used to use "asmtp" for authenticated SMTP, without any
5320 indication (in the protocol name) for TLS use. Now it follows the RFC and
5321 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
5322 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
5323 names appear in log lines as well as in Received: header lines.
5325 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
5326 extensions, and comparison tests, and to fix some bugs.
5328 14. Changes to the "personal" filter test:
5330 (1) The test was buggy in that it was just doing the equivalent of
5331 "contains" tests on header lines. For example, if a user's address was
5332 anne@some.where, the "personal" test would incorrectly be true for
5334 To: susanne@some.where
5336 This test is now done by extracting each address from the header in turn,
5337 and checking the entire address. Other tests that are part of "personal"
5338 are now done using regular expressions (for example, to check local parts
5339 of addresses in From: header lines).
5341 (2) The list of non-personal local parts in From: addresses has been
5342 extended to include "listserv", "majordomo", "*-request", and "owner-*",
5343 taken from the Sieve specification recommendations.
5345 (3) If the message contains any header line starting with "List-" it is
5346 treated as non-personal.
5348 (4) The test for "circular" in the Subject: header line has been removed
5349 because it now seems ill-conceived.
5351 15. Minor typos in src/EDITME comments corrected.
5353 16. Installed latest exipick from John Jetmore.
5355 17. If headers_add on a router specified a text string that was too long for
5356 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
5357 of string_sprintf() is now avoided.
5359 18. $message_body_size was not set (it was always zero) when running the DATA
5360 ACL and the local_scan() function.
5362 19. For the "mail" command in an Exim filter, no default was being set for
5363 the once_repeat time, causing a random time value to be used if "once" was
5364 specified. (If the value happened to be <= 0, no repeat happened.) The
5365 default is now 0s, meaning "never repeat". The "vacation" command was OK
5366 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
5367 (I found it when inspecting the code).
5369 20. There is now an overall timeout for performing a callout verification. It
5370 defaults to 4 times the callout timeout, which applies to individual SMTP
5371 commands during the callout. The overall timeout applies when there is more
5372 than one host that can be tried. The timeout is checked before trying the
5373 next host. This prevents very long delays if there are a large number of
5374 hosts and all are timing out (e.g. when the network connections are timing
5375 out). The value of the overall timeout can be changed by specifying an
5376 additional sub-option for "callout", called "maxwait". For example:
5378 verify = sender/callout=5s,maxwait=20s
5380 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
5381 to the end before writing, but this should make it even safer).
5383 22. Exim was forgetting that it had advertised PIPELINING for the second and
5384 subsequent messages on an SMTP connection. It was also not resetting its
5385 memory on STARTTLS and an internal HELO.
5387 23. When Exim logs an SMTP synchronization error within a session, it now
5388 records whether PIPELINING has been advertised or not.
5390 24. Added 3 instances of "(long int)" casts to time_t variables that were being
5391 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
5392 rather than long int.
5394 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
5396 26. Added the never_mail option to autoreply.
5402 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
5403 crash if the getsockname() call failed; this can happen if a connection is
5404 closed very soon after it is established. The problem was simply in the
5405 order in which certain operations were done, causing Exim to try to write
5406 to the SMTP stream before it had set up the file descriptor. The bug has
5407 been fixed by making things happen in the correct order.
5413 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
5414 before the connection was closed, thus losing the rejection response.
5416 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
5417 some early Solaris releases, but causes trouble in current releases where
5418 socklen_t is defined.
5420 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
5423 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
5426 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
5427 that releasing the top of it at the end releases what was used for sub-
5428 expansions (unless the block got too big). However, discard this block if
5429 the first thing is a variable or header, so that we can use its block when
5430 it is dynamic (useful for very large $message_headers, for example).
5432 6. Lookups now cache *every* query, not just the most recent. A new, separate
5433 store pool is used for this. It can be recovered when all lookup caches are
5434 flushed. Lookups now release memory at the end of their result strings.
5435 This has involved some general refactoring of the lookup sources.
5437 7. Some code has been added to the store_xxx() functions to reduce the amount
5438 of flapping under certain conditions.
5440 8. log_incoming_interface used to affect only the <= reception log lines. Now
5441 it causes the local interface and port to be added to several more SMTP log
5442 lines, for example "SMTP connection from", and rejection lines.
5444 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
5446 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
5448 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
5449 could be overwritten at the end of the current message (or the start of a
5450 new message if it was set in a HELO ACL). The value is now preserved for
5451 the duration of the SMTP connection.
5453 12. If a transport had a headers_rewrite setting, and a matching header line
5454 contained an unqualified address, that address was qualified, even if it
5455 did not match any rewriting rules. The underlying bug was that the values
5456 of the flags that permit the existence of unqualified sender and recipient
5457 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
5458 non-local messages, and by -bnq for local messages) were not being
5459 preserved with the message after it was received.
5461 13. When Exim was logging an SMTP synchronization error, it could sometimes log
5462 "next input=" as part of the text comprising the host identity instead of
5463 the correct text. The code was using the same buffer for two different
5464 strings. However, depending on which order the printing function evaluated
5465 its arguments, the bug did not always show up. Under Linux, for example, my
5466 test suite worked just fine.
5468 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
5469 doesn't work with some older versions of Perl. It has been changed to "my",
5470 which in any case is probably the better facility to use.
5472 15. A really picky compiler found some instances of statements for creating
5473 error messages that either had too many or two few arguments for the format
5476 16. The size of the buffer for calls to the DNS resolver has been increased
5477 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
5478 for addresses that have a lot of PTR records. This alleviates a problem; it
5479 does not fully solve it.
5481 17. A dnsdb lookup for PTR records that receives more data than will fit in the
5482 buffer now truncates the list and logs the incident, which is the same
5483 action as happens when Exim is looking up a host name and its aliases.
5484 Previously in this situation something unpredictable would happen;
5485 sometimes it was "internal error: store_reset failed".
5487 18. If a server dropped the connection unexpectedly when an Exim client was
5488 using GnuTLS and trying to read a response, the client delivery process
5489 crashed while trying to generate an error log message.
5491 19. If a "warn" verb in an ACL added multiple headers to a message in a single
5492 string, for example:
5494 warn message = H1: something\nH2: something
5496 the text was added as a single header line from Exim's point of view
5497 though it ended up OK in the delivered message. However, searching for the
5498 second and subsequent header lines using $h_h2: did not work. This has been
5499 fixed. Similarly, if a system filter added multiple headers in this way,
5500 the routers could not see them.
5502 20. Expanded the error message when iplsearch is called with an invalid key to
5503 suggest using net-iplsearch in a host list.
5505 21. When running tests using -bh, any delays imposed by "delay" modifiers in
5506 ACLs are no longer actually imposed (and a message to that effect is
5509 22. If a "gecos" field in a passwd entry contained escaped characters, in
5510 particular, if it contained a \" sequence, Exim got it wrong when building
5511 a From: or a Sender: header from that name. A second bug also caused
5512 incorrect handling when an unquoted " was present following a character
5513 that needed quoting.
5515 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
5516 was not being matched caselessly.
5518 24. Arranged for all hyphens in the exim.8 source to be escaped with
5521 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
5522 properly. Recipient callout cache records were still being keyed to include
5523 the sender, even when use_sender was set false. This led to far more
5524 callouts that were necessary. The sender is no longer included in the key
5525 when use_sender is false.
5527 26. Added "control = submission" modifier to ACLs.
5529 27. Added the ${base62d: operator to decode base 62 numbers.
5531 28. dnsdb lookups can now access SRV records.
5533 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
5534 the configuration file.
5536 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
5537 (-v) mode. This makes the output for a verbose queue run more intelligible.
5539 31. Added a use_postmaster feature to recipient callouts.
5541 32. Added the $body_zerocount variable, containing the number of binary zero
5542 bytes in the message body.
5544 33. The time of last modification of the "new" subdirectory is now used as the
5545 "mailbox time last read" when there is a quota error for a maildir
5548 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
5550 35. Added +ignore_unknown as a special item in host lists.
5552 36. Code for decoding IPv6 addresses in host lists is now included, even if
5553 IPv6 support is not being compiled. This fixes a bug in which an IPv6
5554 address was recognized as an IP address, but was then not correctly decoded
5555 into binary, causing unexpected and incorrect effects when compared with
5562 1. Very minor rewording of debugging text in manualroute to say "list of
5563 hosts" instead of "hostlist".
5565 2. If verify=header_syntax was set, and a header line with an unqualified
5566 address (no domain) and a large number of spaces between the end of the
5567 name and the colon was received, the reception process suffered a buffer
5568 overflow, and (when I tested it) crashed. This was caused by some obsolete
5569 code that should have been removed. The fix is to remove it!
5571 3. When running in the test harness, delay a bit after writing a bounce
5572 message to get a bit more predictability in the log output.
5574 4. Added a call to search_tidyup() just before forking a reception process. In
5575 theory, someone could use a lookup in the expansion of smtp_accept_max_
5576 per_host which, without the tidyup, could leave open a database connection.
5578 5. Added the variables $recipient_data and $sender_data which get set from a
5579 lookup success in an ACL "recipients" or "senders" condition, or a router
5580 "senders" option, similar to $domain_data and $local_part_data.
5582 6. Moved the writing of debug_print from before to after the "senders" test
5585 7. Change 4.31/66 (moving the time when the Received: is generated) caused
5586 problems for message scanning, either using a data ACL, or using
5587 local_scan() because the Received: header was not generated till after they
5588 were called (in order to set the time as the time of reception completion).
5589 I have revised the way this works. The header is now generated after the
5590 body is received, but before the ACL or local_scan() are called. After they
5591 are run, the timestamp in the header is updated.
5597 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
5598 before starting a queue runner without re-exec. This happened only when
5599 deliver_drop_privilege was set or when the Exim user was set to root. The
5600 effect of the bug was that timeouts during subsequent deliveries caused
5601 crashes instead of being properly handled. The handler is now left at its
5602 default (and expected) setting.
5604 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
5605 message, again when deliver_drop_privilege is set or Exim is run as root.
5606 The bug described in (1) was not present in this case, but the tidying up
5607 of the other signals was missing. I have made the two cases consistent.
5609 3. The ignore_target_hosts setting on a manualroute router was being ignored
5610 for hosts that were looked up using the /MX notation.
5612 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
5615 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
5616 operated on the sender address. After changing the $sender_address to <>
5617 for the sender address verify, Exim was re-instated it as the original
5618 (before rewriting) address, but remembering that it had rewritten it, so it
5619 wasn't rewriting it again. This bug also had the effect of breaking the
5620 sender address verification caching when the sender address was rewritten.
5622 6. The ignore_target_hosts option was being ignored by the ipliteral router.
5623 This has been changed so that if the ip literal address matches
5624 ignore_target_hosts, the router declines.
5626 7. Added expansion conditions match_domain, match_address, and match_local_
5627 part (NOT match_host).
5629 8. The placeholder for the Received: header didn't have a length field set.
5631 9. Added code to Exim itself and to exim_lock to test for a specific race
5632 condition that could lead to file corruption when using MBX delivery. The
5633 issue is with the lockfile that is created in /tmp. If this file is removed
5634 after a process has opened it but before that process has acquired a lock,
5635 there is the potential for a second process to recreate the file and also
5636 acquire a lock. This could lead to two Exim processes writing to the file
5637 at the same time. The added code performs the same test as UW imapd; it
5638 checks after acquiring the lock that its file descriptor still refers to
5639 the same named file.
5641 10. The buffer for building added header lines was of fixed size, 8192 bytes.
5642 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
5645 11. Added the smtp_active_hostname option. If used, this will typically be made
5646 to depend on the incoming interface address. Because $interface_address is
5647 not set up until the daemon has forked a reception process, error responses
5648 that can happen earlier (such as "too many connections") no longer contain
5651 12. If an expansion in a condition on a "warn" statement fails because a lookup
5652 defers, the "warn" statement is abandoned, and the next ACL statement is
5653 processed. Previously this caused the whole ACL to be aborted.
5655 13. Added the iplsearch lookup type.
5657 14. Added ident_timeout as a log selector.
5659 15. Added tls_certificate_verified as a log selector.
5661 16. Added a global option tls_require_ciphers (compare the smtp transport
5662 option of the same name). This controls incoming TLS connections.
5664 17. I finally figured out how to make tls_require_ciphers do a similar thing
5665 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
5666 before starting the TLS session.
5668 18. Tabs are now shown as \t in -bP output.
5670 19. If the log selector return_path_on_delivery was set, Exim crashed when
5671 bouncing a message because it had too many Received: header lines.
5673 20. If two routers both had headers_remove settings, and the first one included
5674 a superfluous trailing colon, the final name in the first list and the
5675 first name in the second list were incorrectly joined into one item (with a
5676 colon in the middle).
5682 1. Added -C and -D options to the exinext utility, mainly to make it easier
5683 to include in the automated testing, but these could be helpful when
5684 multiple configurations are in use.
5686 2. The exinext utility was not formatting the output nicely when there was
5687 an alternate port involved in the retry record key, nor when there was a
5688 message id as well (for retries that were specific to a specific message
5689 and a specific host). It was also confused by IPv6 addresses, because of
5690 the additional colons they contain. I have fixed the IPv4 problem, and
5691 patched it up to do a reasonable job for IPv6.
5693 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
5694 delivery, the log line now contains "pipelined" if PIPELINING was used.
5696 4. An SMTP transport process used to panic and die if the bind() call to set
5697 an explicit outgoing interface failed. This has been changed; it is now
5698 treated in the same way as a connect() failure.
5700 5. A reference to $sender_host_name in the part of a conditional expansion
5701 that was being skipped was still causing a DNS lookup. This no longer
5704 6. The def: expansion condition was not recognizing references to header lines
5705 that used bh_ and bheader_.
5707 7. Added the _cache feature to named lists.
5709 8. The code for checking quota_filecount in the appendfile transport was
5710 allowing one more file than it should have been.
5712 9. For compatibility with Sendmail, the command line option
5720 and sets the incoming protocol and host name (for trusted callers). The
5721 host name and its colon can be omitted when only the protocol is to be set.
5722 Note the Exim already has two private options, -pd and -ps, that refer to
5723 embedded Perl. It is therefore impossible to set a protocol value of "d" or
5724 "s", but I don't think that's a major issue.
5726 10. A number of refactoring changes to the code, none of which should affect
5729 (a) The number of logging options was getting close to filling up the
5730 32-bit word that was used as a bit map. I have split them into two classes:
5731 those that are passed in the argument to log_write(), and those that are
5732 only ever tested independently outside of that function. These are now in
5733 separate 32-bit words, so there is plenty of room for expansion again.
5734 There is no change in the user interface or the logging behaviour.
5736 (b) When building, for example, log lines, the code previously used a
5737 macro that called string_cat() twice, in order to add two strings. This is
5738 not really sufficiently general. Furthermore, there was one instance where
5739 it was actually wrong because one of the argument was used twice, and in
5740 one call a function was used. (As it happened, calling the function twice
5741 did not affect the overall behaviour.) The macro has been replaced by a
5742 function that can join an arbitrary number of extra strings onto a growing
5745 (c) The code for expansion conditions now uses a table and a binary chop
5746 instead of a serial search (which was left over from when there were very
5747 few conditions). Also, it now recognizes conditions like "pam" even when
5748 the relevant support is not compiled in: a suitably worded error message is
5749 given if an attempt is made to use such a condition.
5751 11. Added ${time_interval:xxxxx}.
5753 12. A bug was causing one of the ddress fields not to be passed back correctly
5754 from remote delivery subprocesses. The field in question was not being
5755 subsequently used, so this caused to problems in practice.
5757 13. Added new log selectors queue_time and deliver_time.
5759 14. Might have fixed a bug in maildirsizefile handling that threw up
5760 "unexpected character" debug warnings, and recalculated the data
5761 unnecessarily. In any case, I expanded the warning message to give more
5764 15. Added the message "Restricted characters in address" to the statements in
5765 the default ACL that block characters like @ and % in local parts.
5767 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
5768 Three changes have been made:
5770 (a) There was a serious bug; a negative response to MAIL caused the whole
5771 recipient domain to be cached as invalid, thereby blocking all messages
5772 to all local parts at the same domain, from all senders. This bug has
5773 been fixed. The domain is no longer cached after a negative response to
5774 MAIL if the sender used is not empty.
5776 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
5779 (c) A new callout option, "use_sender" has been added for people who want
5780 the modified behaviour.
5786 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
5789 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
5790 indeed breaks things for older releases.
5792 3. Added additional logging to the case where there is a problem reading data
5793 from a filter that is running in a subprocess using a pipe, in order to
5794 try to track down a specific problem.
5796 4. Testing facility fudge: when running in the test harness and attempting
5797 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
5798 getting "No route to host". Convert this to a timeout.
5800 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
5803 6. Some OS don't have socklen_t but use size_t instead. This affects the
5804 fifth argument of getsockopt() amongst other things. This is now
5805 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
5806 can be set for individual OS. I have set it for SunOS5, OSF1, and
5807 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
5808 some earlier ones do not.
5810 7. Change 4.30/15 was not doing the test caselessly.
5812 8. The standard form for an IPv6 address literal was being rejected by address
5813 parsing in, for example, MAIL and RCPT commands. An example of this kind of
5814 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5815 this, as well as the form without the "IPv6" on the front (but only when
5816 address literals are enabled, of course).
5818 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
5820 10. Exim crashed if a message with an empty sender address specified by -f
5821 encountered a router with an errors_to setting. This could be provoked only
5822 by a command such as
5826 where an empty string was supplied; "<>" did not hit this bug.
5828 11. Installed PCRE release 4.5.
5830 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
5831 remained set. It is now erased.
5833 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
5834 times from message ids (which are base 36 rather than the normal 62).
5836 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
5837 were being counted as actual protocol errors, and logged if the log
5838 selector +smtp_protocol_error was set. One cannot be perfect in this test,
5839 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
5840 and DATA following a set of rejected RCPTs do not count as protocol errors.
5841 In other words, Exim assumes they were pipelined, though this may not
5842 actually be the case. Of course, in all cases the client gets an
5843 appropriate error code.
5845 15. If a lookup fails in an ACL condition, a message about the failure may
5846 be available; it is used if testing the ACL cannot continue, because most
5847 such messages specify what the cause of the deferral is. However, some
5848 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
5849 that caused an old message to be retained and used if a later statement
5850 caused a defer, replacing the real cause of the deferral.
5852 16. If an IP address had so many PTR records that the DNS lookup buffer
5853 was not large enough to hold them, Exim could crash while trying to process
5854 the truncated data. It now detects and logs this case.
5856 17. Further to 4.21/58, another change has been made: if (and only if) the
5857 first line of a message (the first header line) ends with CRLF, a bare LF
5858 in a subsequent header line has a space inserted after it, so as not to
5859 terminate the header.
5861 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
5862 unnecessarily, used atoi() instead of strtol(), and didn't check the
5863 termination when getting file sizes from file names by regex.
5865 19. Completely re-implemented the support for maildirsize files, in the light
5866 of a number of problems with the previous contributed implementation
5867 (4.30/29). In particular:
5869 . If the quota is zero, the maildirsize file is maintained, but no quota is
5872 . If the maildir directory does not exist, it is created before any attempt
5873 to write a maildirsize file.
5875 . The quota value in the file is just a cache; if the quota is changed in
5876 the transport, the new value overrides.
5878 . A regular expression is available for excluding directories from the
5881 20. The autoreply transport checks the characters in options that define the
5882 message's headers; it allows continued headers, but it was checking with
5883 isspace() after an embedded newline instead of explicitly looking for a
5886 21. If all the "regular" hosts to which an address was routed had passed their
5887 expiry times, and had not reached their retry times, the address was
5888 bounced, even if fallback hosts were defined. Now Exim should go on to try
5891 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
5892 equivalent code in the SMTP transport. Some hosts send humungous responses
5893 to HELO/EHLO, more than 1024 it seems.
5895 23. Refactoring: code in filter.c used (void *) for "any old type" but this
5896 gives compiler warnings in some environments. I've now done it "properly",
5899 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
5900 (because of problems with the built-in one) was declared to return uschar *
5901 instead of char *, causing compiler failure.
5903 25. Fixed a file descriptor leak when processing alias/forward files.
5905 26. Fixed a minor format string issue in dbfn.c.
5907 27. Typo in exim.c: ("dmbnz" for "dbmnz").
5909 28. If a filter file refered to $h_xxx or $message_headers, and the headers
5910 contained RFC 2047 "words", Exim's memory could, under certain conditions,
5913 29. When a sender address is verified, it is cached, to save repeating the test
5914 when there is more than one recipient in a message. However, when the
5915 verification involves a callout, it is possible for different callout
5916 options to be set for different recipients. It is too complicated to keep
5917 track of this in the cache, so now Exim always runs a verification when a
5918 callout is required, relying on the callout cache for the optimization.
5919 The overhead is duplication of the address routing, but this should not be
5922 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
5923 to be verified with callout=postmaster, and the main callout worked but the
5924 postmaster check failed, the verification correctly failed. However, if a
5925 subsequent RCPT command asked for sender verification *without* the
5926 postmaster check, incorrect caching caused this verification also to fail,
5929 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
5930 it was not caching the DNS options (qualify_single, search_parents) that
5931 were used when the lookup failed. A subsequent lookup with different
5932 options therefore always gave the same answer, though there were cases
5933 where it should not have. (Example: a "domains = !$mx_any" option on a
5934 dnslookup router: the "domains" option is always processed without any
5935 widening, but the router might have qualify_single set.) Now Exim uses the
5936 cached value only when the same options are set.
5938 32. Added John Jetmore's "exipick" utility to the distribution.
5940 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
5941 than a timeout (e.g. a certificate is required, and is not provided), an
5942 Exim server now closes the connection immediately. Previously it waited for
5943 the client to close - but if the client is SSL, it seems that they each
5944 wait for each other, leading to a delay before one of them times out.
5946 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
5947 maintained 0.8.x compatibility because I don't think many are using it, and
5948 it is clearly obsolete.
5950 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
5953 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
5954 client certificate was expired. A simple patch fixes this, though I don't
5955 understand the full logic of why the verify callback is called multiple
5958 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
5959 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
5960 which causes problems with some clients (such as the Certicom SSL Plus
5961 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
5962 disables the coutermeasure allowing Eudora to connect."
5964 38. Exim was not checking that a write() to a log file succeeded. This could
5965 lead to Bad Things if a log got too big, in particular if it hit a file
5966 size limit. Exim now panics and dies if it cannot write to a log file, just
5967 as it does if it cannot open a log file.
5969 39. Modified OS/Makefile-Linux so that it now contains
5971 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
5973 The two -D definitions ensure that Exim is compiled with large file
5974 support, which makes it possible to handle log files that are bigger than
5977 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
5978 instance) a domain was checked against a named list that involved a lookup,
5979 causing $domain_data to be set, then another domain was checked against the
5980 same list, then the first domain was re-checked, the value of $domain_data
5981 after the final check could be wrong. In particular, if the second check
5982 failed, it could be set empty. This bug probably also applied to
5985 41. The strip_trailing_dot option was not being applied to the address given
5986 with the -f command-line option.
5988 42. The code for reading a message's header from the spool was incrementing
5989 $received_count, but never initializing it. This meant that the value was
5990 incorrect (doubled) while delivering a message in the same process in which
5991 it was received. In the most common configuration of Exim, this never
5992 happens - a fresh exec is done - but it can happen when
5993 deliver_drop_privilege is set.
5995 43. When Exim logs an SMTP synchronization error - client data sent too soon -
5996 it now includes up to 150 characters of the unexpected data in the log
5999 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
6000 and building data strings. The size of both of these buffers was 10 000
6001 bytes - far larger than anybody would *ever* want, thought I. Needless to
6002 say, somebody hit the limit. I have increased the maximum line length to
6003 20 000 and the maximum data length of concatenated lines to 100 000. I have
6004 also fixed two bugs, because there was no checking on these buffers. Tsk,
6005 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
6006 buffer is too small.
6008 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
6009 lsearch lookups. Now it does.
6011 46. When parsing a route_list item in a manualroute router, a fixed-length
6012 buffer was used for the list of hosts. I made this 1024 bytes long,
6013 thinking that nobody would ever have a list of hosts that long. Wrong.
6014 Somebody had a whole pile of complicated expansion conditions, and the
6015 string was silently truncated, leading to an expansion error. It turns out
6016 that it is easier to change to an unlimited length (owing to other changes
6017 that have happened since this code was originally written) than to build
6018 structure for giving a limitation error. The length of the item that
6019 expands into the list of hosts is now unlimited.
6021 47. The lsearch lookup could not handle data where the length of text line was
6022 more than 4095 characters. Such lines were truncated, leading to shortened
6023 data being returned. It should now handle lines of any length.
6025 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
6026 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
6029 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
6030 replacing strings such as BIN_DIRECTORY in a source file: the replacement
6031 no longer happens in comment lines. A list of replacements is now placed
6032 at the head of all of the source files, except those whose only change is
6033 to replace PERL_COMMAND in the very first #! line.
6035 50. Replaced the slow insertion sort in queue.c, for sorting the list of
6036 messages on the queue, with a bottom-up merge sort, using code contributed
6037 by Michael Haardt. This should make operations like -bp somewhat faster on
6038 large queues. It won't affect queue runners, except when queue_run_in_order
6041 51. Installed eximstats 1.31 in the distribution.
6043 52. Added support for SRV lookups to the dnslookup router.
6045 53. If an ACL referred to $message_body or $message_body_end, the value was not
6046 reset for any messages that followed in the same SMTP session.
6048 54. The store-handling optimization for building very long strings was not
6049 differentiating between the different store pools. I don't think this
6050 actually made any difference in practice, but I've tidied it.
6052 55. While running the routers to verify a sender address, $sender_address
6053 was still set to the sender address. This is wrong, because when routing to
6054 send a bounce to the sender, it would be empty. Therefore, I have changed
6055 it so that, while verifying a sender address, $sender_address is set to <>.
6056 (There is no change to what happens when verifying a recipient address.)
6058 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
6059 target A or AAAA records (if not already returned) without resetting the
6060 qualify_single or search_parents options of the DNS resolver. These are
6061 inappropriate in this case because the targets of MX and SRV records must
6062 be FQDNs. A broken DNS record could cause trouble if it happened to have a
6063 target that, when qualified, matched something in the local domain. These
6064 two options are now turned off when doing these lookups.
6066 57. It seems that at least some releases of Reiserfs (which does not have the
6067 concept of a fixed number of inodes) returns zero and not -1 for the
6068 number of available inodes. This interacted badly with check_spool_inodes,
6069 which assumed that -1 was the "no such thing" setting. What I have done is
6070 to check that the total number of inodes is greater than zero before doing
6071 the test of how many are available.
6073 58. When a "warn" ACL statement has a log_message modifier, the message is
6074 remembered, and not repeated. This is to avoid a lot of repetition when a
6075 message has many recipients that cause the same warning to be written.
6076 However, Exim was preserving the list of already written lines for an
6077 entire SMTP session, which doesn't seem right. The memory is now reset if a
6078 new message is started.
6080 59. The "rewrite" debugging flag was not showing the result of rewriting in the
6081 debugging output unless log_rewrite was also set.
6083 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
6084 of (int)(handle) when we know that handle contains (void *)(-1).
6086 61. The Exim daemon panic-logs an error return when it closes the incoming
6087 connection. However "connection reset by peer" seems to be common, and
6088 isn't really an error worthy of noting specially, so that particular error
6091 62. When Exim is trying to find all the local interfaces, it used to panic and
6092 die if the ioctl to get the interface flags failed. However, it seems that
6093 on at least one OS (Solaris 9) it is possible to have an interface that is
6094 included in the list of interfaces, but for which you get a failure error
6095 for this call. This happens when the interface is not "plumbed" into a
6096 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
6097 failure of the "get flags" call assumes that the interface is down.
6099 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
6100 makes life easier for people who are doing arithmetic on fields extracted
6101 from dates, where you often get leading zeros that should not be
6102 interpreted as octal.
6104 64. Added qualify_domain to the redirect router, to override the global
6107 65. If a pathologically long header line contained very many addresses (the
6108 report of this problem mentioned 10 000) and each of them was rewritten,
6109 Exim could use up a very large amount of memory. (It kept on making new
6110 copies of the header line as it rewrote, and never released the old ones.)
6111 At the expense of a bit more processing, the header rewriting function has
6112 been changed so that it no longer eats memory in this way.
6114 66. The generation of the Received: header has been moved from the time that a
6115 message starts to be received, to the time that it finishes. The timestamp
6116 in the Received: header should now be very close to that of the <= log
6117 line. There are two side-effects of this change:
6119 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
6120 logged header lines no longer include the local Received: line, because
6121 it has not yet been created. The same applies to a copy of the message
6122 that is returned to a non-SMTP sender when a message is rejected.
6124 (b) When a filter file is tested using -bf, no additional Received: header
6125 is added to the test message. After some thought, I decided that this
6128 This change does not affect the value of $received_for. It is still set
6129 after address rewriting, but before local_scan() is called.
6131 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
6133 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
6134 gave an unhelpful panic error message, and a defer error. I have managed to
6135 change this behaviour so that it now rejects any supplied certificate,
6136 which seems right, as the list of acceptable certificates is empty.
6138 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
6139 gave an unhelpful defer error. I have not managed to make this reject any
6140 supplied certificates, but the error message it gives is "no certificate
6141 supplied", which is not helpful.
6143 70. exigrep's output now also includes lines that are not associated with any
6144 message, but which match the given pattern. Implemented by a patch from
6145 Martin Sluka, which also tidied up the Perl a bit.
6147 71. Recipient callout verification, like sender verification, was using <> in
6148 the MAIL FROM command. This isn't really the right thing, since the actual
6149 sender may affect whether the remote host accepts the recipient or not. I
6150 have changed it to use the actual sender in the callout; this means that
6151 the cache record is now keyed on a recipient/sender pair, not just the
6152 recipient address. There doesn't seem to be a real danger of callout loops,
6153 since a callout by the remote host to check the sender would use <>.
6154 [SEE ABOVE: changed after hitting problems.]
6156 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
6157 temporary errors. However, in the case of such a code being given after
6158 the end of a data transmission (i.e. after ".") Exim was failing to write
6159 a retry record for the message. (Yes, there was some broken host that was
6160 actually sending 8xx at this point.)
6162 73. An unknown lookup type in a host list could cause Exim to panic-die when
6163 the list was checked. (An example that provoked this was putting <; in the
6164 middle of a list instead of at the start.) If this happened during a DATA
6165 ACL check, a -D file could be left lying around. This kind of configuration
6166 error no longer causes Exim to die; instead it causes a defer error. The
6167 incident is still logged to the main and panic logs.
6169 74. Buglet left over from Exim 3 conversion. The message "too many messages
6170 in one connection" was written to the rejectlog but not the mainlog, except
6171 when address rewriting (yes!) was being logged.
6173 75. Added write_rejectlog option.
6175 76. When a system filter was run not as root (that is, when system_filter_user
6176 was set), the values of the $n variables were not being returned to the
6177 main process; thus, they were not subsequently available in the $sn
6180 77. Added +return_path_on_delivery log selector.
6182 78. A connection timeout was being treated differently from recipients deferred
6183 when testing hosts_max_try with a message that was older than the host's
6184 retry timeout. (The host should not be counted, thus allowing all hosts to
6185 be tried at least once before bouncing.) This may have been the cause of an
6186 occasionally reported bug whereby a message would remain on the queue
6187 longer than the retry timeout, but would be bounced if a delivery was
6188 forced. I say "may" because I never totally pinned down the problem;
6189 setting up timeout/retry tests is difficult. See also the next item.
6191 79. The ultimate address timeout was not being applied to errors that involved
6192 a combination of host plus message (for example, a timeout on a MAIL
6193 command). When an address resolved to a number of possible hosts, and they
6194 were not all tried for each delivery (e.g. because of hosts_max_try), a
6195 message could remain on the queue longer than the retry timeout.
6197 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
6200 81. Fixed an obscure SMTP outgoing bug which required at least the following
6201 conditions: (a) there was another message waiting for the same server;
6202 (b) the server returned 5xx to all RCPT commands in the first message so
6203 that the message was not completed; (c) the server dropped the connection
6204 or gave a negative response to the RSET that Exim sends to abort the
6205 transaction. The observed case was a dropped connection after DATA that had
6206 been sent in pipelining mode. That is, the server had advertised PIPELINING
6207 but was not implementing it correctly. The effect of the bug was incorrect
6208 behaviour, such as trying another host, and this could lead to a crash.
6214 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
6215 and daemon.c were passed as pointers to ints; they should have been
6216 pointers to socklen_t variables (which are typically unsigned ints).
6218 2. Some signed/unsigned type warnings in the os.c file for Linux have been
6221 3. Fixed a really odd bug that affected only the testing scheme; patching a
6222 certain fixed string in the binary changed the value of another string that
6223 happened to be identical to the end of the original first string.
6225 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
6226 name", it returns that address as the IP address. On some operating
6227 systems (e.g. Solaris), it also passes back the IP address string as the
6228 "host name". However, on others (e.g. Linux), it passes back an empty
6229 string. Exim wasn't checking for this, and was changing the host name to an
6230 empty string, assuming it had been canonicalized.
6232 5. Although rare, it is permitted to have more than one PTR record for a given
6233 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
6234 all the names associated with an address, because they do in Solaris.
6235 However, it seems that they do not in Linux for data that comes from the
6236 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
6237 I found this out when I moved to a new Linux workstation and tried to run
6238 the Exim test suite.
6240 To get round this problem I have changed the code so that it now does its
6241 own call to the DNS to look up PTR records when searching for a host name.
6242 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
6243 addresses that are only in /etc/hosts are still found.
6245 This behaviour is, however, controlled by an option called host_lookup_
6246 order, which defaults to "bydns:byaddr". If people want to use the other
6247 order, or indeed, just use one or the other means of lookup, they can
6248 specify it in this variable.
6250 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
6251 some operating systems, this comes back from gethostbyaddr() as an empty
6252 string, and this is what Exim used to test for. However, it seems that in
6253 other systems, "." is yielded. Exim now tests for this case too.
6255 7. The values of check_spool_space and check_log_space are now held internally
6256 as a number of kilobytes instead of an absolute number of bytes. If a
6257 numbers is specified without 'K' or 'M', it is rounded up to the nearest
6258 kilobyte. This means that much larger values can be stored.
6260 8. Exim monitor: an attempt to get the action menu when not actually pointing
6261 at a message produces an empty menu entitled "No message selected". This
6262 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
6263 no entries in it ("Shell widget menu has zero width and/or height"). So I
6264 have added a single, blank menu entry in this case.
6266 9. Added ${quote_local_part.
6268 10. MIME decoding is now applied to the contents of Subject: header lines when
6271 11. Now that a reference to $sender_host_address automatically causes a reverse
6272 lookup to occur if necessary (4.13/18), there is no need to arrange for a
6273 host lookup before query-style lookups in lists that might use this
6274 variable. This has therefore been abolished, and the "net-" prefix is no
6275 longer necessary for query-style lookups.
6277 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
6278 have been a typo for LFLAGS, so it has been changed.
6280 13. The install script calls Exim with "-C /dev/null" in order to find the
6281 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
6282 to be output. However, since Exim outputs its version number before the
6283 error, it didn't break the script. It just looked ugly. I fixed this by
6284 always allowing "-C /dev/null" if the caller is root.
6286 14. Ignore overlarge ACL variable number when reading spool file - insurance
6287 against a later release with more variables having written the file.
6289 15. The standard form for an IPv6 address literal was being rejected by EHLO.
6290 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
6291 this, as well as the form without the "IPv6" on the front.
6293 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
6294 OS/Makefile-Darwin file.
6296 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
6297 only to LDAP libraries that do not have LDAP_OPT_DEREF.
6299 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
6300 variable. However, some OS use ints for inodes. I've added cast to long int
6301 to get rid of the compiler warning.
6303 19. I had forgotten to lock out "/../" in configuration file names when
6304 ALT_CONFIG_PREFIX was set.
6306 20. Routers used for verification do not need to specify transports. However,
6307 if such a router generated a host list, and callout was configured, Exim
6308 crashed, because it could not find a port number from the (non-existent)
6309 transport. It now assumes port 25 in this circumstance.
6311 21. Added the -t option to exigrep.
6313 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
6314 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
6315 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
6316 them is set, LOOKUP_LSEARCH is forced.
6318 23. "exim -bV" now outputs a list of lookups that are included in the binary.
6320 24. Added sender and host information to the "rejected by local_scan()" log
6321 line; previously there was no indication of these.
6323 25. Added .include_if_exists.
6325 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
6326 Linux. It turns out that not all file systems support this. Apparently some
6327 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
6328 do it.) To cope with this, the error EINVAL, which means that sync-ing is
6329 not supported on the file descriptor, is now ignored when Exim is trying to
6330 sync a directory. This applies only to Linux.
6332 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
6334 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
6335 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
6338 29. Added support for maildirsize files from supplied patch (modified a bit).
6340 30. The use of :fail: followed by an empty string could lead Exim to respond to
6341 sender verification failures with (e.g.):
6343 550 Verification failed for <xxx>
6344 550 Sender verify failed
6346 where the first response line was missing the '-' that indicates it is not
6347 the final line of the response.
6349 31. The loop for finding the name of the user that called Exim had a hardwired
6350 limit of 10; it now uses the value of finduser_retries, which is used for
6351 all other user lookups.
6353 32. Added $received_count variable, available in data and not_smtp ACLs, and at
6356 33. Exim was neglecting to zero errno before one call of strtol() when
6357 expanding a string and expecting an integer value. On some systems this
6358 resulted in spurious "integer overflow" errors. Also, it was casting the
6359 result into an int without checking.
6361 34. Testing for a connection timeout using "timeout_connect" in the retry rules
6362 did not work. The code looks as if it has *never* worked, though it appears
6363 to have been documented since at least release 1.62. I have made it work.
6365 35. The "timeout_DNS" error in retry rules, also documented since at least
6366 1.62, also never worked. As it isn't clear exactly what this means, and
6367 clearly it isn't a major issue, I have abolished the feature by treating it
6368 as "timeout", and writing a warning to the main and panic logs.
6370 36. The display of retry rules for -brt wasn't always showing the error code
6373 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
6374 timeout_connect_A, timeout_connect_MX.
6376 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
6377 to the empty sender.
6379 39. The daemon was not analysing the content of -oX till after it had closed
6380 stderr and disconnected from the controlling terminal. This meant that any
6381 syntax errors were only noted on the panic log, and the return code from
6382 the command was 0. By re-arranging the code a little, I've made the
6383 decoding happen first, so such errors now appear on stderr, and the return
6384 code is 1. However, the actual setting up of the sockets still happens in
6385 the disconnected process, so errors there are still only recorded on the
6388 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
6389 connections (as happens on some IP stacks) was logged at start up time as
6390 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
6391 it from "IPv6 and IPv4", which means that two separate sockets are being
6394 41. The debug output for gethostbyname2() or getipnodebyname() failures now
6395 says whether AF_INET or AF_INET6 was passed as an argument.
6397 42. Exiwhat output was messed up when time zones were included in log
6400 43. Exiwhat now gives more information about the daemon's listening ports,
6401 and whether -tls-on-connect was used.
6403 44. The "port" option of the smtp transport is now expanded.
6405 45. A "message" modifier in a "warn" statement in a non-message ACL was being
6406 silently ignored. Now an error message is written to the main and panic
6409 46. There's a new ACL modifier called "logwrite" which writes to a log file
6410 as soon as it is encountered.
6412 47. Added $local_user_uid and $local_user_gid at routing time.
6414 48. Exim crashed when trying to verify a sender address that was being
6417 49. Exim was recognizing only a space character after ".include". It now also
6418 recognizes a tab character.
6420 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
6421 extracting the relevant information from the specification. The man page no
6422 longer contains scrambled data for the -d option, and I've added a section
6423 at the front about calling Exim under different names.
6425 51. Added "extra_headers" argument to the "mail" command in filter files.
6427 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
6430 53. Installed eximstats 1.29.
6432 54. Added transport_filter_timeout as a generic transport option.
6434 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
6435 Cc: header lines. This was required by RFC 822, but it not required by RFC
6438 56. Exim used to add From:, Date:, and Message-Id: header lines to any
6439 incoming messages that did not have them. Now it does so only if the
6440 message originates locally, that is, if there is no associated remote host
6441 address. When Resent- header lines are present, this applies to the Resent-
6442 lines rather than the non-Resent- lines.
6444 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
6445 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
6447 58. Messages for configuration errors now include the name of the main
6448 configuration file - useful now that there may be more than one file in a
6449 list (.included file names were always shown).
6451 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
6452 for those rare installations that do not start the daemon as root or run it
6453 setuid root. I've cut out the call to initgroups() if the daemon is not
6456 60. The Exim user and group can now be bound into the binary as text strings
6457 that are looked up at the start of Exim's processing.
6459 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
6461 62. Added $mailstore_basename variable.
6463 63. Installed patch to sieve.c from Michael Haardt.
6465 64. When Exim failed to open the panic log after failing to open the main log,
6466 the original message it was trying to log was written to stderr and debug
6467 output, but if they were not available (the usual case in production), it
6468 was lost. Now it is written to syslog before the two lines that record the
6469 failures to open the logs.
6471 65. Users' Exim filters run in subprocesses under the user's uid. It is
6472 possible for a "deliver" command or an alias in a "personal" command to
6473 provoke an address rewrite. If logging of address rewriting is configured,
6474 this fails because the process is not running as root or exim. There may be
6475 a better way of dealing with this, but for the moment (because 4.30 needs
6476 to be released), I have disabled address rewrite logging when running a
6477 filter in a non-root, non-exim process.
6483 1. The buildconfig auxiliary program wasn't quoting the value set for
6484 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
6485 not defined. This bug was masked in 4.22 by the effect that was fixed in
6488 2. Some messages that were rejected after a message id was allocated were
6489 shown as "incomplete" by exigrep. It no longer does this for messages that
6490 are rejected by local_scan() or the DATA or non-SMTP ACLs.
6492 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
6493 have allow_domain_literals set, the ID did not get logged in the <= line.
6494 Domain literals are now always recognized in Message-ID: header lines.
6496 4. The first argument for a ${extract expansion item is the key name or field
6497 number. Leading and trailing spaces in this item were not being ignored,
6498 causing some misleading effects.
6500 5. When deliver_drop_privilege was set, single queue runner processes started
6501 manually (i.e. by the command "exim -q") or by the daemon (which uses the
6502 same command in the process it spins off) were not dropping privilege.
6504 6. When the daemon running as "exim" started a queue runner, it always
6505 re-executed Exim in the spun-off process. This is a waste of effort when
6506 deliver_drop_privilege is set. The new process now just calls the
6507 queue-runner function directly.
6513 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
6516 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
6517 ignored. Though the use of -oP was forcing the writing of a pid file, it
6518 was always written to the default place.
6520 3. If the message "no IP address found for host xxxx" is generated during
6521 incoming verification, it is now followed by identification of the incoming
6522 connection (so you can more easily find what provoked it).
6524 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
6526 5. Added some features to "harden" Exim a bit more against certain attacks:
6528 (a) There is now a build-time option called FIXED_NEVER_USERS that can
6529 be put in Local/Makefile. This is like the never_users runtime option,
6530 but it cannot be overridden. The default setting is "root".
6532 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
6533 prefix string with which any file named in a -C command line option
6536 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
6537 is retained for -C and -D only if the caller of Exim is root. Without
6538 it, the exim user may also use -C and -D and retain privilege.
6540 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
6541 command line option is disabled.
6543 6. Macro names set by the -D option must start with an upper case letter, just
6544 like macro names defined in the configuration file.
6546 7. Added "dereference=" facility to LDAP.
6548 8. Two instances of the typo "uknown" in the source files are fixed.
6550 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
6551 the Configure-Makefile script screwed up while processing it.
6553 10. Incorporated PCRE 4.4.
6555 11. The SMTP synchronization check was not operating right at the start of an
6556 SMTP session. For example, it could not catch a HELO sent before the client
6557 waited for the greeting. There is now a check for outstanding input at the
6558 point when the greeting is written. Because of the duplex, asynchronous
6559 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
6560 way, but not yet received, when the check is performed.
6562 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
6563 on TCP/IP sockets, because this apparently causes some broken clients to
6566 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
6567 unchanged) from the Cygwin maintainer.
6569 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
6570 was supported instead of testing for mbx. Effectively a typo.
6572 15. The spa authenticator server code was not checking that the input it
6573 received was valid base64.
6575 16. The debug output line for the "set" modifier in ACLs was not showing the
6576 name of the variable that was being set.
6578 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
6580 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
6581 lookup on its own. Something else was needed to trigger the lookup. For
6582 example, a match in host_lookup or the need for a host name in a host list.
6583 Now, if $sender_host_name is referenced and the host name has not yet been
6584 looked up, a lookup is performed. If the lookup fails, the variable remains
6585 empty, and $host_lookup_failed is set to "1".
6587 19. Added "eqi" as a case-independent comparison operator.
6589 20. The saslauthd authentication condition could segfault if neither service
6590 nor realm was specified.
6592 21. If an overflowing value such as "2048M" was set for message_size_limit, the
6593 error message that was logged was misleading, and incoming SMTP
6594 connections were dropped. The message is now more accurate, and temporary
6595 errors are given to SMTP connections.
6597 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
6598 (except RSET) with a 421 error, until QUIT is received. However, it was
6599 failing to send a response to QUIT.
6601 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
6602 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
6603 ACL is now run after the helo_try_verify_hosts code.
6605 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
6606 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
6607 case-independent, so other case variants are also recognized). Apparently
6608 some systems use these upper case variants.
6610 25. If more than two messages were waiting for the same host, and a transport
6611 filter was specified for the transport, Exim sent two messages over the
6612 same TCP/IP connection, and then failed with "socket operation on non-
6613 socket" when it tried to send the third.
6615 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
6617 27. The extern definition of crypt16() in expand.c was not being excluded when
6618 the OS had its own crypt16() function.
6620 28. Added bounce_return_body as a new option, and bounce_return_size_limit
6621 as a preferred synonym for return_size_limit, both as an option and as an
6624 29. Added LIBS=-liconv to OS/Makefile-OSF1.
6626 30. Changed the default configuration ACL to relax the local part checking rule
6627 for addresses that are not in any local domains. For these addresses,
6628 slashes and pipe symbols are allowed within local parts, but the sequence
6629 /../ is explicitly forbidden.
6631 31. SPA server authentication was not clearing the challenge buffer before
6634 32. log_message in a "warn" ACL statement was writing to the reject log as
6635 well as to the main log, which contradicts the documentation and doesn't
6636 seem right (because no rejection is happening). So I have stopped it.
6638 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
6639 However, I am unable to do any testing of this.
6641 34. Fixed an infelicity in the appendfile transport. When checking directories
6642 for a mailbox, to see if any needed to be created, it was accidentally
6643 using path names with one or more superfluous leading slashes; tracing
6644 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
6646 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
6647 contents are added to the log line that is written for every discarded
6648 recipient. (Previously a log_message setting was ignored.)
6650 36. The ${quote: operator now quotes the string if it is empty.
6652 37. The install script runs exim in order to find its version number. If for
6653 some reason other than non-existence or emptiness, which it checks, it
6654 could not run './exim', it was installing it with an empty version number,
6655 i.e. as "exim-". This error state is now caught, and the installation is
6658 38. An argument was missing from the function that creates an error message
6659 when Exim fails to connect to the socket for saslauthd authentication.
6660 This could cause Exim to crash, or give a corrupted message.
6662 39. Added isip, isip4, and isip6 to ${if conditions.
6664 40. The ACL variables $acl_xx are now saved with the message, and can be
6665 accessed later in routers, transports, and filters.
6667 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
6668 strings in the file are not string-expanded.
6670 42. If a MAIL command specified a SIZE value that was too large to fit into an
6671 int variable, the check against message_size_limit failed. Such values are
6672 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
6673 day this will have to be increased, but I don't think I want to be around
6674 when emails are that large.
6681 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
6682 iconv() is not standard in FreeBSD.
6684 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
6685 IPv6 enabled. The observed symptom was a segmentation fault on return from
6686 the function os_common_find_running_interfaces() in src/os.c.
6688 3. In the check_special_case() function in daemon.c I had used "errno" as an
6689 argument name, which causes warnings on some systems. This was basically a
6690 typo, since it was named "eno" in the comments!
6692 4. The code that waits for the clock to tick (at a resolution of some fraction
6693 of a second) so as to ensure message-id uniqueness was always waiting for
6694 at least one whole tick, when it could have waited for less. [This is
6695 almost certainly not relevant at current processor speeds, where it is
6696 unlikely to ever wait at all. But we try to future-proof.]
6698 5. The function that sleeps for a time interval that includes fractions of a
6699 second contained a race. It did not block SIGALRM between setting the
6700 timer, and suspending (a couple of lines later). If the interval was short
6701 and the sigsuspend() was delayed until after it had expired, the suspension
6702 never ended. On busy systems this could lead to processes getting stuck for
6705 6. Some uncommon configurations may cause a lookup to happen in a queue runner
6706 process, before it forks any delivery processes. The open lookup caching
6707 mechanism meant that the open file or database connection was passed into
6708 the delivery process. The problem was that delivery processes always tidy
6709 up cached lookup data. This could cause a problem for the next delivery
6710 process started by the queue runner, because the external queue runner
6711 process does not know about the closure. So the next delivery process
6712 still has data in the lookup cache. In the case of a file lookup, there was
6713 no problem because closing a file descriptor in a subprocess doesn't affect
6714 the parent. However, if the lookup was caching a connection to a database,
6715 the connection was closed, and the second delivery process was likely to
6716 see errors such as "PGSQL: query failed: server closed the connection
6717 unexpectedly". The problem has been fixed by closing all cached lookups
6718 in a queue runner before running a delivery process.
6720 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
6721 seem to have the "const" qualifier which it has on other OS. I've
6724 8. Change 4.21/2 was too strict. It is only if there are two authenticators
6725 *of the same type* (client or server) with the same public name that an
6726 error should be diagnosed.
6728 9. When Exim looked up a host name for an IP address, but failed to find the
6729 original IP address when looking up the host name (a safety check), it
6730 output the message "<ip address> does not match any IP for NULL", which was
6731 confusing, to say the least. The bug was that the host name should have
6732 appeared instead of "NULL".
6734 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
6735 user that is built into the binary, and the -C or -D options is used, root
6736 privilege is dropped before the configuration file is read. In addition,
6737 logging is switched to stderr instead of the normal log files. If the
6738 configuration then re-defines the Exim user, the unprivileged environment
6739 is probably not what is expected, so Exim logs a panic warning message (but
6742 However, if deliver_drop_privilege is set, the unprivileged state may well
6743 be exactly what is intended, so the warning has been cut out in that case,
6744 and Exim is allowed to try to write to its normal log files.
6750 1. smtp_return_error_details was not giving details for temporary sender
6751 or receiver verification errors.
6753 2. Diagnose a configuration error if two authenticators have the same public
6756 3. Exim used not to create the message log file for a message until the first
6757 delivery attempt. This could be confusing when incoming messages were held
6758 for policy or load reasons. The message log file is now created at the time
6759 the message is received, and an initial "Received" line is written to it.
6761 4. The automatically generated man page for command line options had a minor
6762 bug that caused no ill effects; however, a more serious problem was that
6763 the procedure for building the man page automatically didn't always
6764 operate. Consequently, release 4.20 contains an out-of-date version. This
6765 shouldn't happen again.
6767 5. When building Exim with embedded Perl support, the script that builds the
6768 Makefile was calling 'perl' to find its compile-time parameters, ignoring
6769 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
6771 6. The freeze_tell option was not being used for messages that were frozen on
6772 arrival, either by an ACL or by local_scan().
6774 7. Added the smtp_incomplete_transaction log selector.
6776 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
6777 was accepting AUTH without a new EHLO.
6779 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
6780 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
6783 10. Logging of TCP/IP connections (when configured) now happens in the main
6784 daemon process instead of the child process, so that the TCP/IP connection
6785 count is more accurate (but it can never be perfect).
6787 11. The use of "drop" in a nested ACL was not being handled correctly in the
6788 outer ACL. Now, if condition failure induced by the nested "drop" causes
6789 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
6790 or "require"), the connection is dropped.
6792 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
6793 that yield "discard" can now be used with an "accept" or a "discard" verb,
6794 but an error is generated for any others (because I can't see a useful way
6795 to define what should happen).
6797 13. When an ACL is read dynamically from a file (or anywhere else), the lines
6798 are now processed in the same way as lines in the Exim configuration file.
6799 In particular, continuation lines are supported.
6801 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
6803 15. Added -ti meaning -t -i.
6805 16. Check for letters, digits, hyphens, and dots in the names of dnslist
6806 domains, and warn by logging if others are found.
6808 17. At least on BSD, alignment is not guaranteed for the array of ifreq's
6809 returned from GIFCONF when Exim is trying to find the list of interfaces on
6810 a host. The code in os.c has been modified to copy each ifreq to an aligned
6811 structure in all cases.
6813 Also, in some cases, the returned ifreq's were being copied to a 'struct
6814 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
6815 means the last couple of bytes of an IPv6 address could be chopped if the
6816 ifreq contained only a normal sockaddr (14 bytes storage).
6818 18. Named domain lists were not supported in the hosts_treat_as_local option.
6819 An entry such as +xxxx was not recognized, and was treated as a literal
6822 19. Ensure that header lines added by a DATA ACL are included in the reject log
6823 if the ACL subsequently rejects the message.
6825 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
6826 MD5 (which is deprecated).
6828 21. When testing a filter file using -bf, Exim was writing a message when it
6829 took the sender from a "From " line in the message, but it was not doing so
6830 when it took $return_path from a Return-Path: header line. It now does.
6832 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
6833 with a valid header line field name (a series of printing characters
6834 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
6836 23. Changed "disc" in the source to "disk" to conform to the documentation and
6837 the book and for uniformity.
6839 24. Ignore Sendmail's -Ooption=value command line item.
6841 25. When execve() failed while trying to run a command in a pipe transport,
6842 Exim was returning EX_UNAVAILABLE (69) from the subprocess. However, this
6843 could be confused with a return value of 69 from the command itself. This
6844 has been changed to 127, the value the shell returns if it is asked to run
6845 a non-existent command. The wording for the related log line suggests a
6846 non-existent command as the problem.
6848 26. If received_header_text expands to an empty string, do not add a Received:
6849 header line to the message. (Well, it adds a token one on the spool, but
6850 marks it "old" so that it doesn't get used or transmitted.)
6852 27. Installed eximstats 1.28 (addition of -nt option).
6854 28. There was no check for failure on the call to getsockname() in the daemon
6855 code. This can fail if there is a shortage of resources on the system, with
6856 ENOMEM, for example. A temporary error is now given on failure.
6858 29. Contrary to the C standard, it seems that in some environments, the
6859 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
6860 program. Exim now does this explicitly; it affects the formatting of
6861 timestamps using strftime().
6863 30. If exiqsumm was given junk data, it threw up some uninitialized variable
6864 complaints. I've now initialized all the variables, to avoid this.
6866 32. Header lines added by a system filter were not being "seen" during
6867 transport-time rewrites.
6869 33. The info_callback() function passed to OpenSSL is set up with type void
6870 (*)(SSL *, int, int), as described somewhere. However, when calling the
6871 function (actually a macro) that sets it up, the type void(*)() is
6872 expected. I've put in a cast to prevent warnings from picky compilers.
6874 34. If a DNS black list lookup found a CNAME record, but there were no A
6875 records associated with the domain it pointed at, Exim crashed.
6877 35. If a DNS black list lookup returned more than one A record, Exim ignored
6878 all but the first. It now scans all returned addresses if a particular IP
6879 value is being sought. In this situation, the contents of the
6880 $dnslist_value variable are a list of all the addresses, separated by a
6883 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
6884 to accept a host name and all its aliases if the forward lookup for any of
6885 them yielded the IP address of the incoming connection. Now it accepts only
6886 those names whose forward lookup yields the correct IP address. Any other
6887 names are discarded. This closes a loophole whereby a rogue DNS
6888 administrator could create reverse DNS records to break through a
6889 wildcarded host restriction in an ACL.
6891 37. If a user filter or a system filter that ran in a subprocess used any of
6892 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
6893 the wrong values were passed to the pipe command ($thisaddress had the
6894 value of $0, $0 had the value of $1, etc). This bug was introduced by
6895 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
6897 38. Improved the line breaking for long SMTP error messages from ACLs.
6898 Previously, if there was no break point between 40 and 75 characters, Exim
6899 left the rest of the message alone. Two changes have been made: (a) I've
6900 reduced the minimum length to 35 characters; (b) if it can't find a break
6901 point between 35 and 75 characters, it looks ahead and uses the first one
6902 that it finds. This may give the occasional overlong line, but at least the
6903 remaining text gets split now.
6905 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
6906 file descriptors might be low, and that setting 1000 would always raise it.
6907 It turns out that in some environments, the limit is already over 1000 and
6908 that lowering it causes trouble. So now Exim takes care not to decrease it.
6910 40. When delivering a message, the value of $return_path is set to $sender_
6911 address at the start of routing (routers may change the value). By an
6912 oversight, this default was not being set up when an address was tested by
6913 -bt or -bv, which affected the outcome if any router or filter referred to
6916 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
6917 log only when "message" or "log_message" are set. However, if one of the
6918 conditions was an address verification, or a call to a nested ACL, the
6919 messages generated by the underlying test were being passed through. This
6920 no longer happens. The underlying message is available in $acl_verify_
6921 message for both "message" and "log_message" expansions, so it can be
6922 passed through if needed.
6924 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
6925 new expansion $bh_ to give the encoded byte string without charset
6926 translation. Translation happens only if iconv() is available; HAVE_ICONV
6927 indicates this at build time. HEADERS_CHARSET gives the charset to
6928 translate to; headers_charset can change it in the configuration, and
6929 "headers charset" can change it in an individual filter file.
6931 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
6932 that creates RFC 2047 encoded "words" labels them as that charset instead
6933 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
6934 expansion operator; (ii) when Exim creates a From: line for a local
6935 message; (iii) when a header line is rewritten to include a "phrase" part.
6937 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
6938 buggy, causing it to skip the first lines of messages whose message ID
6939 ended in 'D'. This would not have bitten before Exim release 4.14, because
6940 message IDs were unlikely to end in 'D' before then. The effect was to have
6941 incorrect size information for certain domains.
6943 45. #include "config.h" was missing at the start of the crypt16.c module. This
6944 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
6947 46. If there was a timeout during a "random" callout check, Exim treated it as
6948 a failure of the random address, and carried on sending RSET and the real
6949 address. If the delay was just some slowness somewhere, the response to the
6950 original RCPT would be taken as a response to RSET and so on, causing
6951 mayhem of various kinds.
6953 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
6954 when I implemented it. It didn't allow for the fact that some option values
6955 may legitimately be negative (e.g. size_addition), and it didn't even do
6956 the right test for positive values.
6958 48. Domain names in DNS records are case-independent. Exim always looks them up
6959 in lower case. Some resolvers return domain names in exactly the case they
6960 appear in the zone file, that is, they may contain uppercase letters. Not
6961 all resolvers do this - some return always lower case. Exim was treating a
6962 change of case by a resolver as a change of domain, similar to a widening
6963 of a domain abbreviation. This triggered its re-routing code and so it was
6964 trying to route what was effectively the same domain again. This normally
6965 caused routing to fail (because the router wouldn't handle the domain
6966 twice). Now Exim checks for this case specially, and just changes the
6967 casing of the domain that it ultimately uses when it transmits the message
6970 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
6973 50. If a filter generated a file delivery with a non-absolute name (possible if
6974 no home directory exists for the router), the forbid_file option was not
6977 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
6978 the existing equality matching.
6980 52. Exim was using ints instead of ino_t variables in some places where it was
6981 dealing with inode numbers.
6983 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
6984 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
6985 called TMPDIR, and if it finds it is different, it changes its value.
6987 54. The smtp_printf() function is now made available to local_scan() so
6988 additional output lines can be written before returning. There is also an
6989 smtp_fflush() function to enable the detection of a dropped connection.
6990 The variables smtp_input and smtp_batched_input are exported to
6993 55. Changed the default runtime configuration: the message "Unknown user"
6994 has been removed from the ACL, and instead placed on the localuser router,
6995 using the cannot_route_message feature. This means that any verification
6996 failures that generate their own messages won't get overridden. Similarly,
6997 the "Unrouteable address" message that was in the ACL for unverifiable
6998 relay addresses has also been removed.
7000 56. Added hosts_avoid_esmtp to the smtp transport.
7002 57. The exicyclog script was not checking for the esoteric option
7003 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
7004 will work only if exicyclog is run under the appropriate euid.
7006 58. Following a discussion on the list, the rules by which Exim recognises line
7007 endings on incoming messages have been changed. The -dropcr and drop_cr
7008 options are now no-ops, retained only for backwards compatibility. The
7009 following line terminators are recognized: LF CRLF CR. However, special
7010 processing applies to CR:
7012 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
7013 nor a local message in the state where . is a terminator.
7015 (ii) If a bare CR is encountered in a header line, an extra space is added
7016 after the line terminator so as not to end the header. The reasoning
7017 behind this is that bare CRs in header lines are most likely either
7018 to be mistakes, or people trying to play silly games.
7020 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
7021 was being incorrectly given as 18 bytes larger than it should have been.
7022 This is a VOB (very old bug).
7024 60. This may never have affected anything current, but just in case it has:
7025 When the local host is found other than at the start of a list of hosts,
7026 the local host, those with the same MX, and any that follow, are discarded.
7027 When the list in question was part of a longer list of hosts, the following
7028 hosts (not currently being processed) were also being discarded. This no
7029 longer happens. I'm not sure if this situation could ever has previously
7032 61. Added the "/MX" feature to lists of hosts in the manualroute and query
7035 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
7036 header. This is something that is recommended in a new Internet Draft, and
7037 is something that is documented as being done by Sendmail. There are two
7038 possible values. For messages generated by the autoreply transport, Exim
7041 Auto-Submitted: auto-replied
7043 whereas for all other generated messages (e.g. bounces) it adds
7045 Auto-Submitted: auto-generated
7047 63. The "personal" condition in filters now includes a test for the
7048 Auto-Submitted: header. If it contains the string "auto-" the message it
7049 not considered personal.
7051 64. Added rcpt_include_affixes as a generic transport option.
7053 65. Added queue_only_override (default true).
7055 66. Added the syslog_duplication option.
7057 67. If what should have been the first header line of a message consisted of
7058 a space followed by a colon, Exim was mis-interpreting it as a header line.
7059 It isn't of course - it is syntactically invalid and should therefore be
7060 treated as the start of the message body. The misbehaviour could have
7061 caused a number of strange effects, including loss of data in subsequent
7062 header lines, and spool format errors.
7064 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
7065 client host had authenticated. This control can now be exercised by an ACL
7066 for more flexibility.
7068 69. By default, callouts do not happen when testing with -bh. There is now a
7069 variant, -bhc, which does actually run the callout code, including
7070 consulting and updating the callout cache.
7072 70. Added support for saslauthd authentication, courtesy of Alexander
7075 71. If statvfs() failed on the spool or log directories while checking their
7076 size for availability, Exim confusingly gave the error "space shortage".
7077 Furthermore, in debugging mode it crashed with a floating point exception.
7078 These checks are done if check_{spool,log}_{space,inodes} are set, and when
7079 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
7080 serious problem, Exim now writes to the main and panic logs when this
7081 happens, with details of the failure. It then refuses to accept the
7082 incoming message, giving the message "spool directory problem" or "log
7083 directory problem" with a 421 code for SMTP messages.
7085 72. When Exim is about to re-exec itself, it ensures that the file descriptors
7086 0, 1, and 2 exist, because some OS complain for execs without them (see
7087 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
7088 descriptors. However, the code omitted to check that the open succeeded,
7089 causing mysterious errors if for some reason the permissions on /dev/null
7090 got screwed. Now Exim writes a message to the main and panic logs, and
7091 bombs out if it can't open /dev/null.
7093 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
7094 interact so that it is all more flexible. It is supposed to remain
7095 backwards compatible. Also added extra_local_interfaces.
7097 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
7098 to bomb out with an assertion failure - to the client this appears as a
7099 connection drop. This problem occurs in the part of the code that was taken
7100 from the Samba project. Fortunately, the assertion is in a very simple
7101 function, so I have fixed this by reproducing the function inline in the
7102 one place where it is called, and arranging for authentication to fail
7103 instead of killing the process with assert().
7105 75. The SPA client code was not working when the server requested OEM rather
7106 than Unicode encoding.
7108 76. Added code to make require_files with a specific uid setting more usable in
7109 the case where statting the file as root fails - usually a non-root-mounted
7110 NFS file system. When this happens and the failure is EACCES, Exim now
7111 forks a subprocess and does the per-uid checking as the relevant uid.
7113 77. Added process_log_path.
7115 78. If log_file_path was not explicitly set, a setting of check_log_space or
7116 check_log_inodes was ignored.
7118 79. If a space check for the spool or log partitions fails, the incident is now
7119 logged. Of course, in the latter case the data may get lost...
7121 80. Added the %p formatting code to string_format() so that it can be used to
7122 print addresses in debug_print(). Adjusted all the address printing in the
7123 debugging in store.c to use %p rather than %d.
7125 81. There was a concern that a line of code in smtp_in.c could overflow a
7126 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
7127 initially expressed, the concern was not well-founded, because trailing
7128 spaces are removed early. However, if the trailing spaces were followed by
7129 a NULL, they did not get removed, so the overflow was possible. Two fixes
7132 (a) I re-wrote the offending code in a cleaner fashion.
7133 (b) If an incoming SMTP command contains a NULL character, it is rejected
7136 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
7137 runs initgroups(), so that if the Exim user is in any additional groups,
7138 they will be used during message reception.
7144 The change log for 4.20 and earlier releases has been archived.
projects / exim.git / blob