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.
181 JH/01 Rework error string handling in TLS interface so that the caller in
182 more cases is responsible for logging. This permits library-sourced
183 string to be attached to addresses during delivery, and collapses
184 pairs of long lines into single ones.
186 PP/01 Allow PKG_CONFIG_PATH to be set in Local/Makefile and use it correctly
187 during configuration. Wildcards are allowed and expanded.
189 JH/02 Rework error string handling in DKIM to pass more info back to callers.
190 This permits better logging.
192 JH/03 Rework the transport continued-connection mechanism: when TLS is active,
193 do not close it down and have the child transport start it up again on
194 the passed-on TCP connection. Instead, proxy the child (and any
195 subsequent ones) for TLS via a unix-domain socket channel. Logging is
196 affected: the continued delivery log lines do not have any DNSSEC, TLS
197 Certificate or OCSP information. TLS cipher information is still logged.
199 JH/04 Shorten the log line for daemon startup by collapsing adjacent sets of
200 identical IP addresses on different listening ports. Will also affect
203 PP/02 Bug 2070: uClibc defines __GLIBC__ without providing glibc headers;
204 add noisy ifdef guards to special-case this sillyness.
205 Patch from Bernd Kuhls.
207 JH/05 Tighten up the checking in isip4 (et al): dotted-quad components larger
208 than 255 are no longer allowed.
210 JH/06 Default openssl_options to include +no_ticket, to reduce load on peers.
211 Disable the session-cache too, which might reduce our load. Since we
212 currrectly use a new context for every connection, both as server and
213 client, there is no benefit for these.
214 GnuTLS appears to not support tickets server-side by default (we don't
215 call gnutls_session_ticket_enable_server()) but client side is enabled
216 by default on recent versions (3.1.3 +) unless the PFS priority string
219 PP/03 Add $SOURCE_DATE_EPOCH support for reproducible builds, per spec at
220 <https://reproducible-builds.org/specs/source-date-epoch/>.
222 JH/07 Fix smtp transport use of limited max_rcpt under mua_wrapper. Previously
223 the check for any unsuccessful recipients did not notice the limit, and
224 erroneously found still-pending ones.
226 JH/08 Pipeline CHUNKING command and data together, on kernels that support
227 MSG_MORE. Only in-clear (not on TLS connections).
229 JH/09 Avoid using a temporary file during transport using dkim. Unless a
230 transport-filter is involved we can buffer the headers in memory for
231 creating the signature, and read the spool data file once for the
232 signature and again for transmission.
234 JH/10 Enable use of sendfile in Linux builds as default. It was disabled in
235 4.77 as the kernel support then wasn't solid, having issues in 64bit
236 mode. Now, it's been long enough. Add support for FreeBSD also.
238 JH/11 Bug 2104: Fix continued use of a transport connection with TLS. In the
239 case where the routing stage had gathered several addresses to send to
240 a host before calling the transport for the first, we previously failed
241 to close down TLS in the old transport process before passing the TCP
242 connection to the new process. The new one sent a STARTTLS command
243 which naturally failed, giving a failed delivery and bloating the retry
244 database. Investigation and fix prototype from Wolfgang Breyha.
246 JH/12 Fix check on SMTP command input synchronisation. Previously there were
247 false-negatives in the check that the sender had not preempted a response
248 or prompt from Exim (running as a server), due to that code's lack of
249 awareness of the SMTP input buffering.
251 PP/04 Add commandline_checks_require_admin option.
252 Exim drops privileges sanely, various checks such as -be aren't a
253 security problem, as long as you trust local users with access to their
254 own account. When invoked by services which pass untrusted data to
255 Exim, this might be an issue. Set this option in main configuration
256 AND make fixes to the calling application, such as using `--` to stop
259 JH/13 Do pipelining under TLS. Previously, although safe, no advantage was
260 taken. Now take care to pack both (client) MAIL,RCPT,DATA, and (server)
261 responses to those, into a single TLS record each way (this usually means
262 a single packet). As a side issue, smtp_enforce_sync now works on TLS
265 PP/05 OpenSSL/1.1: use DH_bits() for more accurate DH param sizes. This
266 affects you only if you're dancing at the edge of the param size limits.
267 If you are, and this message makes sense to you, then: raise the
268 configured limit or use OpenSSL 1.1. Nothing we can do for older
271 JH/14 For the "sock" variant of the malware scanner interface, accept an empty
272 cmdline element to get the documented default one. Previously it was
275 JH/15 Fix a crash in the smtp transport caused when two hosts in succession
276 are unsuable for non-message-specific reasons - eg. connection timeout,
277 banner-time rejection.
279 JH/16 Fix logging of delivery remote port, when specified by router, under
282 PP/06 Repair manualroute's ability to take options in any order, even if one
283 is the name of a transport.
286 HS/01 Cleanup, prevent repeated use of -p/-oMr (CVE-2017-1000369)
288 JH/17 Change the list-building routines interface to use the expanding-string
289 triplet model, for better allocation and copying behaviour.
291 JH/18 Prebuild the data-structure for "builtin" macros, for faster startup.
292 Previously it was constructed the first time a possibly-matching string
293 was met in the configuration file input during startup; now it is done
296 JH/19 Bug 2141: Use the full-complex API for Berkeley DB rather than the legacy-
297 compatible one, to avoid the (poorly documented) possibility of a config
298 file in the working directory redirecting the DB files, possibly correpting
299 some existing file. CVE-2017-10140 assigned for BDB.
301 JH/20 Bug 2147: Do not defer for a verify-with-callout-and-random which is not
302 cache-hot. Previously, although the result was properly cached, the
303 initial verify call returned a defer.
305 JH/21 Bug 2151: Avoid using SIZE on the MAIL for a callout verify, on any but
306 the main verify for receipient in uncached-mode.
308 JH/22 Retire historical build files to an "unsupported" subdir. These are
309 defined as "ones for which we have no current evidence of testing".
311 JH/23 DKIM: enforce the DNS pubkey record "h" permitted-hashes optional field,
312 if present. Previously it was ignored.
314 JH/24 Start using specified-initialisers in C structure init coding. This is
315 a C99 feature (it's 2017, so now considered safe).
317 JH/25 Use one-bit bitfields for flags in the "addr" data structure. Previously
318 if was a fixed-sized field and bitmask ops via macros; it is now more
321 PP/07 GitHub PR 56: Apply MariaDB build fix.
322 Patch provided by Jaroslav Škarvada.
324 PP/08 Bug 2161: Fix regression in sieve quoted-printable handling introduced
325 during Coverity cleanups [4.87 JH/47]
326 Diagnosis and fix provided by Michael Fischer v. Mollard.
328 JH/26 Fix DKIM bug: when the pseudoheader generated for signing was exactly
329 the right size to place the terminating semicolon on its own folded
330 line, the header hash was calculated to an incorrect value thanks to
331 the (relaxed) space the fold became.
333 HS/02 Fix Bug 2130: large writes from the transport subprocess where chunked
334 and confused the parent.
336 JH/27 Fix SOCKS bug: an unitialized pointer was deref'd by the transport process
337 which could crash as a result. This could lead to undeliverable messages.
339 JH/28 Logging: "next input sent too soon" now shows where input was truncated
342 JH/29 Fix queue_run_in_order to ignore the PID portion of the message ID. This
343 matters on fast-turnover and PID-randomising systems, which were getting
344 out-of-order delivery.
346 JH/30 Fix a logging bug on aarch64: an unsafe routine was previously used for
347 a possibly-overlapping copy. The symptom was that "Remote host closed
348 connection in response to HELO" was logged instead of the actual 4xx
351 JH/31 Fix CHUNKING code to properly flush the unwanted chunk after an error.
352 Previously only that bufferd was discarded, resulting in SYMTP command
355 JH/32 DKIM: when a message has multiple signatures matching an identity given
356 in dkim_verify_signers, run the dkim acl once for each. Previously only
357 one run was done. Bug 2189.
359 JH/33 Downgrade an unfound-list name (usually a typo in the config file) from
360 "panic the current process" to "deliberately defer". The panic log is
361 still written with the problem list name; the mail and reject logs now
362 get a temp-reject line for the message that was being handled, saying
363 something like "domains check lookup or other defer". The SMTP 451
364 message is still "Temporary local problem".
366 JH/34 Bug 2199: Fix a use-after-free while reading smtp input for header lines.
367 A crafted sequence of BDAT commands could result in in-use memory beeing
368 freed. CVE-2017-16943.
370 HS/03 Bug 2201: Fix checking for leading-dot on a line during headers reading
371 from SMTP input. Previously it was always done; now only done for DATA
372 and not BDAT commands. CVE-2017-16944.
374 JH/35 Bug 2201: Flush received data in BDAT mode after detecting an error fatal
375 to the message (such as an overlong header line). Previously this was
376 not done and we did not exit BDAT mode. Followon from the previous item
377 though a different problem.
383 JH/01 Bug 1922: Support IDNA2008. This has slightly different conversion rules
384 than -2003 did; needs libidn2 in addition to libidn.
386 JH/02 The path option on a pipe transport is now expanded before use.
388 PP/01 GitHub PR 50: Do not call ldap_start_tls_s on ldapi:// connections.
389 Patch provided by "Björn", documentation fix added too.
391 JH/03 Bug 2003: fix Proxy Protocol v2 handling: the address size field was
392 missing a wire-to-host endian conversion.
394 JH/04 Bug 2004: fix CHUNKING in non-PIPELINEING mode. Chunk data following
395 close after a BDAT command line could be taken as a following command,
396 giving a synch failure. Fix by only checking for synch immediately
397 before acknowledging the chunk.
399 PP/02 GitHub PR 52: many spelling fixes, which include fixing parsing of
400 no_require_dnssec option and creation of _HAVE_TRANSPORT_APPEND_MAILDIR
401 macro. Patches provided by Josh Soref.
403 JH/05 Have the EHLO response advertise VRFY, if there is a vrfy ACL defined.
404 Previously we did not; the RFC seems ambiguous and VRFY is not listed
405 by IANA as a service extension. However, John Klensin suggests that we
408 JH/06 Bug 2017: Fix DKIM verification in -bh test mode. The data feed into
409 the dkim code may be unix-mode line endings rather than smtp wire-format
410 CRLF, so prepend a CR to any bare LF.
412 JH/07 Rationalise the coding for callout smtp conversations and transport ones.
413 As a side-benfit, callouts can now use PIPELINING hence fewer round-trips.
415 JH/08 Bug 2016: Fix DKIM verification vs. CHUNKING. Any BDAT commands after
416 the first were themselves being wrongly included in the feed into dkim
417 processing; with most chunk sizes in use this resulted in an incorrect
418 body hash calculated value.
420 JH/09 Bug 2014: permit inclusion of a DKIM-Signature header in a received
421 DKIM signature block, for verification. Although advised against by
422 standards it is specifically not ruled illegal.
424 JH/10 Bug 2025: Fix reception of (quoted) local-parts with embedded spaces.
426 JH/11 Bug 2029: Fix crash in DKIM verification when a message signature block is
427 missing a body hash (the bh= tag).
429 JH/12 Bug 2018: Re-order Proxy Protocol startup versus TLS-on-connect startup.
430 It seems that HAProxy sends the Proxy Protocol information in clear and
431 only then does a TLS startup, so do the same.
433 JH/13 Bug 2027: Avoid attempting to use TCP Fast Open for non-transport client
434 TCP connections (such as for Spamd) unless the daemon successfully set
435 Fast Open mode on its listening sockets. This fixes breakage seen on
436 too-old kernels or those not configured for Fast Open, at the cost of
437 requiring both directions being enabled for TFO, and TFO never being used
438 by non-daemon-related Exim processes.
440 JH/14 Bug 2000: Reject messages recieved with CHUNKING but with malformed line
441 endings, at least on the first header line. Try to canonify any that get
442 past that check, despite the cost.
444 JH/15 Angle-bracket nesting (an error inserted by broken sendmails) levels are
445 now limited to an arbitrary five deep, while parsing addresses with the
446 strip_excess_angle_brackets option enabled.
448 PP/03 Bug 2018: For Proxy Protocol and TLS-on-connect, do not over-read and
449 instead leave the unprompted TLS handshake in socket buffer for the
450 TLS library to consume.
452 PP/04 Bug 2018: Also handle Proxy Protocol v2 safely.
454 PP/05 FreeBSD compat: handle that Ports no longer create /usr/bin/perl
456 JH/16 Drop variables when they go out of scope. Memory management drops a whole
457 region in one operation, for speed, and this leaves assigned pointers
458 dangling. Add checks run only under the testsuite which checks all
459 variables at a store-reset and panics on a dangling pointer; add code
460 explicitly nulling out all the variables discovered. Fixes one known
461 bug: a transport crash, where a dangling pointer for $sending_ip_address
462 originally assigned in a verify callout, is re-used.
464 PP/06 Drop '.' from @INC in various Perl scripts.
466 PP/07 Switch FreeBSD iconv to always use the base-system libc functions.
468 PP/08 Reduce a number of compilation warnings under clang; building with
469 CC=clang CFLAGS+=-Wno-dangling-else -Wno-logical-op-parentheses
470 should be warning-free.
472 JH/17 Fix inbound CHUNKING when DKIM disabled at runtime.
474 HS/01 Fix portability problems introduced by PP/08 for platforms where
475 realloc(NULL) is not equivalent to malloc() [SunOS et al].
477 HS/02 Bug 1974: Fix missing line terminator on the last received BDAT
478 chunk. This allows us to accept broken chunked messages. We need a more
479 general solution here.
481 PP/09 Wrote util/chunking_fixqueue_finalnewlines.pl to help recover
482 already-broken messages in the queue.
484 JH/18 Bug 2061: Fix ${extract } corrupting an enclosing ${reduce } $value.
486 JH/19 Fix reference counting bug in routing-generated-address tracking.
492 JH/01 Use SIZE on MAIL FROM in a cutthrough connection, if the destination
493 supports it and a size is available (ie. the sending peer gave us one).
495 JH/02 The obsolete acl condition "demime" is removed (finally, after ten
496 years of being deprecated). The replacements are the ACLs
497 acl_smtp_mime and acl_not_smtp_mime.
499 JH/03 Upgrade security requirements imposed for hosts_try_dane: previously
500 a downgraded non-dane trust-anchor for the TLS connection (CA-style)
501 or even an in-clear connection were permitted. Now, if the host lookup
502 was dnssec and dane was requested then the host is only used if the
503 TLSA lookup succeeds and is dnssec. Further hosts (eg. lower priority
504 MXs) will be tried (for hosts_try_dane though not for hosts_require_dane)
505 if one fails this test.
506 This means that a poorly-configured remote DNS will make it incommunicado;
507 but it protects against a DNS-interception attack on it.
509 JH/04 Bug 1810: make continued-use of an open smtp transport connection
510 non-noisy when a race steals the message being considered.
512 JH/05 If main configuration option tls_certificate is unset, generate a
513 self-signed certificate for inbound TLS connections.
515 JH/06 Bug 165: hide more cases of password exposure - this time in expansions
516 in rewrites and routers.
518 JH/07 Retire gnutls_require_mac et.al. These were nonfunctional since 4.80
519 and logged a warning sing 4.83; now they are a configuration file error.
521 JH/08 Bug 1836: Fix crash in VRFY handling when handed an unqualified name
522 (lacking @domain). Apply the same qualification processing as RCPT.
524 JH/09 Bug 1804: Avoid writing msglog files when in -bh or -bhc mode.
526 JH/10 Support ${sha256:} applied to a string (as well as the previous
529 JH/11 Cutthrough: avoid using the callout hints db on a verify callout when
530 a cutthrough deliver is pending, as we always want to make a connection.
531 This also avoids re-routing the message when later placing the cutthrough
532 connection after a verify cache hit.
533 Do not update it with the verify result either.
535 JH/12 Cutthrough: disable when verify option success_on_redirect is used, and
536 when routing results in more than one destination address.
538 JH/13 Cutthrough: expand transport dkim_domain option when testing for dkim
539 signing (which inhibits the cutthrough capability). Previously only
540 the presence of an option was tested; now an expansion evaluating as
541 empty is permissible (obviously it should depend only on data available
542 when the cutthrough connection is made).
544 JH/14 Fix logging of errors under PIPELINING. Previously the log line giving
545 the relevant preceding SMTP command did not note the pipelining mode.
547 JH/15 Fix counting of empty lines in $body_linecount and $message_linecount.
548 Previously they were not counted.
550 JH/16 DANE: treat a TLSA lookup response having all non-TLSA RRs, the same
551 as one having no matching records. Previously we deferred the message
552 that needed the lookup.
554 JH/17 Fakereject: previously logged as a normal message arrival "<="; now
555 distinguished as "(=".
557 JH/18 Bug 1867: make the fail_defer_domains option on a dnslookup router work
558 for missing MX records. Previously it only worked for missing A records.
560 JH/19 Bug 1850: support Radius libraries that return REJECT_RC.
562 JH/20 Bug 1872: Ensure that acl_smtp_notquit is run when the connection drops
563 after the data-go-ahead and data-ack. Patch from Jason Betts.
565 JH/21 Bug 1846: Send DMARC forensic reports for reject and quarantine results,
566 even for a "none" policy. Patch from Tony Meyer.
568 JH/22 Fix continued use of a connection for further deliveries. If a port was
569 specified by a router, it must also match for the delivery to be
572 JH/23 Bug 1874: fix continued use of a connection for further deliveries.
573 When one of the recipients of a message was unsuitable for the connection
574 (has no matching addresses), we lost track of needing to mark it
575 deferred. As a result mail would be lost.
577 JH/24 Bug 1832: Log EHLO response on getting conn-close response for HELO.
579 JH/25 Decoding ACL controls is now done using a binary search; the source code
580 takes up less space and should be simpler to maintain. Merge the ACL
581 condition decode tables also, with similar effect.
583 JH/26 Fix problem with one_time used on a redirect router which returned the
584 parent address unchanged. A retry would see the parent address marked as
585 delivered, so not attempt the (identical) child. As a result mail would
588 JH/27 Fix a possible security hole, wherein a process operating with the Exim
589 UID can gain a root shell. Credit to http://www.halfdog.net/ for
590 discovery and writeup. Ubuntu bug 1580454; no bug raised against Exim
593 JH/28 Enable {spool,log} filesystem space and inode checks as default.
594 Main config options check_{log,spool}_{inodes,space} are now
595 100 inodes, 10MB unless set otherwise in the configuration.
597 JH/29 Fix the connection_reject log selector to apply to the connect ACL.
598 Previously it only applied to the main-section connection policy
601 JH/30 Bug 1897: fix callouts connection fallback from TLS to cleartext.
603 PP/01 Changed default Diffie-Hellman parameters to be Exim-specific, created
604 by me. Added RFC7919 DH primes as an alternative.
606 PP/02 Unbreak build via pkg-config with new hash support when crypto headers
607 are not in the system include path.
609 JH/31 Fix longstanding bug with aborted TLS server connection handling. Under
610 GnuTLS, when a session startup failed (eg because the client disconnected)
611 Exim did stdio operations after fclose. This was exposed by a recent
612 change which nulled out the file handle after the fclose.
614 JH/32 Bug 1909: Fix OCSP proof verification for cases where the proof is
615 signed directly by the cert-signing cert, rather than an intermediate
616 OCSP-signing cert. This is the model used by LetsEncrypt.
618 JH/33 Bug 1914: Ensure socket is nonblocking before draining after SMTP QUIT.
620 HS/01 Fix leak in verify callout under GnuTLS, about 3MB per recipient on
621 an incoming connection.
623 HS/02 Bug 1802: Do not half-close the connection after sending a request
626 HS/03 Use "auto" as the default EC curve parameter. For OpenSSL < 1.0.2
627 fallback to "prime256v1".
629 JH/34 SECURITY: Use proper copy of DATA command in error message.
630 Could leak key material. Remotely exploitable. CVE-2016-9963.
636 JH/01 Bug 1664: Disable OCSP for GnuTLS library versions at/before 3.3.16
637 and 3.4.4 - once the server is enabled to respond to an OCSP request
638 it does even when not requested, resulting in a stapling non-aware
639 client dropping the TLS connection.
641 TF/01 Code cleanup: Overhaul the debug_selector and log_selector machinery to
642 support variable-length bit vectors. No functional change.
644 TF/02 Improve the consistency of logging incoming and outgoing interfaces.
645 The I= interface field on outgoing lines is now after the H= remote
646 host field, same as incoming lines. There is a separate
647 outgoing_interface log selector which allows you to disable the
650 JH/02 Bug 728: Close logfiles after a daemon-process "exceptional" log write.
651 If not running log_selector +smtp_connection the mainlog would be held
652 open indefinitely after a "too many connections" event, including to a
653 deleted file after a log rotate. Leave the per net connection logging
654 leaving it open for efficiency as that will be quickly detected by the
655 check on the next write.
657 HS/01 Bug 1671: Fix post transport crash.
658 Processing the wait-<transport> messages could crash the delivery
659 process if the message IDs didn't exist for some reason. When
660 using 'split_spool_directory=yes' the construction of the spool
661 file name failed already, exposing the same netto behaviour.
663 JH/03 Bug 425: Capture substrings in $regex1, $regex2 etc from regex &
664 mime_regex ACL conditions.
666 JH/04 Bug 1686: When compiled with EXPERIMENTAL_DSN_INFO: Add extra information
667 to DSN fail messages (bounces): remote IP, remote greeting, remote response
668 to HELO, local diagnostic string.
670 JH/05 Downgrade message for a TLS-certificate-based authentication fail from
671 log line to debug. Even when configured with a tls authenticator many
672 client connections are expected to not authenticate in this way, so
673 an authenticate fail is not an error.
675 HS/02 Add the Exim version string to the process info. This way exiwhat
676 gives some more detail about the running daemon.
678 JH/06 Bug 1395: time-limit caching of DNS lookups, to the TTL value. This may
679 matter for fast-change records such as DNSBLs.
681 JH/07 Bug 1678: Always record an interface option value, if set, as part of a
682 retry record, even if constant. There may be multiple transports with
683 different interface settings and the retry behaviour needs to be kept
686 JH/08 Bug 1586: exiqgrep now refuses to run if there are unexpected arguments.
688 JH/09 Bug 1700: ignore space & tab embedded in base64 during decode.
690 JH/10 Bug 840: fix log_defer_output option of pipe transport
692 JH/11 Bug 830: use same host for all RCPTS of a message, even under
693 hosts_randomize. This matters a lot when combined with mua_wrapper.
695 JH/12 Bug 1706: percent and underbar characters are no longer escaped by the
696 ${quote_pgsql:<string>} operator.
698 JH/13 Bug 1708: avoid misaligned access in cached lookup.
700 JH/14 Change header file name for freeradius-client. Relevant if compiling
701 with Radius support; from the Gentoo tree and checked under Fedora.
703 JH/15 Bug 1712: Introduce $prdr_requested flag variable
705 JH/16 Bug 1714: Permit an empty string as expansion result for transport
706 option transport_filter, meaning no filtering.
708 JH/17 Bug 1713: Fix non-PDKIM_DEBUG build. Patch from Jasen Betts.
710 JH/18 Bug 1709: When built with TLS support, the tls_advertise_hosts option now
711 defaults to "*" (all hosts). The variable is now available when not built
712 with TLS, default unset, mainly to enable keeping the testsuite sane.
713 If a server certificate is not supplied (via tls_certificate) an error is
714 logged, and clients will find TLS connections fail on startup. Presumably
715 they will retry in-clear.
716 Packagers of Exim are strongly encouraged to create a server certificate
717 at installation time.
719 HS/03 Add -bP config_file as a synonym for -bP configure_file, for consistency
720 with the $config_file variable.
722 JH/19 Two additional event types: msg:rcpt:defer and msg:rcpt:host:defer. Both
723 in transport context, after the attempt, and per-recipient. The latter type
724 is per host attempted. The event data is the error message, and the errno
725 information encodes the lookup type (A vs. MX) used for the (first) host,
726 and the trailing two digits of the smtp 4xx response.
728 GF/01 Bug 1715: Fix for race condition in exicyclog, where exim could attempt
729 to write to mainlog (or rejectlog, paniclog) in the window between file
730 creation and permissions/ownership being changed. Particularly affects
731 installations where exicyclog is run as root, rather than exim user;
732 result is that the running daemon panics and dies.
734 JH/20 Bug 1701: For MySQL lookups, support MySQL config file option group names.
736 JH/21 Bug 1720: Add support for priority groups and weighted-random proxy
737 selection for the EXPERIMENTAL_SOCKS feature, via new per-proxy options
738 "pri" and "weight". Note that the previous implicit priority given by the
739 list order is no longer honoured.
741 JH/22 Bugs 963, 1721: Fix some corner cases in message body canonicalization
744 JH/23 Move SOCKS5 support from Experimental to mainline, enabled for a build
745 by defining SUPPORT_SOCKS.
747 JH/26 Move PROXY support from Experimental to mainline, enabled for a build
748 by defining SUPPORT_PROXY. Note that the proxy_required_hosts option
749 is renamed to hosts_proxy, and the proxy_{host,target}_{address,port}.
750 variables are renamed to proxy_{local,external}_{address,port}.
752 JH/27 Move Internationalisation support from Experimental to mainline, enabled
753 for a build by defining SUPPORT_I18N
755 JH/28 Bug 1745: Fix redis lookups to handle (quoted) spaces embedded in parts
756 of the query string, and make ${quote_redis:} do that quoting.
758 JH/29 Move Events support from Experimental to mainline, enabled by default
759 and removable for a build by defining DISABLE_EVENT.
761 JH/30 Updated DANE implementation code to current from Viktor Dukhovni.
763 JH/31 Fix bug with hosts_connection_nolog and named-lists which were wrongly
764 cached by the daemon.
766 JH/32 Move Redis support from Experimental to mainline, enabled for a build
767 by defining LOOKUP_REDIS. The libhiredis library is required.
769 JH/33 Bug 1748: Permit ACL dnslists= condition in non-smtp ACLs if explicit
770 keys are given for lookup.
772 JH/34 Bug 1192: replace the embedded copy of PolarSSL RSA routines in the DKIM
773 support, by using OpenSSL or GnuTLS library ones. This means DKIM is
774 only supported when built with TLS support. The PolarSSL SHA routines
775 are still used when the TLS library is too old for convenient support.
777 JH/35 Require SINGLE_DH_USE by default in OpenSSL (main config option
778 openssl_options), for security. OpenSSL forces this from version 1.1.0
779 server-side so match that on older versions.
781 JH/36 Bug 1778: longstanding bug in memory use by the ${run } expansion: A fresh
782 allocation for $value could be released as the expansion processing
783 concluded, but leaving the global pointer active for it.
785 JH/37 Bug 1769: Permit a VRFY ACL to override the default 252 response,
786 and to use the domains and local_parts ACL conditions.
788 JH/38 Fix cutthrough bug with body lines having a single dot. The dot was
789 incorrectly not doubled on cutthrough transmission, hence seen as a
790 body-termination at the receiving system - resulting in truncated mails.
791 Commonly the sender saw a TCP-level error, and retransmitted the message
792 via the normal store-and-forward channel. This could result in duplicates
793 received - but deduplicating mailstores were liable to retain only the
794 initial truncated version.
796 JH/39 Bug 1781: Fix use of DKIM private-keys having trailing '=' in the base-64.
798 JH/40 Fix crash in queryprogram router when compiled with EXPERIMENTAL_SRS.
800 JH/41 Bug 1792: Fix selection of headers to sign for DKIM: bottom-up. While
801 we're in there, support oversigning also; bug 1309.
803 JH/42 Bug 1796: Fix error logged on a malware scanner connection failure.
805 HS/04 Add support for keep_environment and add_environment options.
807 JH/43 Tidy coding issues detected by gcc --fsanitize=undefined. Some remain;
808 either intentional arithmetic overflow during PRNG, or testing config-
811 JH/44 Bug 1800: The combination of a -bhc commandline option and cutthrough
812 delivery resulted in actual delivery. Cancel cutthrough before DATA
815 JH/45 Fix cutthrough, when connection not opened by verify and target hard-
816 rejects a recipient: pass the reject to the originator.
818 JH/46 Multiple issues raised by Coverity. Some were obvious or plausible bugs.
819 Many were false-positives and ignorable, but it's worth fixing the
822 JH/47 Fix build on HP-UX and older Solaris, which need (un)setenv now also
823 for the new environment-manipulation done at startup. Move the routines
824 from being local to tls.c to being global via the os.c file.
826 JH/48 Bug 1807: Fix ${extract } for the numeric/3-string case. While preparsing
827 an extract embedded as result-arg for a map, the first arg for extract
828 is unavailable so we cannot tell if this is a numbered or keyed
829 extraction. Accept either.
835 JH/01 Bug 1545: The smtp transport option "retry_include_ip_address" is now
838 JH/02 The smtp transport option "multi_domain" is now expanded.
840 JH/03 The smtp transport now requests PRDR by default, if the server offers
843 JH/04 Certificate name checking on server certificates, when exim is a client,
844 is now done by default. The transport option tls_verify_cert_hostnames
845 can be used to disable this per-host. The build option
846 EXPERIMENTAL_CERTNAMES is withdrawn.
848 JH/05 The value of the tls_verify_certificates smtp transport and main options
849 default to the word "system" to access the system default CA bundle.
850 For GnuTLS, only version 3.0.20 or later.
852 JH/06 Verification of the server certificate for a TLS connection is now tried
853 (but not required) by default. The verification status is now logged by
854 default, for both outbound TLS and client-certificate supplying inbound
857 JH/07 Changed the default rfc1413 lookup settings to disable calls. Few
860 JH/08 The EXPERIMENTAL_DSN compile option is no longer needed; all Delivery
861 Status Notification (bounce) messages are now MIME format per RFC 3464.
862 Support for RFC 3461 DSN options NOTIFY,ENVID,RET,ORCPT can be advertised
863 under the control of the dsn_advertise_hosts option, and routers may
864 have a dsn_lasthop option.
866 JH/09 A timeout of 2 minutes is now applied to all malware scanner types by
867 default, modifiable by a malware= option. The list separator for
868 the options can now be changed in the usual way. Bug 68.
870 JH/10 The smtp_receive_timeout main option is now expanded before use.
872 JH/11 The incoming_interface log option now also enables logging of the
873 local interface on delivery outgoing connections.
875 JH/12 The cutthrough-routing facility now supports multi-recipient mails,
876 if the interface and destination host and port all match.
878 JH/13 Bug 344: The verify = reverse_host_lookup ACL condition now accepts a
881 JH/14 Bug 1573: The spam= ACL condition now additionally supports Rspamd.
882 Patch from Andrew Lewis.
884 JH/15 Bug 670: The spamd_address main option (for the spam= ACL condition)
885 now supports optional time-restrictions, weighting, and priority
886 modifiers per server. Patch originally by <rommer@active.by>.
888 JH/16 The spamd_address main option now supports a mixed list of local
889 and remote servers. Remote servers can be IPv6 addresses, and
890 specify a port-range.
892 JH/17 Bug 68: The spamd_address main option now supports an optional
893 timeout value per server.
895 JH/18 Bug 1581: Router and transport options headers_add/remove can
896 now have the list separator specified.
898 JH/19 Bug 392: spamd_address, and clamd av_scanner, now support retry
901 JH/20 Bug 1571: Ensure that $tls_in_peerdn is set, when verification fails
904 JH/21 Support for the A6 type of dns record is withdrawn.
906 JH/22 Bug 608: The result of a QUIT or not-QUIT toplevel ACL now matters
907 rather than the verbs used.
909 JH/23 Bug 1572: Increase limit on SMTP confirmation message copy size
910 from 255 to 1024 chars.
912 JH/24 Verification callouts now attempt to use TLS by default.
914 HS/01 DNSSEC options (dnssec_require_domains, dnssec_request_domains)
915 are generic router options now. The defaults didn't change.
917 JH/25 Bug 466: Add RFC2322 support for MIME attachment filenames.
918 Original patch from Alexander Shikoff, worked over by JH.
920 HS/02 Bug 1575: exigrep falls back to autodetection of compressed
921 files if ZCAT_COMMAND is not executable.
923 JH/26 Bug 1539: Add timeout/retry options on dnsdb lookups.
925 JH/27 Bug 286: Support SOA lookup in dnsdb lookups.
927 JH/28 Bug 1588: Do not use the A lookup following an AAAA for setting the FQDN.
928 Normally benign, it bites when the pair was led to by a CNAME;
929 modern usage is to not canonicalize the domain to a CNAME target
930 (and we were inconsistent anyway for A-only vs AAAA+A).
932 JH/29 Bug 1632: Removed the word "rejected" from line logged for ACL discards.
934 JH/30 Check the forward DNS lookup for DNSSEC, in addition to the reverse,
935 when evaluating $sender_host_dnssec.
937 JH/31 Check the HELO verification lookup for DNSSEC, adding new
938 $sender_helo_dnssec variable.
940 JH/32 Bug 1397: Enable ECDHE on OpenSSL, just the NIST P-256 curve.
942 JH/33 Bug 1346: Note MAIL cmd seen in -bS batch, to avoid smtp_no_mail log.
944 JH/34 Bug 1648: Fix a memory leak seen with "mailq" and large queues.
946 JH/35 Bug 1642: Fix support of $spam_ variables at delivery time. Was
947 documented as working, but never had. Support all but $spam_report.
949 JH/36 Bug 1659: Guard checking of input smtp commands again pseudo-command
950 added for tls authenticator.
952 HS/03 Add perl_taintmode main config option
958 TL/01 When running the test suite, the README says that variables such as
959 no_msglog_check are global and can be placed anywhere in a specific
960 test's script, however it was observed that placement needed to be near
961 the beginning for it to behave that way. Changed the runtest perl
962 script to read through the entire script once to detect and set these
963 variables, reset to the beginning of the script, and then run through
964 the script parsing/test process like normal.
966 TL/02 The BSD's have an arc4random API. One of the functions to induce
967 adding randomness was arc4random_stir(), but it has been removed in
968 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
969 function when detected.
971 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
972 cause callback expansion.
974 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
975 syntax errors in an expansion can be treated as a string instead of
976 logging or causing an error, due to the internal use of bool_lax
977 instead of bool when processing it.
979 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
980 server certificates when making smtp deliveries.
982 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
984 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
986 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
988 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
989 Merged patch from Sebastian Wiedenroth.
991 JH/05 Fix results-pipe from transport process. Several recipients, combined
992 with certificate use, exposed issues where response data items split
993 over buffer boundaries were not parsed properly. This eventually
994 resulted in duplicates being sent. This issue only became common enough
995 to notice due to the introduction of connection certificate information,
996 the item size being so much larger. Found and fixed by Wolfgang Breyha.
998 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
999 size buffer was used, resulting in syntax errors when an expansion
1002 JH/07 Add support for directories of certificates when compiled with a GnuTLS
1003 version 3.3.6 or later.
1005 JH/08 Rename the TPDA experimental facility to Event Actions. The #ifdef
1006 is EXPERIMENTAL_EVENT, the main-configuration and transport options
1007 both become "event_action", the variables become $event_name, $event_data
1008 and $event_defer_errno. There is a new variable $verify_mode, usable in
1009 routers, transports and related events. The tls:cert event is now also
1010 raised for inbound connections, if the main configuration event_action
1013 TL/06 In test suite, disable OCSP for old versions of openssl which contained
1014 early OCSP support, but no stapling (appears to be less than 1.0.0).
1016 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
1017 server certificate names available under the smtp transport option
1018 "tls_verify_cert_hostname" now do not permit multi-component wildcard
1021 JH/10 Time-related extraction expansions from certificates now use the main
1022 option "timezone" setting for output formatting, and are consistent
1023 between OpenSSL and GnuTLS compilations. Bug 1541.
1025 JH/11 Fix a crash in mime ACL when meeting a zero-length, quoted or RFC2047-
1026 encoded parameter in the incoming message. Bug 1558.
1028 JH/12 Bug 1527: Autogrow buffer used in reading spool files. Since they now
1029 include certificate info, eximon was claiming there were spoolfile
1032 JH/13 Bug 1521: Fix ldap lookup for single-attr request, multiple-attr return.
1034 JH/14 Log delivery-related information more consistently, using the sequence
1035 "H=<name> [<ip>]" wherever possible.
1037 TL/07 Bug 1547: Omit RFCs from release. Draft and RFCs have licenses which
1038 are problematic for Debian distribution, omit them from the release
1041 JH/15 Updates and fixes to the EXPERIMENTAL_DSN feature.
1043 JH/16 Fix string representation of time values on 64bit time_t architectures.
1046 JH/17 Fix a null-indirection in certextract expansions when a nondefault
1047 output list separator was used.
1052 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
1053 checkers that were complaining about end of non-void function with no
1056 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
1057 This was a regression introduced in 4.83 by another bugfix.
1059 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
1061 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
1062 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
1068 TF/01 Correctly close the server side of TLS when forking for delivery.
1070 When a message was received over SMTP with TLS, Exim failed to clear up
1071 the incoming connection properly after forking off the child process to
1072 deliver the message. In some situations the subsequent outgoing
1073 delivery connection happened to have the same fd number as the incoming
1074 connection previously had. Exim would try to use TLS and fail, logging
1075 a "Bad file descriptor" error.
1077 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
1078 utilities have not been installed.
1080 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
1081 temporary space as the ACL may create new global variables.
1083 TL/01 LDAP support uses per connection or global context settings, depending
1084 upon the detected version of the libraries at build time.
1086 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
1087 to extract and use the src ip:port in logging and expansions as if it
1088 were a direct connection from the outside internet. PPv2 support was
1089 updated based on HAProxy spec change in May 2014.
1091 JH/02 Add ${listextract {number}{list}{success}{fail}}.
1093 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
1094 Properly escape header and check for NULL return.
1096 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
1099 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
1101 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
1102 characters in header names, implemented as a verify condition.
1103 Contributed by Michael Fischer v. Mollard.
1105 TL/05 Rename SPF condition results err_perm and err_temp to standardized
1106 results permerror and temperror. Previous values are deprecated but
1107 still accepted. In a future release, err_perm and err_temp will be
1108 completely removed, which will be a backward incompatibility if the
1109 ACL tests for either of these two old results. Patch contributed by
1110 user bes-internal on the mailing list.
1112 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
1114 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
1115 selectors, in both main and reject logs.
1117 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
1120 JH/07 Add malware type "sock" for talking to simple daemon.
1122 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
1124 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
1125 routers/transports under cutthrough routing.
1127 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
1128 numbers. Touch up "bool" conditional to keep the same definition.
1130 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
1132 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
1134 JH/12 Expand items in router/transport headers_add or headers_remove lists
1135 individually rather than the list as a whole. Bug 1452.
1137 Required for reasonable handling of multiple headers_ options when
1138 they may be empty; requires that headers_remove items with embedded
1139 colons must have them doubled (or the list-separator changed).
1141 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
1142 view the policy declared in the DMARC record. Currently, $dmarc_status
1143 is a combined value of both the record presence and the result of the
1146 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
1148 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
1149 dnslookup router and the smtp transport (applying to the forward
1152 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
1153 of ldap servers used for a specific lookup. Patch provided by Heiko
1156 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
1157 New variable $lookup_dnssec_authenticated for observability.
1159 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
1160 Patch submitted by Lars Timman.
1162 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
1164 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
1165 Requires trusted mode and valid format message id, aborts otherwise.
1166 Patch contributed by Heiko Schlichting.
1168 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
1169 certextract with support for various fields. Bug 1358.
1171 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
1172 is requested by default, modifiable by smtp transport option
1175 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
1176 operate on certificate variables to give certificate fingerprints
1177 Also new ${sha256:cert_variable}.
1179 JH/23 The PRDR feature is moved from being Experimental into the mainline.
1181 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
1182 Christian Aistleitner.
1184 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
1186 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
1187 file. Patch from Wolfgang Breyha.
1189 JH/25 Expand the coverage of the delivery $host and $host_address to
1190 client authenticators run in verify callout. Bug 1476.
1192 JH/26 Port service names are now accepted for tls_on_connect_ports, to
1193 align with daemon_smtp_ports. Bug 72.
1195 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
1196 support and error reporting did not work properly.
1198 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
1199 and is readable. Patch from Andrew Colin Kissa.
1201 TL/14 Enhance documentation of ${run expansion and how it parses the
1202 commandline after expansion, particularly in the case when an
1203 unquoted variable expansion results in an empty value.
1205 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
1207 PP/02 Fix internal collision of T_APL on systems which support RFC3123
1208 by renaming away from it. Addresses GH issue 15, reported by
1211 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
1213 TL/15 SECURITY: prevent double expansion in math comparison functions
1214 (can expand unsanitized data). Not remotely exploitable.
1221 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
1223 PP/02 Make -n do something, by making it not do something.
1224 When combined with -bP, the name of an option is not output.
1226 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
1229 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
1230 $sender_host_name and config options to manage this, and basic check
1233 PP/05 DSCP support for outbound connections and control modifier for inbound.
1235 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
1236 (Only plugin which currently uses this is kerberos4, which nobody should
1237 be using, but we should make it available and other future plugins might
1238 conceivably use it, even though it would break NAT; stuff *should* be
1239 using channel bindings instead).
1241 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
1242 name; added for Sendmail compatibility; requires admin caller.
1243 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
1244 just ignore it); requires trusted caller.
1245 Also parse but ignore: -Ac -Am -X<logfile>
1248 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
1250 TL/02 Add +smtp_confirmation as a default logging option.
1252 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
1253 Patch by Magnus Holmgren from 2007-02-20.
1255 TL/04 Bugzilla 1281 - Spec typo.
1256 Bugzilla 1283 - Spec typo.
1257 Bugzilla 1290 - Spec grammar fixes.
1259 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
1261 TL/06 Add Experimental DMARC support using libopendmarc libraries.
1263 TL/07 Fix an out of order global option causing a segfault. Reported to dev
1264 mailing list by by Dmitry Isaikin.
1266 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
1268 JH/02 Support "G" suffix to numbers in ${if comparisons.
1270 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
1272 NM/01 Bugzilla 1197 - Spec typo
1273 Bugzilla 1196 - Spec examples corrections
1275 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
1277 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
1278 gnutls_enable_pkcs11, but renamed to more accurately indicate its
1281 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
1282 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
1284 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
1285 "acl {{name}{arg}...}", and optional args on acl condition
1288 JH/05 Permit multiple router/transport headers_add/remove lines.
1290 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
1292 JH/07 Avoid using a waiting database for a single-message-only transport.
1293 Performance patch from Paul Fisher. Bugzilla 1262.
1295 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
1298 JH/09 Add $headers_added variable, with content from use of ACL modifier
1299 add_header (but not yet added to the message). Bugzilla 199.
1301 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
1302 Pulled from Bugzilla 817 by Wolfgang Breyha.
1304 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
1306 (nb: this is the same fix as in Exim 4.80.1)
1308 JH/11 Add A= logging on delivery lines, and a client_set_id option on
1311 JH/12 Add optional authenticated_sender logging to A= and a log_selector
1314 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
1316 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
1317 advertise SMTP AUTH mechanism to us, instead of a generic
1318 protocol violation error. Also, make Exim more robust to bad
1319 data from the Dovecot auth socket.
1321 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
1323 When a queue runner is handling a message, Exim first routes the
1324 recipient addresses, during which it prunes them based on the retry
1325 hints database. After that it attempts to deliver the message to
1326 any remaining recipients. It then updates the hints database using
1329 So if a recipient address works intermittently, it can get repeatedly
1330 deferred at routing time. The retry hints record remains fresh so the
1331 address never reaches the final cutoff time.
1333 This is a fairly common occurrence when a user is bumping up against
1334 their storage quota. Exim had some logic in its local delivery code
1335 to deal with this. However it did not apply to per-recipient defers
1336 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
1338 This change adds a proper retry rule check during routing so that the
1339 final cutoff time is checked against the message's age. We only do
1340 this check if there is an address retry record and there is not a
1341 domain retry record; this implies that previous attempts to handle
1342 the address had the retry_use_local_parts option turned on. We use
1343 this as an approximation for the destination being like a local
1344 delivery, as in LMTP.
1346 I suspect this new check makes the old local delivery cutoff check
1347 redundant, but I have not verified this so I left the code in place.
1349 TF/02 Correct gecos expansion when From: is a prefix of the username.
1351 Test 0254 submits a message to Exim with the header
1355 When I ran the test suite under the user fanf2, Exim expanded
1356 the header to contain my full name, whereas it should have added
1357 a Resent-Sender: header. It erroneously treats any prefix of the
1358 username as equal to the username.
1360 This change corrects that bug.
1362 GF/01 DCC debug and logging tidyup
1363 Error conditions log to paniclog rather than rejectlog.
1364 Debug lines prefixed by "DCC: " to remove any ambiguity.
1366 TF/03 Avoid unnecessary rebuilds of lookup-related code.
1368 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
1369 Bug spotted by Jeremy Harris; was flawed since initial commit.
1370 Would have resulted in OCSP responses post-SNI triggering an Exim
1371 NULL dereference and crash.
1373 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
1375 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
1376 Bug detection, analysis and fix by Samuel Thibault.
1377 Bugzilla 1331, Debian bug #698092.
1379 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
1381 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
1382 Server implementation by Todd Lyons, client by JH.
1383 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
1384 config variable "prdr_enable" controls whether the server
1385 advertises the facility. If the client requests PRDR a new
1386 acl_data_smtp_prdr ACL is called once for each recipient, after
1387 the body content is received and before the acl_smtp_data ACL.
1388 The client is controlled by both of: a hosts_try_prdr option
1389 on the smtp transport, and the server advertisement.
1390 Default client logging of deliveries and rejections involving
1391 PRDR are flagged with the string "PRDR".
1393 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
1394 fclose(). Diagnosis by Todd Lyons.
1396 PP/17 Update configure.default to handle IPv6 localhost better.
1397 Patch by Alain Williams (plus minor tweaks).
1400 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
1401 This is now consistent with GnuTLS, and is now documented: the
1402 previous undocumented portable approach to treating the option as
1403 unset was to force an expansion failure. That still works, and
1404 an empty string is now equivalent.
1406 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
1407 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
1408 not performing validation itself.
1410 PP/20 Added force_command boolean option to pipe transport.
1411 Patch from Nick Koston, of cPanel Inc.
1413 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
1416 TF/04 Added udpsend ACL modifier and hexquote expansion operator
1418 PP/21 Fix eximon continuous updating with timestamped log-files.
1419 Broken in a format-string cleanup in 4.80, missed when I repaired the
1420 other false fix of the same issue.
1421 Report and fix from Heiko Schlichting.
1424 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
1425 Report from Prashanth Katuri.
1427 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
1428 It's SecureTransport, so affects any MacOS clients which use the
1429 system-integrated TLS libraries, including email clients.
1431 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
1432 using a MIME ACL for non-SMTP local injection.
1433 Report and assistance in diagnosis by Warren Baker.
1435 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
1437 JH/16 Fix comparisons for 64b. Bugzilla 1385.
1439 TL/09 Add expansion variable $authenticated_fail_id to keep track of
1440 last id that failed so it may be referenced in subsequent ACL's.
1442 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
1445 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
1446 ldap library initialization, allowing self-signed CA's to be
1447 used. Also properly sets require_cert option later in code by
1448 using NULL (global ldap config) instead of ldap handle (per
1449 session). Bug diagnosis and testing by alxgomz.
1451 TL/12 Enhanced documentation in the ratelimit.pl script provided in
1452 the src/util/ subdirectory.
1454 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
1455 renamed to Transport Post Delivery Action by Jeremy Harris, as
1458 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
1459 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
1460 redis_servers = needs to be configured which will be used by the redis
1461 lookup. Patch from Warren Baker, of The Packet Hub.
1463 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
1465 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
1466 hostname or reverse DNS when processing a host list. Used suggestions
1467 from multiple comments on this bug.
1469 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
1471 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
1472 Missed a few lines, added it to make the runtest require no keyboard
1475 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
1476 contains upper case chars. Make router use caseful_local_part.
1478 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
1479 support when GnuTLS has been built with p11-kit.
1485 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
1487 This, or similar/improved, will also be change PP/11 of 4.82.
1493 PP/01 Handle short writes when writing local log-files.
1494 In practice, only affects FreeBSD (8 onwards).
1495 Bugzilla 1053, with thanks to Dmitry Isaikin.
1497 NM/01 Bugzilla 949 - Documentation tweak
1499 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
1502 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
1504 PP/02 Implemented gsasl authenticator.
1506 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
1508 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
1509 `pkg-config foo` for cflags/libs.
1511 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
1512 with rest of GSASL and with heimdal_gssapi.
1514 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
1515 `pkg-config foo` for cflags/libs for the TLS implementation.
1517 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
1518 properties get this fed in as external SSF. A number of robustness
1519 and debugging improvements to the cyrus_sasl authenticator.
1521 PP/08 cyrus_sasl server now expands the server_realm option.
1523 PP/09 Bugzilla 1214 - Log authentication information in reject log.
1524 Patch by Jeremy Harris.
1526 PP/10 Added dbmjz lookup type.
1528 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
1530 PP/12 MAIL args handles TAB as well as SP, for better interop with
1531 non-compliant senders.
1532 Analysis and variant patch by Todd Lyons.
1534 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
1535 Bug report from Lars Müller <lars@samba.org> (via SUSE),
1536 Patch from Dirk Mueller <dmueller@suse.com>
1538 PP/13 tls_peerdn now print-escaped for spool files.
1539 Observed some $tls_peerdn in wild which contained \n, which resulted
1540 in spool file corruption.
1542 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
1543 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
1544 or write after TLS renegotiation, which otherwise led to messages
1547 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
1548 as a tracking header (ie: a signed header comes before the signature).
1549 Patch from Wolfgang Breyha.
1551 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
1552 comma-sep list; embedded commas doubled.
1554 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
1556 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
1558 Report and patch from Dmitry Banschikov.
1560 PP/16 Removed "dont_insert_empty_fragments" from "openssl_options".
1561 Removed SSL_clear() after SSL_new() which led to protocol negotiation
1562 failures. We appear to now support TLS1.1+ with Exim.
1564 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
1565 lets Exim select keys and certificates based upon TLS SNI from client.
1566 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
1567 before an outbound SMTP session. New log_selector, +tls_sni.
1569 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
1570 NULL dereference. Report and patch from Alun Jones.
1572 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
1573 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
1574 Not seeing resolver debug output on NetBSD, but suspect this is a
1575 resolver implementation change.
1577 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
1578 Left warnings. Added "eximon gdb" invocation mode.
1580 PP/21 Defaulting "accept_8bitmime" to true, not false.
1582 PP/22 Added -bw for inetd wait mode support.
1584 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
1585 locate the relevant includes and libraries. Made this the default.
1587 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
1588 Bugzilla 1246, report and most of solution from Tomasz Kusy.
1590 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
1591 This may cause build issues on older platforms.
1593 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
1594 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
1595 gnutls_require_mac & gnutls_require_protocols (no longer supported).
1596 Added SNI support via GnuTLS too.
1597 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
1599 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
1601 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
1602 Applied second patch from Janne, implementing suggestion to default
1603 multiple-strings-in-record handling to match SPF spec.
1605 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
1607 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
1608 read-only, out of scope).
1609 Patch from Wolfgang Breyha, report from Stuart Northfield.
1611 PP/29 Fix three issues highlighted by clang analyser static analysis.
1612 Only crash-plausible issue would require the Cambridge-specific
1613 iplookup router and a misconfiguration.
1614 Report from Marcin Mirosław.
1616 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
1618 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
1619 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
1620 As part of this, removing so much warning spew let me fix some minor
1621 real issues in debug logging.
1623 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
1624 assignment on my part. Fixed.
1626 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
1627 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
1628 Janne Snabb (who went above and beyond: thank you).
1630 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
1631 string otherwise requires a connection and a bunch more work and it's
1632 relatively easy to get wrong. Should also expose TLS library linkage
1635 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
1636 64-bit ${eval} (JH/03).
1638 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
1639 GNU libc to support some of the 64-bit stuff, should not lead to
1640 conflicts. Defined before os.h is pulled in, so if a given platform
1641 needs to override this, it can.
1643 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
1644 protection layer was required, which is not implemented.
1645 Bugzilla 1254, patch from Wolfgang Breyha.
1647 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
1648 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
1649 tls_dhparam take prime identifiers. Also unbreak combination of
1650 OpenSSL+DH_params+TLSSNI.
1652 PP/39 Disable SSLv2 by default in OpenSSL support.
1658 PP/01 Solaris build fix for Oracle's LDAP libraries.
1659 Bugzilla 1109, patch from Stephen Usher.
1661 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
1663 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
1666 TF/02 Fix a couple more cases where we did not log the error message
1667 when unlink() failed. See also change 4.74-TF/03.
1669 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
1670 lock up or crash if it happened to be inside a call to libc when it
1671 got a SIGUSR1 from exiwhat.
1673 The SIGUSR1 handler appends the current process status to the process
1674 log which is later printed by exiwhat. It used to use the general
1675 purpose logging code to do this, but several functions it calls are
1676 not safe for signals.
1678 The new output code in the SIGUSR1 handler is specific to the process
1679 log, and simple enough that it's easy to inspect for signal safety.
1680 Removing some special cases also simplifies the general logging code.
1681 Removing the spurious timestamps from the process log simplifies
1684 TF/04 Improved ratelimit ACL condition.
1686 The /noupdate option has been deprecated in favour of /readonly which
1687 has clearer semantics. The /leaky, /strict, and /readonly update modes
1688 are mutually exclusive. The update mode is no longer included in the
1689 database key; it just determines when the database is updated. (This
1690 means that when you upgrade Exim will forget old rate measurements.)
1692 Exim now checks that the per_* options are used with an update mode that
1693 makes sense for the current ACL. For example, when Exim is processing a
1694 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
1695 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
1696 must specify per_mail/readonly. If you omit the update mode it defaults to
1697 /leaky where that makes sense (as before) or /readonly where required.
1699 The /noupdate option is now undocumented but still supported for
1700 backwards compatibility. It is equivalent to /readonly except that in
1701 ACLs where /readonly is required you may specify /leaky/noupdate or
1702 /strict/noupdate which are treated the same as /readonly.
1704 A useful new feature is the /count= option. This is a generalization
1705 of the per_byte option, so that you can measure the throughput of other
1706 aggregate values. For example, the per_byte option is now equivalent
1707 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
1709 The per_rcpt option has been generalized using the /count= mechanism
1710 (though it's more complicated than the per_byte equivalence). When it is
1711 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
1712 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
1713 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
1714 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
1715 Note that using per_rcpt with a non-readonly update mode in more than
1716 one ACL will cause the recipients to be double-counted. (The per_mail
1717 and per_byte options don't have this problem.)
1719 The handling of very low rates has changed slightly. If the computed rate
1720 is less than the event's count (usually one) then this event is the first
1721 after a long gap. In this case the rate is set to the same as this event's
1722 count, so that the first message of a spam run is counted properly.
1724 The major new feature is a mechanism for counting the rate of unique
1725 events. The new per_addr option counts the number of different
1726 recipients that someone has sent messages to in the last time period. It
1727 behaves like per_rcpt if all the recipient addresses are different, but
1728 duplicate recipient addresses do not increase the measured rate. Like
1729 the /count= option this is a general mechanism, so the per_addr option
1730 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
1731 example, measure the rate that a client uses different sender addresses
1732 with the options per_mail/unique=$sender_address. There are further
1733 details in the main documentation.
1735 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
1737 TF/06 Removed a few PCRE remnants.
1739 TF/07 Automatically extract Exim's version number from tags in the git
1740 repository when doing development or release builds.
1742 PP/02 Raise smtp_cmd_buffer_size to 16kB.
1743 Bugzilla 879. Patch from Paul Fisher.
1745 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
1746 Heavily based on revision 40f9a89a from Simon Arlott's tree.
1749 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
1751 PP/05 Variable $av_failed, true if the AV scanner deferred.
1752 Bugzilla 1078. Patch from John Horne.
1754 PP/06 Stop make process more reliably on build failure.
1755 Bugzilla 1087. Patch from Heiko Schlittermann.
1757 PP/07 Make maildir_use_size_file an _expandable_ boolean.
1758 Bugzilla 1089. Patch from Heiko Schlittermann.
1760 PP/08 Handle ${run} returning more data than OS pipe buffer size.
1761 Bugzilla 1131. Patch from Holger Weiß.
1763 PP/09 Handle IPv6 addresses with SPF.
1764 Bugzilla 860. Patch from Wolfgang Breyha.
1766 PP/10 GnuTLS: support TLS 1.2 & 1.1.
1768 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
1771 PP/11 match_* no longer expand right-hand-side by default.
1772 New compile-time build option, EXPAND_LISTMATCH_RHS.
1773 New expansion conditions, "inlist", "inlisti".
1775 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
1777 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
1779 PP/14 fix log_write() format string regression from TF/03.
1780 Bugzilla 1152. Patch from Dmitry Isaikin.
1786 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
1788 PP/02 Harmonised TLS library version reporting; only show if debugging.
1789 Layout now matches that introduced for other libraries in 4.74 PP/03.
1791 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
1793 PP/04 New "dns_use_edns0" global option.
1795 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
1798 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
1799 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
1801 TK/01 Updated PolarSSL code to 0.14.2.
1802 Bugzilla 1097. Patch from Andreas Metzler.
1804 PP/07 Catch divide-by-zero in ${eval:...}.
1805 Fixes bugzilla 1102.
1807 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
1810 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
1811 format-string attack -- SECURITY: remote arbitrary code execution.
1813 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
1814 time unintentionally subject to list matching rules, letting the header
1815 cause arbitrary Exim lookups (of items which can occur in lists, *not*
1816 arbitrary string expansion). This allowed for information disclosure.
1818 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
1819 INT_MIN/-1 -- value coerced to INT_MAX.
1825 NM/01 Workaround for PCRE version dependency in version reporting
1828 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
1829 This fixes portability to compilers other than gcc, notably
1830 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
1832 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
1833 makefiles for portability to HP-UX and POSIX correctness.
1835 PP/01 Permit LOOKUP_foo enabling on the make command-line.
1836 Also via indented variable definition in the Makefile.
1837 (Debugging by Oliver Heesakkers).
1839 PP/02 Restore caching of spamd results with expanded spamd_address.
1840 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
1842 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
1843 Improves build reliability. Fix from: Frank Elsner
1845 NM/02 Fix wide character breakage in the rfc2047 coding
1846 Fixes bug 1064. Patch from Andrey N. Oktyabrski
1848 NM/03 Allow underscore in dnslist lookups
1849 Fixes bug 1026. Patch from Graeme Fowler
1851 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
1852 Code patches from Adam Ciarcinski of NetBSD.
1854 NM/04 Fixed exiqgrep to cope with mailq missing size issue
1857 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
1858 is logged, to avoid truncation. Patch from John Horne.
1860 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
1861 Patch from Jakob Hirsch.
1863 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
1864 SQL string expansion failure details.
1865 Patch from Andrey Oktyabrski.
1867 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
1868 Patch from Simon Arlott.
1870 PP/09 New lookups functionality failed to compile on old gcc which rejects
1871 extern declarations in function scope.
1872 Patch from Oliver Fleischmann
1874 PP/10 Use sig_atomic_t for flags set from signal handlers.
1875 Check getgroups() return and improve debugging.
1876 Fixed developed for diagnosis in bug 927 (which turned out to be
1879 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
1880 Patch from Mark Zealey.
1882 PP/12 Bugzilla 1056: Improved spamd server selection.
1883 Patch from Mark Zealey.
1885 PP/13 Bugzilla 1086: Deal with maildir quota file races.
1886 Based on patch from Heiko Schlittermann.
1888 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
1889 Patch from Uwe Doering, sign-off by Michael Haardt.
1891 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
1892 variable declaration deep within a block. Bug and patch from
1895 PP/15 lookups-Makefile IRIX compatibility coercion.
1897 PP/16 Make DISABLE_DKIM build knob functional.
1899 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
1900 Patch by Simon Arlott
1902 TF/03 Fix valgrind.h portability to C89 compilers that do not support
1903 variable argument macros. Our copy now differs from upstream.
1909 TF/01 Failure to get a lock on a hints database can have serious
1910 consequences so log it to the panic log.
1912 TF/02 Log LMTP confirmation messages in the same way as SMTP,
1913 controlled using the smtp_confirmation log selector.
1915 TF/03 Include the error message when we fail to unlink a spool file.
1917 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
1918 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
1919 for maintaining out-of-tree patches for some time.
1921 PP/01 Bugzilla 139: Documentation and portability issues.
1922 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
1923 Handle per-OS dynamic-module compilation flags.
1925 PP/02 Let /dev/null have normal permissions.
1926 The 4.73 fixes were a little too stringent and complained about the
1927 permissions on /dev/null. Exempt it from some checks.
1928 Reported by Andreas M. Kirchwitz.
1930 PP/03 Report version information for many libraries, including
1931 Exim version information for dynamically loaded libraries. Created
1932 version.h, now support a version extension string for distributors
1933 who patch heavily. Dynamic module ABI change.
1935 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
1936 privilege escalation vulnerability whereby the Exim run-time user
1937 can cause root to append content of the attacker's choosing to
1940 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
1943 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
1944 If dropping privileges for untrusted macros, we disabled normal logging
1945 on the basis that it would fail; for the Exim run-time user, this is not
1946 the case, and it resulted in successful deliveries going unlogged.
1947 Fixed. Reported by Andreas Metzler.
1953 PP/01 Date: & Message-Id: revert to normally being appended to a message,
1954 only prepend for the Resent-* case. Fixes regression introduced in
1955 Exim 4.70 by NM/22 for Bugzilla 607.
1957 PP/02 Include check_rfc2047_length in configure.default because we're seeing
1958 increasing numbers of administrators be bitten by this.
1960 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
1962 PP/03 Bugzilla 994: added openssl_options main configuration option.
1964 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
1966 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
1968 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
1970 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
1971 without a peer certificate, leading to a segfault because of an
1972 assumption that peers always have certificates. Be a little more
1973 paranoid. Problem reported by Martin Tscholak.
1975 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
1976 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
1977 NB: ClamAV planning to remove STREAM in "middle of 2010".
1978 CL also introduces -bmalware, various -d+acl logging additions and
1979 more caution in buffer sizes.
1981 PP/09 Implemented reverse_ip expansion operator.
1983 PP/10 Bugzilla 937: provide a "debug" ACL control.
1985 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
1987 PP/12 Bugzilla 973: Implement --version.
1989 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
1991 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
1993 PP/15 Bugzilla 816: support multiple condition rules on Routers.
1995 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
1996 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
1997 ignore trailing whitespace.
1999 JJ/02 prevent non-panic DKIM error from being sent to paniclog
2001 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
2004 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
2005 Notification from Dr Andrew Aitchison.
2007 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
2008 ExtendedDetectionInfo response format.
2009 Notification from John Horne.
2011 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
2014 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
2015 XSL and documented dependency on system catalogs, with examples of how
2018 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
2021 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
2022 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
2023 configuration file which is writeable by the Exim user or group.
2025 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
2026 of configuration files to cover files specified with the -C option if
2027 they are going to be used with root privileges, not just the default
2030 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
2031 option (effectively making it always true).
2033 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
2034 files to be used while preserving root privileges.
2036 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
2037 that rogue child processes cannot use them.
2039 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
2040 run-time user, instead of root.
2042 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
2043 Exim run-time user without dropping privileges.
2045 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
2046 result string, instead of calling string_vformat() twice with the same
2049 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
2050 for other users. Others should always drop root privileges if they use
2051 -C on the command line, even for a whitelisted configure file.
2053 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
2055 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
2061 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
2062 $data_path, and $header_path variables; fixed documentation bugs and
2065 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
2066 exipick to access non-standard spools, including the "frozen" queue
2069 NM/01 Bugzilla 965: Support mysql stored procedures.
2070 Patch from Alain Williams
2072 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
2074 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
2075 Patch from Andreas Metzler
2077 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
2078 Patch from Kirill Miazine
2080 NM/05 Bugzilla 671: Added umask to procmail example.
2082 JJ/03 installed exipick 20100323.0, fixing doc bug
2084 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
2085 directory. Notification and patch from Dan Rosenberg.
2087 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
2089 TK/02 Improve log output when DKIM signing operation fails.
2091 MH/01 Treat the transport option dkim_domain as a colon separated
2092 list, not as a single string, and sign the message with each element,
2093 omitting multiple occurences of the same signer.
2095 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
2096 Bugzilla 985, 986. Patch by Simon Arlott
2098 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
2099 Patch by Simon Arlott
2101 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
2102 MBX locking. Notification from Dan Rosenberg.
2108 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
2110 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
2112 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
2114 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
2116 NM/04 Bugzilla 915: Flag broken perl installation during build.
2122 TK/01 Added patch by Johannes Berg that expands the main option
2123 "spamd_address" if it starts with a dollar sign.
2125 TK/02 Write list of recipients to X-Envelope-Sender header when building
2126 the mbox-format spool file for content scanning (suggested by Jakob
2129 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
2130 (http://www.dcc-servers.net/) support via dccifd. Activated by
2131 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
2133 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
2134 by Mark Daniel Reidel <mr@df.eu>.
2136 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
2137 When building exim an external PCRE library is now needed -
2138 PCRE is a system library on the majority of modern systems.
2139 See entry on PCRE_LIBS in EDITME file.
2141 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
2142 conversation. Added nologin parameter to request.
2143 Patch contributed by Kirill Miazine.
2145 TF/01 Do not log submission mode rewrites if they do not change the address.
2147 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
2149 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
2150 log files in place. Contributed by Roberto Lima.
2152 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
2154 TF/03 Bugzilla 615: When checking the local_parts router precondition
2155 after a local_part_suffix or local_part_prefix option, Exim now
2156 does not use the address's named list lookup cache, since this
2157 contains cached lookups for the whole local part.
2159 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
2160 Robert Millan. Documentation is in experimental-spec.txt.
2162 TF/04 Bugzilla 668: Fix parallel build (make -j).
2164 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
2166 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
2167 Patch provided by Jan Srzednicki.
2169 TF/05 Leading white space used to be stripped from $spam_report which
2170 wrecked the formatting. Now it is preserved.
2172 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
2173 that they are available at delivery time.
2175 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
2177 TF/08 TLS error reporting now respects the incoming_interface and
2178 incoming_port log selectors.
2180 TF/09 Produce a more useful error message if an SMTP transport's hosts
2181 setting expands to an empty string.
2183 NM/06 Bugzilla 744: EXPN did not work under TLS.
2184 Patch provided by Phil Pennock.
2186 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
2187 Patch provided by Richard Godbee.
2189 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
2190 acl_smtp_notquit, added index entry.
2192 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
2193 Patch provided by Eugene Bujak.
2195 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
2196 accept(). Patch provided by Maxim Dounin.
2198 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
2199 Patch provided by Phil Pennock.
2201 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
2203 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
2204 Patch provided by Brad "anomie" Jorsch.
2206 NM/14 Bugzilla 622: Added timeout setting to transport filter.
2207 Patch provided by Dean Brooks.
2209 TK/05 Add native DKIM support (does not depend on external libraries).
2211 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
2212 Patch provided by Graeme Fowler.
2214 NM/16 Bugzilla 851: Documentation example syntax fix.
2216 NM/17 Changed NOTICE file to remove references to embedded PCRE.
2218 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
2221 NM/19 Bugzilla 745: TLS version reporting.
2222 Patch provided by Phil Pennock.
2224 NM/20 Bugzilla 167: bool: condition support.
2225 Patch provided by Phil Pennock.
2227 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
2228 clients. Patch provided by Phil Pennock.
2230 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
2231 Patch provided by Brad "anomie" Jorsch.
2233 NM/23 Bugzilla 687: Fix misparses in eximstats.
2234 Patch provided by Heiko Schlittermann.
2236 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
2237 Patch provided by Heiko Schlittermann.
2239 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
2240 plus update to original patch.
2242 NM/26 Bugzilla 799: Documentation correction for ratelimit.
2244 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
2245 Patch provided by David Brownlee.
2247 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
2249 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
2251 NM/30 Bugzilla 888: TLS documentation bugfixes.
2253 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
2255 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
2256 Unlike the original bugzilla I have changed all shell scripts in src tree.
2258 NM/33 Bugzilla 898: Transport filter timeout fix.
2259 Patch by Todd Rinaldo.
2261 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
2262 Patch by Serge Demonchaux.
2264 NM/35 Bugzilla 39: Base64 decode bug fixes.
2265 Patch by Jakob Hirsch.
2267 NM/36 Bugzilla 909: Correct connect() call in dcc code.
2269 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
2271 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
2273 NM/39 Bugzilla 911: Fixed MakeLinks build script.
2279 TK/01 Add preliminary DKIM support. Currently requires a forked version of
2280 ALT-N's libdkim that I have put here:
2281 http://duncanthrax.net/exim-experimental/
2283 Note to Michael Haardt: I had to rename some vars in sieve.c. They
2284 were called 'true' and it seems that C99 defines that as a reserved
2285 keyword to be used with 'bool' variable types. That means you could
2286 not include C99-style headers which use bools without triggering
2287 build errors in sieve.c.
2289 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
2290 as mailq or other aliases. Changed the --help handling significantly
2291 to do whats expected. exim_usage() emits usage/help information.
2293 SC/01 Added the -bylocaldomain option to eximstats.
2295 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
2297 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
2299 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
2305 PH/01 Another patch from the Sieve maintainer.
2307 PH/02 When an IPv6 address is converted to a string for single-key lookup
2308 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
2309 dots are used instead of colons so that keys in lsearch files need not
2310 contain colons. This was done some time before quoting was made available
2311 in lsearch files. However, iplsearch files do require colons in IPv6 keys
2312 (notated using the quote facility) so as to distinguish them from IPv4
2313 keys. This meant that lookups for IP addresses in host lists did not work
2314 for iplsearch lookups.
2316 This has been fixed by arranging for IPv6 addresses to be expressed with
2317 colons if the lookup type is iplsearch. This is not incompatible, because
2318 previously such lookups could never work.
2320 The situation is now rather anomalous, since one *can* have colons in
2321 ordinary lsearch keys. However, making the change in all cases is
2322 incompatible and would probably break a number of configurations.
2324 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
2327 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
2328 conversion specification without a maximum field width, thereby enabling
2329 a rogue spamd server to cause a buffer overflow. While nobody in their
2330 right mind would setup Exim to query an untrusted spamd server, an
2331 attacker that gains access to a server running spamd could potentially
2332 exploit this vulnerability to run arbitrary code as the Exim user.
2334 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
2335 $primary_hostname instead of what libspf2 thinks the hosts name is.
2337 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
2338 a directory entry by the name of the lookup key. Previously, if a
2339 symlink pointed to a non-existing file or a file in a directory that
2340 Exim lacked permissions to read, a lookup for a key matching that
2341 symlink would fail. Now it is enough that a matching directory entry
2342 exists, symlink or not. (Bugzilla 503.)
2344 PH/03 The body_linecount and body_zerocount variables are now exported in the
2347 PH/04 Added the $dnslist_matched variable.
2349 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
2350 This means they are set thereafter only if the connection becomes
2353 PH/06 Added the client_condition to authenticators so that some can be skipped
2354 by clients under certain conditions.
2356 PH/07 The error message for a badly-placed control=no_multiline_responses left
2357 "_responses" off the end of the name.
2359 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
2361 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
2362 (without spaces) instead of just copying the configuration text.
2364 PH/10 Added the /noupdate option to the ratelimit ACL condition.
2366 PH/11 Added $max_received_linelength.
2368 PH/12 Added +ignore_defer and +include_defer to host lists.
2370 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
2371 way in which PCRE > 7.0 is built.
2373 PH/14 Implemented queue_only_load_latch.
2375 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
2376 MAIL command. The effect was to mangle the value on 64-bit systems.
2378 PH/16 Another patch from the Sieve maintainer.
2380 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
2382 PH/18 If a system quota error occurred while trying to create the file for
2383 a maildir delivery, the message "Mailbox is full" was not appended to the
2384 bounce if the delivery eventually timed out. Change 4.67/27 below applied
2385 only to a quota excession during the actual writing of the file.
2387 PH/19 It seems that peer DN values may contain newlines (and other non-printing
2388 characters?) which causes problems in log lines. The DN values are now
2389 passed through string_printing() before being added to log lines.
2391 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
2392 and InterBase are left for another time.)
2394 PH/21 Added message_body_newlines option.
2396 PH/22 Guard against possible overflow in moan_check_errorcopy().
2398 PH/23 POSIX allows open() to be a macro; guard against that.
2400 PH/24 If the recipient of an error message contained an @ in the local part
2401 (suitably quoted, of course), incorrect values were put in $domain and
2402 $local_part during the evaluation of errors_copy.
2408 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
2409 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
2412 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
2413 issue a MAIL command.
2415 PH/02 In an ACL statement such as
2417 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
2419 if a client was not listed at all, or was listed with a value other than
2420 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
2421 the condition was not true (as it should be), so access was not denied.
2422 The bug was that the ! inversion was incorrectly passed on to the second
2423 item. This has been fixed.
2425 PH/03 Added additional dnslists conditions == and =& which are different from
2426 = and & when the dns lookup returns more than one IP address.
2428 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
2429 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
2431 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
2432 FSYNC, which compiles an option called disable_fsync that allows for
2433 bypassing fsync(). The documentation is heavily laced with warnings.
2435 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
2437 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
2438 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
2439 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
2440 including adding "make clean"; (3) Added -fPIC when compiling the test
2441 dynamically loaded module, to get rid of a warning.
2443 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
2444 message fails, move_frozen_messages = true and ignore_bounce_errors_after
2445 = 0s. The bug is otherwise harmless.
2447 PH/07 There was a bug in the dovecot authenticator such that the value of
2448 $auth1 could be overwritten, and so not correctly preserved, after a
2449 successful authentication. This usually meant that the value preserved by
2450 the server_setid option was incorrect.
2452 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
2454 PH/09 Installed PCRE release 7.0.
2456 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
2457 run for batched SMTP input. It is now run at the start of every message
2458 in the batch. While fixing this I discovered that the process information
2459 (output by running exiwhat) was not always getting set for -bs and -bS
2460 input. This is fixed, and it now also says "batched" for BSMTP.
2462 PH/11 Added control=no_pipelining.
2464 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
2465 patch, slightly modified), and move the expansion of helo_data till after
2466 the connection is made in the smtp transport (so it can use these
2469 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
2471 PH/14 Added log_selector = +pid.
2473 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
2475 PH/16 Add ${if forany and ${if forall.
2477 PH/17 Added dsn_from option to vary the From: line in DSNs.
2479 PH/18 Flush SMTP output before performing a callout, unless control =
2480 no_callout_flush is set.
2482 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
2483 was true (the default) a successful delivery failed to delete the retry
2484 item, thus causing premature timeout of the address. The bug is now
2487 PH/20 Added hosts_avoid_pipelining to the smtp transport.
2489 PH/21 Long custom messages for fakedefer and fakereject are now split up
2490 into multiline responses in the same way that messages for "deny" and
2491 other ACL rejections are.
2493 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
2494 with slight modification.
2496 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
2497 draft, changing the syntax and factoring some duplicate code".
2499 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
2500 for deliveries of the second and subsequent messages over the same SMTP
2503 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
2504 ${reduce, with only minor "tidies".
2506 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
2508 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
2509 expansion side effects.
2511 PH/27 When a message times out after an over-quota error from an Exim-imposed
2512 quota, the bounce message says "mailbox is full". This message was not
2513 being given when it was a system quota that was exceeded. It now should
2516 MH/03 Made $recipients available in local_scan(). local_scan() already has
2517 better access to the recipient list through recipients_list[], but
2518 $recipients can be useful in postmaster-provided expansion strings.
2520 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
2521 in the case of a MAIL command with additional options following the
2522 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
2523 were accidentally chopped off.
2525 PH/29 SMTP synchronization checks are implemented when a command is read -
2526 there is a check that no more input is waiting when there shouldn't be
2527 any. However, for some commands, a delay in an ACL can mean that it is
2528 some time before the response is written. In this time, more input might
2529 arrive, invalidly. So now there are extra checks after an ACL has run for
2530 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
2531 pipelining has not been advertised.
2533 PH/30 MH's patch to allow iscntrl() characters to be list separators.
2535 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
2536 returned in the SMTP response when smtp_return_error_details was false.
2537 This has been fixed.
2539 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
2540 instead of the C I/O that was originally supplied, because problems were
2541 reported on Solaris.
2543 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
2544 Exim which did not show up earlier: it was assuming that a call to
2545 SSL_CTX_set_info_callback() might give an error value. In fact, there is
2546 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
2547 was a macro that became an assignment, so it seemed to work. This has
2548 changed to a proper function call with a void return, hence the compile
2549 error. Exim's code has been fixed.
2551 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
2554 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
2556 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
2558 PH/37 If a message is not accepted after it has had an id assigned (e.g.
2559 because it turns out to be too big or there is a timeout) there is no
2560 "Completed" line in the log. When some messages of this type were
2561 selected by exigrep, they were listed as "not completed". Others were
2562 picked up by some special patterns. I have improved the selection
2563 criteria to be more general.
2565 PH/38 The host_find_failed option in the manualroute router can now be set
2566 to "ignore", to completely ignore a host whose IP address cannot be
2567 found. If all hosts are ignored, the behaviour is controlled by the new
2568 host_all_ignored option.
2570 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
2571 homing or because of multiple MX records with /mx) generated more than
2572 one IP address, and the following item turned out to be the local host,
2573 all the secondary addresses of the first item were incorrectly removed
2574 from the list, along with the local host and any following hosts (which
2575 is what is supposed to happen).
2577 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
2578 whoever called Exim into the -H file. In the case of the daemon it was
2579 behaving confusingly. When first started, it used values for whoever
2580 started the daemon, but after a SIGHUP it used the Exim user (because it
2581 calls itself on a restart). I have changed the code so that it now always
2584 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
2585 message are rejected with the same error (e.g. no authentication or bad
2586 sender address), and a DATA command is nevertheless sent (as can happen
2587 with PIPELINING or a stupid MUA), the error message that was given to the
2588 RCPT commands is included in the rejection of the DATA command. This is
2589 intended to be helpful for MUAs that show only the final error to their
2592 PH/42 Another patch from the Sieve maintainer.
2594 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
2595 Eximstats - Fixed some broken HTML links and added missing column headers
2597 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
2598 columns for Rejects, Temp Rejects, Ham, and Spam rows.
2600 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
2602 PH/43 Yet another patch from the Sieve maintainer.
2604 PH/44 I found a way to check for a TCP/IP connection going away before sending
2605 the response to the final '.' that terminates a message, but only in the
2606 case where the client has not sent further data following the '.'
2607 (unfortunately, this is allowed). However, in many cases there won't be
2608 any further data because there won't be any more messages to send. A call
2609 to select() can be used: if it shows that the input is "ready", there is
2610 either input waiting, or the socket has been closed. An attempt to read
2611 the next input character can distinguish the two cases. Previously, Exim
2612 would have sent an OK response which the client would never have see.
2613 This could lead to message repetition. This fix should cure that, at
2614 least in a lot of common cases.
2616 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
2617 advertised in response to EHLO.
2623 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
2624 fixed by 4.65/MH/01 (is this a record?) are fixed:
2626 (i) An empty string was always treated as zero by the numeric comparison
2627 operators. This behaviour has been restored.
2629 (ii) It is documented that the numeric comparison operators always treat
2630 their arguments as decimal numbers. This was broken in that numbers
2631 starting with 0 were being interpreted as octal.
2633 While fixing these problems I realized that there was another issue that
2634 hadn't been noticed. Values of message_size_limit (both the global option
2635 and the transport option) were treated as octal if they started with 0.
2636 The documentation was vague. These values are now always treated as
2637 decimal, and I will make that clear in the documentation.
2643 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
2644 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
2647 MH/01 Don't check that the operands of numeric comparison operators are
2648 integers when their expansion is in "skipping" mode (fixes bug
2649 introduced by 4.64-PH/07).
2651 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
2652 child addresses, Exim now panics and dies. Previously, because the count
2653 is held in a short int, deliveries were likely to be lost. As such a
2654 large number of recipients for a single message is ridiculous
2655 (performance will be very, very poor), I have chosen to impose a limit
2656 rather than extend the field.
2662 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
2663 leftover -K file (the existence of which was triggered by #402).
2664 While we were at it, introduced process PID as part of the -K
2665 filename. This should rule out race conditions when creating
2668 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
2669 processing considerably. Previous code took too long for large mails,
2670 triggering a timeout which in turn triggers #401.
2672 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
2673 in the DK code in transports.c. sendfile() is not really portable,
2674 hence the _LINUX specificness.
2676 TF/01 In the add_headers option to the mail command in an Exim filter,
2677 there was a bug that Exim would claim a syntax error in any
2678 header after the first one which had an odd number of characters
2681 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
2682 callout verification, Exim cached a "reject" for the entire domain. This
2683 is correct for most verifications, but it is not correct for a recipient
2684 verification with use_sender or use_postmaster set, because in that case
2685 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
2686 case of MAIL FROM:<> rejection from other early rejections (e.g.
2687 rejection of HELO). When verifying a recipient using a non-null MAIL
2688 address, the cache is ignored if it shows MAIL FROM:<> rejection.
2689 Whatever the result of the callout, the value of the domain cache is
2690 left unchanged (for any other kind of callout, getting as far as trying
2691 RCPT means that the domain itself is ok).
2693 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
2696 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
2697 manifest itself as EPIPE rather than ECONNECT. When tidying away a
2698 session, the daemon ignores ECONNECT errors and logs others; it now
2699 ignores EPIPE as well.
2701 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
2702 (quoted-printable decoding).
2704 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
2705 later the small subsequent patch to fix an introduced bug.
2707 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
2709 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
2711 PH/08 An error is now given if message_size_limit is specified negative.
2713 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
2714 to be given (somewhat) arbitrary names.
2716 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
2719 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
2720 miscellaneous code fixes
2722 PH/10 Added the log_reject_target ACL modifier to specify where to log
2725 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
2726 hostname. This is wrong, because it relates to the incoming message (and
2727 probably the interface on which it is arriving) and not to the outgoing
2728 callout (which could be using a different interface). This has been
2729 changed to use the value of the helo_data option from the smtp transport
2730 instead - this is what is used when a message is actually being sent. If
2731 there is no remote transport (possible with a router that sets up host
2732 addresses), $smtp_active_hostname is used.
2734 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
2735 tweaks were necessary in order to get it to work (see also 21 below):
2736 (a) The code assumed that strncpy() returns a negative number on buffer
2737 overflow, which isn't the case. Replaced with Exim's string_format()
2739 (b) There were several signed/unsigned issues. I just did the minimum
2740 hacking in of casts. There is scope for a larger refactoring.
2741 (c) The code used strcasecmp() which is not a standard C function.
2742 Replaced with Exim's strcmpic() function.
2743 (d) The code set only $1; it now sets $auth1 as well.
2744 (e) A simple test gave the error "authentication client didn't specify
2745 service in request". It would seem that Dovecot has changed its
2746 interface. Fortunately there's a specification; I followed it and
2747 changed what the client sends and it appears to be working now.
2749 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
2752 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
2753 address A is aliased to B and C, where B exists and C does not. Without
2754 -v the output is "A verified" because verification stops after a
2755 successful redirection if more than one address is generated. However,
2756 with -v the child addresses are also verified. Exim was outputting "A
2757 failed to verify" and then showing the successful verification for C,
2758 with its parentage. It now outputs "B failed to verify", showing B's
2759 parentage before showing the successful verification of C.
2761 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
2762 look up a TXT record in a specific list after matching in a combined
2765 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
2766 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
2767 they consult the DNS. I had assumed they would set it the way they
2768 wanted; and indeed my experiments on Linux seem to show that in some
2769 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
2770 To be on the safe side, however, I have now made the interface to
2771 host_find_byname() similar to host_find_bydns(), with an argument
2772 containing the DNS resolver options. The host_find_byname() function now
2773 sets these options at its start, just as host_find_bydns() does. The smtp
2774 transport options dns_qualify_single and dns_search_parents are passed to
2775 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
2776 of host_find_byname() use the default settings of RES_DEFNAMES
2777 (qualify_single) but not RES_DNSRCH (search_parents).
2779 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
2780 spool_read_header() do less string testing, by means of a preliminary
2781 switch on the second character of optional "-foo" lines. (This is
2782 overdue, caused by the large number of possibilities that now exist.
2783 Originally there were few.) While I was there, I also converted the
2784 str(n)cmp tests so they don't re-test the leading "-" and the first
2785 character, in the hope this might squeeze out yet more improvement.
2787 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
2788 flag allowing group syntax was set by the header_syntax check but not
2789 turned off, possible causing trouble later; (2) The flag was not being
2790 set at all for the header_verify test, causing "group"-style headers to
2791 be rejected. I have now set it in this case, and also caused header_
2792 verify to ignore an empty address taken from a group. While doing this, I
2793 came across some other cases where the code for allowing group syntax
2794 while scanning a header line wasn't quite right (mostly, not resetting
2795 the flag correctly in the right place). These bugs could have caused
2796 trouble for malformed header lines. I hope it is now all correct.
2798 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
2799 with the "reply" argument non-NULL. The code, however (which originally
2800 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
2801 but it didn't always do it. This confused somebody who was copying the
2802 code for some other use. I have removed all the tests.
2804 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
2805 feature that was used to support insecure browsers during the U.S. crypto
2806 embargo. It requires special client support, and Exim is probably the
2807 only MTA that supported it -- and would never use it because real RSA is
2808 always available. This code has been removed, because it had the bad
2809 effect of slowing Exim down by computing (never used) parameters for the
2810 RSA_EXPORT functionality.
2812 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
2813 authenticator to fail if there's a tab character in the incoming data
2814 (there should never be unless someone is messing about, as it's supposed
2815 to be base64-encoded). Also added, on Timo's advice, the "secured" option
2816 if the connection is using TLS or if the remote IP is the same as the
2817 local IP, and the "valid-client-cert option" if a client certificate has
2820 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
2821 authenticators. This can be used for authorization after authentication
2822 succeeds. (In the case of plaintext, it servers for both authentication
2825 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
2826 if any retry times were supplied.
2828 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
2829 connection, where there is no client IP address to check. In this
2830 situation, the verify now always succeeds.
2832 PH/25 Applied John Jetmore's -Mset patch.
2834 PH/26 Added -bem to be like -Mset, but loading a message from a file.
2836 PH/27 In a string expansion for a processed (not raw) header when multiple
2837 headers of the same name were present, leading whitespace was being
2838 removed from all of them, but trailing whitespace was being removed only
2839 from the last one. Now trailing whitespace is removed from each header
2840 before concatenation. Completely empty headers in a concatenation (as
2841 before) are ignored.
2843 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
2844 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
2846 PH/29 [Removed. This was a change that I later backed out, and forgot to
2847 correct the ChangeLog entry (that I had efficiently created) before
2848 committing the later change.]
2850 PH/30 Exim was sometimes attempting to deliver messages that had suffered
2851 address errors (4xx response to RCPT) over the same connection as other
2852 messages routed to the same hosts. Such deliveries are always "forced",
2853 so retry times are not inspected. This resulted in far too many retries
2854 for the affected addresses. The effect occurred only when there were more
2855 hosts than the hosts_max_try setting in the smtp transport when it had
2856 the 4xx errors. Those hosts that it had tried were not added to the list
2857 of hosts for which the message was waiting, so if all were tried, there
2858 was no problem. Two fixes have been applied:
2860 (i) If there are any address or message errors in an SMTP delivery, none
2861 of the hosts (tried or untried) are now added to the list of hosts
2862 for which the message is waiting, so the message should not be a
2863 candidate for sending over the same connection that was used for a
2864 successful delivery of some other message. This seems entirely
2865 reasonable: after all the message is NOT "waiting for some host".
2866 This is so "obvious" that I'm not sure why it wasn't done
2867 previously. Hope I haven't missed anything, but it can't do any
2868 harm, as the worst effect is to miss an optimization.
2870 (ii) If, despite (i), such a delivery is accidentally attempted, the
2871 routing retry time is respected, so at least it doesn't keep
2872 hammering the server.
2874 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
2875 in ${readsocket because some servers need this prod.
2877 PH/32 Added some extra debug output when updating a wait-xxx database.
2879 PH/33 The hint "could be header name not terminated by colon", which has been
2880 given for certain expansion errors for a long time, was not being given
2881 for the ${if def:h_colon_omitted{... case.
2883 PH/34 The spec says: "With one important exception, whenever a domain list is
2884 being scanned, $domain contains the subject domain." There was at least
2885 one case where this was not true.
2887 PH/35 The error "getsockname() failed: connection reset by peer" was being
2888 written to the panic log as well as the main log, but it isn't really
2889 panic-worthy as it just means the connection died rather early on. I have
2890 removed the panic log writing for the ECONNRESET error when getsockname()
2893 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
2894 runs only) independently of the message's sender address. This meant
2895 that, if the 4xx error was in fact related to the sender, a different
2896 message to the same recipient with a different sender could confuse
2897 things. In particular, this can happen when sending to a greylisting
2898 server, but other circumstances could also provoke similar problems.
2899 I have changed the default so that the retry time for these errors is now
2900 based a combination of the sender and recipient addresses. This change
2901 can be overridden by setting address_retry_include_sender=false in the
2904 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
2905 remote server are returned as part of bounce messages. This was not
2906 happening for LMTP over a pipe (the lmtp transport), but now it is the
2907 same for both kinds of LMTP.
2909 PH/38 Despite being documented as not happening, Exim was rewriting addresses
2910 in header lines that were in fact CNAMEs. This is no longer the case.
2912 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
2913 and queue runs started by the daemon processed all messages. This has
2914 been fixed so that -R and -S can now usefully be given with -q<time>.
2916 PH/40 Import PCRE release 6.7 (fixes some bugs).
2918 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
2920 PH/42 Give an error if -q is specified more than once.
2922 PH/43 Renamed the variables $interface_address and $interface_port as
2923 $received_ip_address and $received_port, to make it clear that these
2924 values apply to message reception, and not to the outgoing interface when
2925 a message is delivered. (The old names remain recognized, of course.)
2927 PH/44 There was no timeout on the connect() call when using a Unix domain
2928 socket in the ${readsocket expansion. There now is.
2930 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
2931 be meaningful with "accept".
2933 SC/01 Eximstats V1.43
2934 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
2936 SC/02 Eximstats V1.44
2937 Use a glob alias rather than an array ref in the generated
2938 parser. This improves both readability and performance.
2940 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
2941 Collect SpamAssassin and rejection statistics.
2942 Don't display local sender or destination tables unless
2943 there is data to show.
2944 Added average volumes into the top table text output.
2946 SC/04 Eximstats V1.46
2947 Collect data on the number of addresses (recipients)
2948 as well as the number of messages.
2950 SC/05 Eximstats V1.47
2951 Added 'Message too big' to the list of mail rejection
2952 reasons (thanks to Marco Gaiarin).
2954 SC/06 Eximstats V1.48
2955 Mainlog lines which have GMT offsets and are too short to
2956 have a flag are now skipped.
2958 SC/07 Eximstats V1.49 (Alain Williams)
2959 Added the -emptyok flag.
2961 SC/08 Eximstats V1.50
2962 Fixes for obtaining the IP address from reject messages.
2964 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
2965 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
2966 whitespace changes from 4.64-PH/27
2968 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
2971 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
2972 are found, allow negative numbers in numeric criteria)
2974 JJ/06 exipick.20061117.2, added new $message_body_missing variable
2976 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
2977 to match changes made in 4.64-PH/43
2979 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
2981 PH/47 Put in an explicit test for a DNS lookup of an address record where the
2982 "domain" is actually an IP address, and force a failure. This locks out
2983 those revolvers/nameservers that support "A-for-A" lookups, in
2984 contravention of the specifications.
2986 PH/48 When a host name was looked up from an IP address, and the subsequent
2987 forward lookup of the name timed out, the host name was left in
2988 $sender_host_name, contrary to the specification.
2990 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
2991 restricted to single-key lookups, Exim was not diagnosing an error if
2992 * or *@ was used with a query-style lookup.
2994 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
2996 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
2997 long ago, but noone interested enough thought of it. Let's just say that
2998 the "1.1" means that there are some new functions that weren't there at
2999 some point in the past.
3001 PH/51 Error processing for expansion failure of helo_data from an smtp
3002 transport during callout processing was broken.
3004 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
3005 tested/used via the -bh/-bhc/-bs options.
3007 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
3008 bug, fixed in subsequent PCRE releases).
3010 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
3011 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
3013 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
3019 SC/01 Use a glob alias rather than an array ref in eximstats generated
3020 parser. This improves both readability and performance.
3022 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
3023 Don't display local sender or destination tables in eximstats unless
3024 there is data to show.
3025 Added average volumes into the eximstats top table text output.
3027 SC/03 Collect data on the number of addresses (recipients) as well
3028 as the number of messages in eximstats.
3030 TF/01 Correct an error in the documentation for the redirect router. Exim
3031 does (usually) call initgroups() when daemonizing.
3033 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
3034 with consistent privilege compared to when running as a daemon.
3036 TF/03 Note in the spec that $authenticated_id is not set for local
3037 submissions from trusted users.
3039 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
3040 Thanks to Dean Brooks <dean@iglou.com> for the patch.
3042 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
3043 by adding some example configuration directives to the default
3044 configuration file. A little bit of work is required to uncomment the
3045 directives and define how usernames and passwords are checked, but
3046 there is now a framework to start from.
3048 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
3049 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
3050 without this. I don't know how relevant this is to other LDAP libraries.
3052 PH/02 Add the verb name to the "unknown ACL verb" error.
3054 PH/03 Magnus Holmgren's patch for filter_prepend_home.
3056 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
3058 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
3059 directory not expanded when it should be if an expanded home directory
3060 was set for the address (which is overridden by the transport).
3062 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
3065 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
3066 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
3067 because it is too late at that time, and has no effect.
3069 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
3070 security issue with \' (bugzilla #107). I could not use the
3071 PQescapeStringConn() function, because it needs a PGconn value as one of
3074 PH/08 When testing addresses using -bt, indicate those final addresses that
3075 are duplicates that would not cause an additional delivery. At least one
3076 person was confused, thinking that -bt output corresponded to deliveries.
3077 (Suppressing duplicates isn't a good idea as you lose the information
3078 about possibly different redirections that led to the duplicates.)
3080 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
3081 systems where poll() doesn't work, in particular OS X.
3083 PH/10 Added more information to debugging output for retry time not reached.
3085 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
3086 operations in malware.c.
3088 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
3091 PH/13 If write_rejectlog was set false when logging was sent to syslog with
3092 syslog_duplication set false, log lines that would normally be written
3093 both the the main log and to the reject log were not written to syslog at
3096 PH/14 In the default configuration, change the use of "message" in ACL warn
3097 statements to "add_header".
3099 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
3100 not followed by a command (e.g. "seen endif").
3102 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
3103 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
3106 PH/17 Added extra conditions to the default value of delay_warning_condition
3110 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
3111 { match{$h_precedence:}{(?i)bulk|list|junk} } \
3112 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
3115 The Auto-Submitted: and various List- headers are standardised, whereas I
3116 don't think Precedence: ever was.
3118 PH/18 Refactored debugging code in route_finduser() to show more information,
3119 in particular, the error code if getpwnam() issues one.
3121 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
3122 This is apparently needed in addition to the PH/07 change above to avoid
3123 any possible encoding problems.
3125 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
3126 but not after initializing Perl.
3128 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
3129 output them only if debugging. By default they are written stderr,
3130 apparently, which is not desirable.
3132 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
3135 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
3138 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
3140 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
3141 authenticated or an ident call has been made. Suppress the default
3142 values for $authenticated_id and $authenticated_sender (but permit -oMai
3143 and -oMas) when testing with -bh.
3145 PH/24 Re-jigged the order of the tests in the default configuration so that the
3146 tests for valid domains and recipients precede the DNS black list and CSA
3147 tests, on the grounds that those ones are more expensive.
3149 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
3150 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
3151 This bug exists in every version of Exim that I still have, right back to
3154 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
3155 However, an attempt to turn on case-sensitivity in a regex key by
3156 including (?-i) didn't work because the subject string was already
3157 lowercased, and the effects were non-intuitive. It turns out that a
3158 one-line patch can be used to allow (?-i) to work as expected.
3164 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
3165 other effects) broke the use of negated acl sub-conditions.
3167 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
3170 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
3171 "Deny" causes Exim to reject the incoming connection with a 554 error.
3172 Unfortunately, if there is a major crisis, such as a disk failure,
3173 tcp-wrappers gives "deny", whereas what one would like would be some
3174 kind of temporary error. A kludge has been added to help with this.
3175 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
3176 554 error is used if errno is still zero or contains ENOENT (which occurs
3177 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
3180 PH/03 Add -lutil to the default FreeBSD LIBS setting.
3182 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
3183 errors. Otherwise a message that provokes a temporary error (when other
3184 messages do not) can cause a whole host to time out.
3186 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
3187 addresses were routed directly to files or pipes from a redirect router.
3188 File deliveries just didn't batch; pipe deliveries might have suffered
3191 PH/06 A failure to get a lock for a hints database would erroneously always say
3192 "Failed to get write lock", even when it was really a read lock.
3194 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
3195 of 0600. This has been changed to use the value of the lockfile_mode
3196 option (which defaults to 0600).
3198 PH/08 Applied small patch from the Sieve maintainer.
3200 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
3201 folder from quota calculations, a direct delivery into this folder messed
3202 up the contents of the maildirsize file. This was because the regex was
3203 used only to exclude .Trash (or whatever) when the size of the mailbox
3204 was calculated. There was no check that a delivery was happening into an
3205 excluded directory. This bug has been fixed by ignoring all quota
3206 processing for deliveries into excluded directories.
3208 PH/10 Added the maildirfolder_create_regex option to appendfile.
3214 PH/01 The code for finding all the local interface addresses on a FreeBSD
3215 system running IPv6 was broken. This may well have applied to all BSD
3216 systems, as well as to others that have similar system calls. The broken
3217 code found IPv4 interfaces correctly, but gave incorrect values for the
3218 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
3219 that it would not match correctly against @[] and not recognize the IPv6
3222 PH/02 The ipliteral router was not recognizing addresses of the form user@
3223 [ipv6:....] because it didn't know about the "ipv6:" prefix.
3225 PH/03 Added disable_ipv6.
3227 PH/04 Changed $reply_address to use the raw form of the headers instead of the
3228 decoded form, because it is most often used to construct To: headers
3229 lines in autoreplies, and the decoded form may well be syntactically
3230 invalid. However, $reply_address has leading white space removed, and all
3231 newlines turned into spaces so that the autoreply transport does not
3234 PH/05 If group was specified without a user on a router, and no group or user
3235 was specified on a transport, the group from the router was ignored.
3237 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
3238 for visible compile-time settings that can be used to change these
3239 numbers, for those that want even more. Backwards compatibility with old
3240 spool files has been maintained. However, going back to a previous Exim
3241 release will lost any variables that are in spool files.
3243 PH/07 Two small changes when running in the test harness: increase delay when
3244 passing a TCP/IP connection to a new process, in case the original
3245 process has to generate a bounce, and remove special handling of
3246 127.0.0.2 (sic), which is no longer necessary.
3248 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
3249 be the same on different OS.
3251 PH/09 Moved a debug statement in filter processing to avoid a race problem when
3254 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
3255 whether --show-vars was specified or not
3257 JJ/02 exipick: Added support for new ACL variable spool format introduced
3260 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
3261 syntactically invalid From: or Reply-to: line, and a filter used this to
3262 generate an autoreply, and therefore failed to obtain an address for the
3263 autoreply, Exim could try to deliver to a non-existent relative file
3264 name, causing unrelated and misleading errors. What now happens is that
3265 it logs this as a hard delivery error, but does not attempt to create a
3268 PH/11 The exinext utility has a -C option for testing purposes, but although
3269 the given file was scanned by exinext itself; it wasn't being passed on
3270 when Exim was called.
3272 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
3273 an end-of-file indication when reading a command response.
3275 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
3276 compiled. In many other places in Exim, IPv6 addresses are always
3277 recognized, so I have changed this. It also means that IPv4 domain
3278 literals of the form [IPV4:n.n.n.n] are now always recognized.
3280 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
3281 used if the router is not running as root, for example, when verifying at
3282 ACL time, or when using -bh. The debugging output from this situation was
3283 non-existent - all you got was a failure to exec. I have made two
3286 (a) Failures to set uid/gid, the current directory, or a process leader
3287 in a subprocess such as that created by queryprogram now generate
3288 suitable debugging output when -d is set.
3290 (b) The queryprogram router detects when it is not running as root,
3291 outputs suitable debugging information if -d is set, and then runs
3292 the subprocess without attempting to change uid/gid.
3294 PH/15 Minor change to Makefile for building test_host (undocumented testing
3297 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
3298 additional section of a DNS packet that returns MX or SRV records.
3299 Instead, it always explicitly searches for A/AAAA records. This avoids
3300 major problems that occur when a DNS server includes only records of one
3301 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
3302 fixed another bug: if SRV records were looked up and the corresponding
3303 address records were *not* found in the additional section, the port
3304 values from the SRV records were lost.
3306 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
3307 using the correct key (the original address) when searching the retry
3308 rules in order to find which one to use for generating the retry hint.
3310 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
3311 adding the default one. Similarly, if it contains a Reply-To: header, the
3312 errors_reply_to option, if set, is not used.
3314 PH/19 When calculating a retry time, Exim used to measure the "time since
3315 failure" by looking at the "first failed" field in the retry record. Now
3316 it does not use this if it is later than than the arrival time of the
3317 message. Instead it uses the arrival time. This makes for better
3318 behaviour in cases where some deliveries succeed, thus re-setting the
3319 "first failed" field. An example is a quota failure for a huge message
3320 when small messages continue to be delivered. Without this change, the
3321 "time since failure" will always be short, possible causing more frequent
3322 delivery attempts for the huge message than are intended.
3323 [Note: This change was subsequently modified - see PH/04 for 4.62.]
3325 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
3326 $1, $2, $3) because the numerical variables can be reset during some
3327 expansion items (e.g. "match"), thereby losing the authentication data.
3329 PH/21 Make -bV show the size of off_t variables so that the test suite can
3330 decide whether to run tests for quotas > 2G.
3332 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
3333 mailbox_size, and mailbox_filecount in the appendfile transport. If a
3334 filecount value is greater than 2G or if a quota value is greater than 2G
3335 on a system where the size of off_t is not greater than 4, a panic error
3338 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
3339 never match. The debug and -bh output now contains an explicit error
3340 message indicating a malformed IPv4 address or mask.
3342 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
3343 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
3344 PH/23 above applies.
3346 PH/25 Do not write to syslog when running in the test harness. The only
3347 occasion when this arises is a failure to open the main or panic logs
3348 (for which there is an explicit test).
3350 PH/26 Added the /no_tell option to "control=freeze".
3352 PH/27 If a host name lookup failed very early in a connection, for example, if
3353 the IP address matched host_lookup and the reverse lookup yielded a name
3354 that did not have a forward lookup, an error message of the form "no IP
3355 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
3356 could be logged. Now it outputs the IP address instead of "NULL".
3358 PH/28 An enabling patch from MH: add new function child_open_exim2() which
3359 allows the sender and the authenticated sender to be set when
3360 submitting a message from within Exim. Since child_open_exim() is
3361 documented for local_scan(), the new function should be too.
3363 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
3364 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
3365 results in an empty string is now treated as unset.
3367 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
3369 PH/31 Added sender_verify_fail logging option.
3371 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
3372 needed by RFC 822 but not by RFC 2822 was commented out. I have now
3373 tidied the source and removed it altogether.
3375 PH/33 When a queue run was abandoned because the load average was too high, a
3376 log line was always written; now it is written only if the queue_run log
3377 selector is set. In addition, the log line for abandonment now contains
3378 information about the queue run such as the pid. This is always present
3379 in "start" and "stop" lines but was omitted from the "abandon" line.
3381 PH/34 Omit spaces between a header name and the colon in the error message that
3382 is given when verify = headers_syntax fails (if there are lots of them,
3383 the message gets confusing).
3385 PH/35 Change the default for dns_check_names_pattern to allow slashes within
3386 names, as there are now some PTR records that contain slashes. This check
3387 is only to protect against broken name servers that fall over on strange
3388 characters, so the fact that it applies to all lookups doesn't matter.
3390 PH/36 Now that the new test suite is complete, we can remove some of the
3391 special code in Exim that was needed for the old test suite. For example,
3392 sorting DNS records because real resolvers return them in an arbitrary
3393 order. The new test suite's fake resolver always returns records in the
3396 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
3397 bounces) except when queue_only is set, to avoid logging races between
3398 the different processes.
3400 PH/38 Panic-die if .include specifies a non-absolute path.
3402 PH/39 A tweak to the "H" retry rule from its user.
3404 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
3405 a label. They prevented compilation on older perls.
3407 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
3408 a warning to be raised on newish perls.
3410 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
3411 on queue. Changes to match documented behaviour of showing count of
3412 messages matching specified criteria.
3414 PH/40 Changed the default ident timeout from 30s to 5s.
3416 PH/41 Added support for the use of login_cap features, on those BSD systems
3417 that have them, for controlling the resources used by pipe deliveries.
3419 PH/42 The content-scanning code uses fopen() to create files in which to put
3420 message data. Previously it was not paying any attention to the mode of
3421 the files. Exim runs with umask(0) because the rest of the code creates
3422 files with open(), and sets the required mode explicitly. Thus, these
3423 files were ending up world-writeable. This was not a big issue, because,
3424 being within the spool directory, they were not world-accessible. I have
3425 created a function called modefopen, which takes an additional mode
3426 argument. It sets umask(777), creates the file, chmods it to the required
3427 mode, then resets the umask. All the relevant calls to fopen() in the
3428 content scanning code have been changed to use this function.
3430 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
3431 to 24 hours. This avoids potential overflow problems when processing G
3432 and H retry rules. I suspect nobody ever tinkers with this value.
3434 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
3436 PH/45 When the plaintext authenticator is running as a client, the server's
3437 challenges are checked to ensure they are valid base64 strings. By
3438 default, the authentication attempt is cancelled if an invalid string is
3439 received. Setting client_ignore_invalid_base64 true ignores these errors.
3440 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
3441 they are received. Thus, the responses can be made to depend on the
3442 challenges. If an invalid string is ignored, an empty string is placed in
3445 PH/46 Messages that are created by the autoreply transport now contains a
3446 References: header, in accordance with RFCs 2822 and 3834.
3448 PH/47 Added authenticated_sender_force to the smtp transport.
3450 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
3452 PH/49 Installed latest patch from the Sieve maintainer.
3454 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
3455 was also set, the appendfile transport was unnecessarily scanning a
3456 directory of message files (e.g. for maildir delivery) to find the count
3457 of files (along with the size), even though it did not need this
3458 information. It now does the scan only if it needs to find either the
3459 size of the count of files.
3461 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
3463 PH/52 Two bugs concerned with error handling when the smtp transport is
3466 (i) Exim was not creating retry information for temporary errors given
3467 for individual recipients after the DATA command when the smtp transport
3468 was used in LMTP mode. This meant that they could be retried too
3469 frequently, and not timed out correctly.
3471 (ii) Exim was setting the flag that allows error details to be returned
3472 for LMTP errors on RCPT commands, but not for LMTP errors for individual
3473 recipients that were returned after the DATA command.
3475 PH/53 This is related to PH/52, but is more general: for any failing address,
3476 when detailed error information was permitted to be returned to the
3477 sender, but the error was temporary, then after the final timeout, only
3478 "retry timeout exceeded" was returned. Now it returns the full error as
3479 well as "retry timeout exceeded".
3481 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
3482 do this, and (what is worse) MTAs that accept it.
3484 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
3485 will now be deprecated.
3487 PH/56 New os.c-cygwin from the Cygwin maintainer.
3489 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
3490 formats (previously only available in exim formats via -bpr, -bpru,
3491 and -bpra. Now also available in native and exiqgrep formats)
3493 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
3494 with very large, slow to parse queues
3496 JJ/08 exipick: added ! as generic prefix to negate any criteria format
3498 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
3500 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
3501 responses to authentication challenges, though it was showing the
3502 challenges; (ii) I've removed the CR characters from the debug output for
3505 PH/58 Allow for the insertion of a newline as well as a space when a string
3506 is turned into more than one encoded-word during RFC 2047 encoding. The
3507 Sieve code now uses this.
3509 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
3510 data_4xx, lost_connection, tls_required.
3512 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
3513 message was being sent as an SMTP response.
3515 PH/61 Add -l and -k options to exicyclog.
3517 PH/62 When verifying, if an address was redirected to one new address, so that
3518 verification continued, and the new address failed or deferred after
3519 having set something in $address_data, the value of $address_data was not
3520 passed back to the ACL. This was different to the case when no
3521 redirection occurred. The value is now passed back in both cases.
3523 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
3524 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
3525 use login_cap.h, so on its own it isn't the distinguishing feature. The
3526 new name refers directly to the setclassresources() function.
3528 PH/65 Added configuration files for NetBSD3.
3530 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
3532 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
3533 is preferred over IPv4.
3535 PH/68 The bounce_return_message and bounce_return_body options were not being
3536 honoured for bounces generated during the reception of non-SMTP messages.
3537 In particular, this applied to messages rejected by the ACL. This bug has
3538 been fixed. However, if bounce_return_message is true and bounce_return_
3539 body is false, the headers that are returned for a non-SMTP message
3540 include only those that have been read before the error was detected.
3541 (In the case of an ACL rejection, they have all been read.)
3543 PH/69 The HTML version of the specification is now built in a directory called
3544 spec_html instead of spec.html, because the latter looks like a path with
3545 a MIME-type, and this confuses some software.
3547 PH/70 Catch two compiler warnings in sieve.c.
3549 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
3550 function verify_get_ident() calls ip_connect() to connect a socket, but
3551 if the "connect()" function timed out, ip_connect() used to close the
3552 socket. However, verify_get_ident() also closes the socket later, and in
3553 between Exim writes to the log, which may get opened at this point. When
3554 the socket was closed in ip_connect(), the log could get the same file
3555 descriptor number as the socket. This naturally causes chaos. The fix is
3556 not to close the socket in ip_connect(); the socket should be closed by
3557 the function that creates it. There was only one place in the code where
3558 this was missing, in the iplookup router, which I don't think anybody now
3559 uses, but I've fixed it anyway.
3561 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
3562 well as to direct DNS lookups. Otherwise the handling of names in host
3563 lists is inconsistent and therefore confusing.
3569 PH/01 Two changes to the default runtime configuration:
3571 (1) Move the checks for relay_from_hosts and authenticated clients from
3572 after to before the (commented out) DNS black list checks.
3574 (2) Add control=submission to the relay_from_hosts and authenticated
3575 clients checks, on the grounds that messages accepted by these
3576 statements are most likely to be submissions.
3578 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
3580 (1) Generate an error if the third argument for the ${prvs expansion is
3583 (2) Treat a missing third argument of ${prvscheck as if it were an empty
3586 (3) Reset the variables that are obtained from the first argument of
3587 ${prvscheck and used in the second argument before leaving the code,
3588 because their memory is reclaimed, so using them afterwards may do
3591 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
3592 one (it's much easier than Tom thought :-).
3594 (5) Because of (4), we can now allow for the use of $prvscheck_result
3595 inside the third argument.
3597 PH/03 For some reason, the default setting of PATH when running a command from
3598 a pipe transport was just "/usr/bin". I have changed it to
3601 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
3602 anything to be listed in the output from -bV.
3604 PH/05 When a filter generated an autoreply, the entire To: header line was
3605 quoted in the delivery log line, like this:
3607 => >A.N.Other <ano@some.domain> <original@ddress> ...
3609 This has been changed so that it extracts the operative address. There
3610 may be more than one such address. If so, they are comma-separated, like
3613 => >ano@some.domain,ona@other.domain <original@ddress> ...
3615 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
3616 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
3617 not being looked up in the rDNS to get a host name, Exim was showing the
3618 IP address twice in Received: lines, even though the IP addresses were
3619 identical. For example:
3621 Received: from [1.2.3.4] (helo=[1.2.3.4])
3623 However, if the real host name was known, it was omitting the HELO data
3624 if it matched the actual IP address. This has been tidied up so that it
3625 doesn't show the same IP address twice.
3627 PH/07 When both +timestamp and +memory debugging was on, the value given by
3628 $tod_xxx expansions could be wrong, because the tod_stamp() function was
3629 called by the debug printing, thereby overwriting the timestamp buffer.
3630 Debugging no longer uses the tod_stamp() function when +timestamp is set.
3632 PH/08 When the original message was included in an autoreply transport, it
3633 always said "this is a copy of the message, including all the headers",
3634 even if body_only or headers_only was set. It now gives an appropriate
3637 PH/09 Applied a patch from the Sieve maintainer which:
3639 o fixes some comments
3640 o adds the (disabled) notify extension core
3641 o adds some debug output for the result of if/elsif tests
3642 o points to the current vacation draft in the documentation
3643 and documents the missing references header update
3647 o fixes a bug in processing the envelope test (when testing
3648 multiple envelope elements, the last element determined the
3651 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
3652 Electronic Mail") by including:
3654 Auto-submitted: auto-generated
3656 in the messages that it generates (bounce messages and others, such as
3657 warnings). In the case of bounce messages for non-SMTP messages, there was
3658 also a typo: it was using "Auto_submitted" (underscore instead of
3659 hyphen). Since every message generated by Exim is necessarily in response
3660 to another message, thes have all been changed to:
3662 Auto-Submitted: auto-replied
3664 in accordance with these statements in the RFC:
3666 The auto-replied keyword:
3668 - SHOULD be used on messages sent in direct response to another
3669 message by an automatic process,
3671 - MUST NOT be used on manually-generated messages,
3673 - MAY be used on Delivery Status Notifications (DSNs) and Message
3674 Disposition Notifications (MDNs),
3676 - MUST NOT be used on messages generated by automatic or periodic
3677 processes, except for messages which are automatic responses to
3680 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
3681 to the default Received: header definition.
3683 PH/12 Added log selector acl_warn_skipped (default on).
3685 PH/13 After a successful wildlsearch lookup, discard the values of numeric
3686 variables because (a) they are in the wrong storage pool and (b) even if
3687 they were copied, it wouldn't work properly because of the caching.
3689 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
3690 checking when decoding. Apparently there are clients that generate
3691 overlong encoded strings. Why am I not surprised?
3693 PH/15 If the first argument of "${if match_address" was not empty, but did not
3694 contain an "@" character, Exim crashed. Now it writes a panic log message
3695 and treats the condition as false.
3697 PH/16 In autoreply, treat an empty string for "once" the same as unset.
3699 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
3700 extension "envelope-auth". The code is finished and in agreement with
3701 other implementations, but there is no documentation so far and in fact,
3702 nobody wrote the draft yet. This extension is currently #undef'ed, thus
3703 not changing the active code.
3705 Print executed "if" and "elsif" statements when debugging is used. This
3706 helps a great deal to understand what a filter does.
3708 Document more things not specified clearly in RFC3028. I had all this
3709 sorted out, when out of a sudden new issues came to my mind. Oops."
3711 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
3714 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
3715 canonical form (as documented). However, after a host name lookup from
3716 the IP address, check_host() was doing a simple string comparison with
3717 addresses acquired from the DNS when checking that the found name did
3718 have the original IP as one of its addresses. Since any found IPv6
3719 addresses are likely to be in abbreviated form, the comparison could
3720 fail. Luckily, there already exists a function for doing the comparison
3721 by converting both addresses to binary, so now that is used instead of
3722 the text comparison.
3724 PH/20 There was another similar case to PH/19, when a complete host name was
3725 given in a host list; looking up its IP address could give an abbreviated
3726 form, whereas the current host's name might or might not be abbreviated.
3727 The same fix has been applied.
3733 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
3734 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
3737 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
3738 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
3740 PH/03 Typo: missing ".o" in src/pcre/Makefile.
3742 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
3743 header line, restrict the check to what is listed in RFCs 2369 and 2929.
3744 Also, for "Auto-Submitted", treat anything other than "no" as
3745 non-personal, in accordance with RFC 3834. (Previously it treated
3746 anything starting "auto-" as non-personal.)
3748 TF/01 The control=submission/name=... option had a problem with syntax
3749 errors if the name included a slash character. The /name= option
3750 now slurps the rest of the string, so it can include any characters
3751 but it must come last in the list of options (after /sender_retain
3754 PH/05 Some modifications to the interface to the fake nameserver for the new
3762 TK/01 Added the "success_on_redirect" address verification option. See
3763 NewStuff for rationale and an example.
3765 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
3767 PH/02 Patch to exigrep to allow it to work on syslog lines.
3769 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
3770 fread() to skip over the body file's header line, because in Cygwin the
3771 header line is locked and is inaccessible.
3773 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
3774 co-exist for some time) to make it clear that it is the Exim ID that is
3775 referenced, not the Message-ID: header line.
3777 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
3778 string_format() function, because snprintf() does not exist on all
3781 PH/06 The use of forbid_filter_existstest now also locks out the use of the
3782 ${stat: expansion item.
3784 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
3785 protocol synchronization error", to keep the pedants happy.
3787 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
3788 well as for IRIX systems, when gcc is being used. See the host.c source
3791 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
3793 PH/10 Named domain lists were not working if used in a queue_smtp_domains
3796 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
3797 transport and to the smtp transport in LMTP mode.
3799 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
3801 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
3802 run a filter in a subprocess. This could lead to confusion in subsequent
3803 lookups in the parent process. There should also be a search_tidyup() at
3804 the end of the subprocess.
3806 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
3807 only if the host matched helo_try_verify_hosts, which caused the
3808 verification to occur when the EHLO/HELO command was issued. The ACL just
3809 tested the remembered result. Now, if a previous verification attempt has
3810 not happened, "verify = helo" does it there and then.
3812 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
3814 TK/03 Fix log output including CR from clamd.
3816 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
3817 exist provoked a memory error which could cause a segfault.
3819 PH/15 Installed PCRE 6.2
3821 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
3823 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
3824 of the problem. Specifically, suggested +O2 rather than +O1 for the
3827 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
3829 PH/20 If a delivery was routed to a non-standard port by means of an SRV
3830 record, the port was not correctly logged when the outgoing_port log
3831 selector was set (it logged the transort's default port).
3833 PH/21 Added support for host-specific ports to manualroute, queryprogram,
3834 fallback_hosts, and "hosts" in the smtp transport.
3836 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
3837 host errors such as "Connection refused".
3839 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
3840 authentication with radiusclient 0.4.9:
3842 - Error returned from rc_read_config was caught wrongly
3843 - Username/password not passed on to radius server due to wrong length.
3845 The presumption is that some radiusclient API changes for 4.51/PH/17
3846 were not taken care of correctly. The code is still untested by me (my
3847 Linux distribution still has 0.3.2 of radiusclient), but it was
3848 contributed by a Radius user.
3850 PH/24 When doing a callout, the value of $domain wasn't set correctly when
3851 expanding the "port" option of the smtp transport.
3853 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
3854 while reading a MIME header. Thanks to Tom Hughes for a patch.
3856 PH/24 Include config.h inside local_scan.h so that configuration settings are
3859 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
3860 that in an ACL for RCPT (for example), you can examine exactly what was
3863 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
3864 commands, but it was not correctly comparing the address with the actual
3865 client host address. Thus, it would show the EHLO address in Received:
3866 header lines when this was not necessary.
3868 PH/27 Added the % operator to ${eval:}.
3870 PH/28 Exim tries to create and chdir to its spool directory when it starts;
3871 it should be ignoring failures (because with -C, for example, it has lost
3872 privilege). It wasn't ignoring creation failures other than "already
3875 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
3878 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
3879 because an input line was too long, either on its own, or by virtue of
3880 too many continuations, the temporary file was not being removed, and the
3881 return code was incorrect.
3883 PH/31 Missing "BOOL" in function definition in filtertest.c.
3885 PH/32 Applied Sieve patches from the maintainer.
3887 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
3889 PH/33 Added "verify = not_blind".
3891 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
3892 Local/Makefile (with some defaults set). These are used in built scripts
3893 such as exicyclog, but they have never been used in the exim_install
3894 script (though there are many overriding facilities there). I have
3895 arranged that the exim_install script now takes note of these two
3898 PH/35 Installed configuration files for Dragonfly.
3900 PH/36 When a locally submitted message by a trusted user did not contain a
3901 From: header, and the sender address was obtained from -f or from an SMTP
3902 MAIL command, and the trusted user did not use -F to supply a sender
3903 name, $originator_name was incorrectly used when constructing a From:
3904 header. Furthermore, $originator_name was used for submission mode
3905 messages from external hosts without From: headers in a similar way,
3906 which is clearly wrong.
3908 PH/37 Added control=suppress_local_fixups.
3910 PH/38 When log_selector = +received_sender was set, and the addition of the
3911 sender made the log line's construction buffer exactly full, or one byte
3912 less than full, an overflow happened when the terminating "\n" was
3915 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
3916 when the result of a list match is failure because a DNS lookup failed.
3918 PH/40 RM_COMMAND is now used in the building process.
3920 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
3921 the "build-* directories that it finds.
3923 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
3924 address in a domain literal was a prefix of an interface address.
3926 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
3927 when verifying a sender address, unless rewrite_headers is false.
3929 PH/44 Wrote a long comment about why errors_to addresses are verified as
3930 recipients, not senders.
3932 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
3933 the ratelimit ACL was added.
3935 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
3937 PH/46 Added extra information about PostgreSQL errors to the error string.
3939 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
3940 avoiding the need to install special zones in a real server. This is
3941 backwards compatible; if it can't find the fake resolver, it drops back.
3942 Thus, both old and new test suites can be run.
3944 TF/02 Added util/ratelimit.pl
3946 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
3947 clock is set back in time.
3949 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
3950 Candler <B.Candler@pobox.com>.
3952 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
3953 OK for addresses that are the result of redirections.
3955 PH/48 A number of further additions for the benefit of the new test suite,
3956 including a fake gethostbyname() that interfaces to the fake DNS resolver
3959 TF/06 The fix for widen_domains has also been applied to qualify_single and
3960 search_parents which are the other dnslookup options that can cause
3963 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
3966 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
3968 TF/07 Exim produced the error message "an SRV record indicated no SMTP
3969 service" if it encountered an MX record with an empty target hostname.
3970 The message is now "an MX or SRV record indicated no SMTP service".
3972 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
3973 if the DNS of the sending site is misconfigured. This is quite a
3974 common situation. This change restores the behaviour of treating a
3975 helo verification defer as a failure.
3977 PH/51 If self=fail was set on a router, the bounce message did not include the
3978 actual error message.
3984 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
3986 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
3987 command itself ended in error, the underlying message about the transport
3988 filter timeout was being overwritten with the pipe command error. Now the
3989 underlying error message should be appended to the second error message.
3991 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
3993 PH/02 Reduce the amount of output that "make" produces by default. Full output
3994 can still be requested.
3996 PH/03 The warning log line about a condition test deferring for a "warn" verb
3997 was being output only once per connection, rather than after each
3998 occurrence (because it was using the same function as for successful
3999 "warn" verbs). This seems wrong, so I have changed it.
4001 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
4002 it should not have, which might have caused a crash in the right
4003 circumstances, but probably never did.
4005 PH/04 Installed a modified version of Tony Finch's patch to make submission
4006 mode fix the return path as well as the Sender: header line, and to
4007 add a /name= option so that you can make the user's friendly name appear
4010 TF/03 Added the control = fakedefer ACL modifier.
4012 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
4013 Mark Lowes for thorough testing.
4015 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
4017 TK/03 Merged latest SRS patch from Miles Wilton.
4019 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
4020 with the definition in sysexits.h (which is #included earlier).
4021 Fortunately, Exim does not actually use EX_OK. The code used to try to
4022 preserve the sysexits.h value, by assuming that macro definitions were
4023 scanned for macro replacements. I have been disabused of this notion,
4024 so now the code just undefines EX_OK before #including unistd.h.
4026 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
4027 in the smtp transport. When a block could not be written in a single
4028 write() function, the timeout was being re-applied to each part-write.
4029 This seems wrong - if the receiver was accepting one byte at a time it
4030 would take for ever. The timeout is now adjusted when this happens. It
4031 doesn't have to be particularly precise.
4033 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
4034 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
4036 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
4037 without a domain if the check to <postmaster@domain> fails.
4039 SC/01 Eximstats: added -xls and the ability to specify output files
4040 (patch written by Frank Heydlauf).
4042 SC/02 Eximstats: use FileHandles for outputting results.
4044 SC/03 Eximstats: allow any combination of xls, txt, and html output.
4046 SC/04 Eximstats: fixed display of large numbers with -nvr option
4048 SC/05 Eximstats: fixed merging of reports with empty tables.
4050 SC/06 Eximstats: added the -include_original_destination flag
4052 SC/07 Eximstats: removed tabs and trailing whitespace.
4054 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
4056 TK/06 MBOX spool code: Add real "From " MBOX separator line
4057 so the .eml file is really in mbox format (even though
4058 most programs do not really care). Patch from Alex Miller.
4060 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
4061 The latter is generated from $received_to and is only set if the
4062 message has one envelope recipient. SA can use these headers,
4063 obviously out-of-the-box. Patch from Alex Miller.
4065 PH/08 The ${def test on a variable was returning false if the variable's
4066 value was "0", contrary to what the specification has always said!
4067 The result should be true unless the variable is empty.
4069 PH/09 The syntax error of a character other than { following "${if
4070 def:variable_name" (after optional whitespace) was not being diagnosed.
4071 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
4072 accidental colon was present, for example, could give incorrect results.
4074 PH/10 Tidied the code in a number of places where the st_size field of a stat()
4075 result is used (not including appendfile, where other changes are about
4078 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
4079 This involved changing a lot of size variables from int to off_t. It
4080 should work with maildirs and everything.
4082 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
4083 spamd dying while we are connected to it.
4085 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
4088 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
4089 function for BDB 4.3.
4091 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
4093 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
4094 the sources for PCRE have been re-arranged and more files are now
4097 PH/15 The code I had for printing potentially long long variables in PH/11
4098 above was not the best (it lost precision). The length of off_t variables
4099 is now inspected at build time, and an appropriate printing format (%ld
4100 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
4101 to be "long long int" or "long int". This is needed for the internal
4102 formatting function string_vformat().
4104 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
4105 the configuration file to be ":syslog", then the script "guesses" where
4106 the logs files are, rather than using the compiled in default. In our
4107 case the guess is not the same as the compiled default, so the script
4108 suddenly stopped working when I started to use syslog. The patch checks
4109 to see if log_file_path is "". If so, it attempts to read it from exim
4110 with no configuration file to get the compiled in version, before it
4111 falls back to the previous guessing code."
4113 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
4114 implementing BATV in an Exim configuration. See NewStuff for the gory
4117 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
4118 Makefile that are specific to HP-UX.
4120 PH/18 If the "use_postmaster" option was set for a recipient callout together
4121 with the "random" option, the postmaster address was used as the MAIL
4122 FROM address for the random test, but not for the subsequent recipient
4123 test. It is now used for both.
4125 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
4126 patch removes a few documentation additions to RFC 3028, because the
4127 latest draft now contains them. It adds the new en;ascii-case comparator
4128 and a new error check for 8bit text in MIME parts. Comparator and
4129 require names are now matched exactly. I enabled the subaddress
4130 extension, but it is not well tested yet (read: it works for me)."
4132 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
4133 rework some of the code of TK/09 above to avoid the hardwired use of
4134 "%lld" and "long long". Replaced the call to snprintf() with a call to
4137 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
4138 records point to non-existent hosts", "retry timeout exceeded", and
4139 "retry time not reached for any host after a long failure period".
4141 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
4142 experimental DomainKeys support:
4144 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
4145 (2) On an error such as an illegally used "control", the wrong name for
4146 the control was given.
4148 These problems did NOT occur unless DomainKeys support was compiled.
4150 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
4152 PH/24 Added ${if match_ip condition.
4154 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
4155 databases so that it will be absolutely obvious if a crash occurs in the
4156 DB library. This is a regular occurrence (often caused by mis-matched
4159 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
4160 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
4161 picked up on a user's system that detects such things. There doesn't seem
4162 to be a gcc warning option for this - only an attribute that has to be
4163 put on the function's prototype. It seems that in Fedora Core 4 they have
4164 set this on a number of new functions. No doubt there will be more in due
4167 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
4168 specify a transport. However, if an address that was verified by such a
4169 router was the subject of a callout, Exim crashed because it tried to
4170 read the rcpt_include_affixes from the non-existent transport. Now it
4171 just assumes that the setting of that option is false. This bug was
4172 introduced by 4.51/PH/31.
4174 PH/28 Changed -d+all to exclude +memory, because that information is very
4175 rarely of interest, but it makes the output a lot bigger. People tend to
4176 do -d+all out of habit.
4178 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
4179 code in os-type was giving problems when libc.so lives in lib64, like on
4182 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
4183 aren't the modern standard, and it seems that some systems' include files
4184 don't always have them. Exim was already checking for some of the newer
4185 ones like T_AAAA, and defining it itself. I've added checks for all the
4186 record types that Exim uses.
4188 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
4189 not automatically generating a new one, as it is supposed to. This
4190 prevented TLS from working. If the file did exist, but contained invalid
4191 data, a new version was generated, as expected. It was only the case of a
4192 non-existent file that was broken.
4194 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
4195 with a change in libdomainkeys > 0.64.
4197 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
4198 from DNS. If the selector record carries the flag, it now has
4199 precedence over the domain-wide flag.
4201 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
4203 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
4204 the use of a port name that isn't defined in /etc/services) occurred, the
4205 message was deferred as in a normal delivery, and thus remained on the
4206 spool, instead of being failed because of the mua_wrapper setting. This
4207 is now fixed, and I tidied up some of the mua_wrapper messages at the
4210 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
4211 the messages in a hash of arrays rather than using individual hashes.
4212 This is a bit cleaner and results in dramatic memory savings, albeit
4213 at a slight CPU cost.
4215 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
4216 as requested by Marc Sherman.
4218 SC/10 Eximstats: added histograms for user specified patterns as requested
4221 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
4223 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
4224 fopen() in the content-scanning modules that did not already have it.
4230 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
4231 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
4233 TK/02 Fix ACL "control" statement not being available in MIME ACL.
4235 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
4237 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
4238 to test Sieve filters that use "vacation".
4240 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
4241 that changes the way the GnuTLS parameters are stored in the cache file.
4242 The new format can be generated externally. For backward compatibility,
4243 if the data in the cache doesn't make sense, Exim assumes it has read an
4244 old-format file, and it generates new data and writes a new file. This
4245 means that you can't go back to an older release without removing the
4248 PH/03 A redirect router that has both "unseen" and "one_time" set does not
4249 work if there are any delivery delays because "one_time" forces the
4250 parent to be marked "delivered", so its unseen clone is never tried
4251 again. For this reason, Exim now forbids the simultaneous setting of
4254 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
4255 redirected to themselves ("homonym" addresses). Read the long ChangeLog
4256 entry if you want to know the details. The fix, however, neglected to
4257 consider the case when local delivery batching is involved. The test for
4258 "previously delivered" was not happening when checking to see if an
4259 address could be batched with a previous (undelivered) one; under
4260 certain circumstances this could lead to multiple deliveries to the same
4263 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
4264 in its include files, and this causes problems building Exim.
4266 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
4267 header_syntax) but Exim was just ignoring anything given after a slash.
4268 In particular, this caused confusion with an attempt to use "verify =
4269 reverse_host_lookup/defer_ok". An error is now given when options are
4270 supplied for verify items that do not have them. (Maybe reverse_host_
4271 lookup should have a defer_ok option, but that's a different point.)
4273 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
4274 defined by RFC 821) to 2048, because there were problems with some AUTH
4275 commands, and RFC 1869 says the size should be increased for extended
4276 SMTP commands that take arguments.
4278 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
4281 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
4282 "unknown" error; now it says that the functionality isn't in the binary.
4284 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
4285 an address' error message when a string expansion fails (syntax or
4286 whatever). Otherwise the password may appear in the log. Following change
4287 PH/42 below, there is no longer a chance of it appearing in a bounce
4290 PH/11 Installed exipick version 20050225.0 from John Jetmore.
4292 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
4293 of its addresses was ever tried. (Bugzilla bug #2.)
4295 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
4296 the result incorrectly in the debug output. (It correctly added a newline
4297 to what was transported.)
4299 TF/01 Added $received_time.
4301 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
4302 commented out examples of how to interface to a virus scanner and to
4303 SpamAssassin. Also added commented examples of av_scanner and
4304 spamd_address settings.
4306 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
4307 and controls are allowed in which ACLs. There were a couple of minor
4308 errors. Some of the entries in the conditions table (which is a table of
4309 where they are NOT allowed) were getting very unwieldy; rewrote them as a
4310 negation of where the condition IS allowed.
4312 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
4314 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
4315 header file does not have a version number, so I've had to invent a new
4316 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
4317 API. The code is untested by me (my Linux distribution still has 0.3.2 of
4318 radiusclient), but it was contributed by a Radius user.
4320 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
4321 files or directories, for OpenSSL.
4323 PH/19 When an Exim process that is running as root has to create an Exim log
4324 file, it does so in a subprocess that runs as exim:exim so as to get the
4325 ownership right at creation (otherwise, other Exim processes might see
4326 the file with the wrong ownership). There was no test for failure of this
4327 fork() call, which would lead to the process getting stuck as it waited
4328 for a non-existent subprocess. Forks do occasionally fail when resources
4329 run out. I reviewed all the other calls to fork(); they all seem to check
4332 PH/20 When checking for unexpected SMTP input at connect time (before writing
4333 the banner), Exim was not dealing correctly with a non-positive return
4334 from the read() function. If the client had disconnected by this time,
4335 the result was a log entry for a synchronization error with an empty
4336 string after "input=" when read() returned zero. If read() returned -1
4337 (an event I could not check), uninitialized data bytes were printed.
4338 There were reports of junk text (parts of files, etc) appearing after
4341 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
4343 PH/22 Added support for macro redefinition, and (re)definition in between
4344 driver and ACL definitions.
4346 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
4347 forgetting to use the resulting value; it was using the unexpanded value.
4349 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
4350 hadn't been configured. The fix is from Juergen Kreileder, who
4351 understands it better than I do:
4353 "Here's what I see happening with three configured cyrus_sasl
4354 authenticators configured (plain, login, cram-md5):
4356 On startup auth_cyrus_sasl_init() gets called for each of these.
4357 This means three calls to sasl_listmech() without a specified mech_list.
4358 => SASL tests which mechs of all available mechs actually work
4359 => three warnings about OTP not working
4360 => the returned list contains: plain, login, cram-md5, digest-md5, ...
4362 With the patch, sasl_listmech() also gets called three times. But now
4363 SASL's mech_list option is set to the server_mech specified in the the
4364 authenticator. Or in other words, the answer from sasl_listmech()
4365 gets limited to just the mech you're testing for (which is different
4367 => the return list contains just 'plain' or 'login', 'cram-md5' or
4368 nothing depending on the value of ob->server_mech.
4370 I've just tested the patch: Authentication still works fine,
4371 unavailable mechs specified in the exim configuration are still
4372 caught, and the auth.log warnings about OTP are gone."
4374 PH/25 When debugging is enabled, the contents of the command line are added
4375 to the debugging output, even when log_selector=+arguments is not
4378 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
4379 answer is "GNU", and only if the return is "GNU/something" is the answer
4382 PH/27 $acl_verify_message is now set immediately after the failure of a
4383 verification in an ACL, and so is available in subsequent modifiers. In
4384 particular, the message can be preserved by coding like this:
4386 warn !verify = sender
4387 set acl_m0 = $acl_verify_message
4389 Previously, $acl_verify_message was set only while expanding "message"
4390 and "log_message" when a very denied access.
4392 PH/28 Modified OS/os.c-Linux with
4394 -#ifndef OS_LOAD_AVERAGE
4395 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
4397 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
4398 nomenclature these days.)
4400 PH/29 Installed patch from the Sieve maintainer that adds the options
4401 sieve_useraddress and sieve_subaddress to the redirect router.
4403 PH/30 In these circumstances:
4404 . Two addresses routed to the same list of hosts;
4405 . First host does not offer TLS;
4406 . First host accepts first address;
4407 . First host gives temporary error to second address;
4408 . Second host offers TLS and a TLS session is established;
4409 . Second host accepts second address.
4410 Exim incorrectly logged both deliveries with the TLS parameters (cipher
4411 and peerdn, if requested) that were in fact used only for the second
4414 PH/31 When doing a callout as part of verifying an address, Exim was not paying
4415 attention to any local part prefix or suffix that was matched by the
4416 router that accepted the address. It now behaves in the same way as it
4417 does for delivery: the affixes are removed from the local part unless
4418 rcpt_include_affixes is set on the transport.
4420 PH/32 Add the sender address, as F=<...>, to the log line when logging a
4421 timeout during the DATA phase of an incoming message.
4423 PH/33 Sieve envelope tests were broken for match types other than :is. I have
4424 applied a patch sanctioned by the Sieve maintainer.
4426 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
4427 the uid or gid is negative. A case of a negative gid caused this to be
4428 noticed. The fix allows for either to be negative.
4430 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
4431 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
4434 PH/36 The change PH/12 above was broken. Fixed it.
4436 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
4437 the grounds that routing the same address twice would always produce the
4438 same answer. This might have been true once, but it is certainly no
4439 longer true now. Routing a child address may depend on the previous
4440 routing that produced that child. Some complicated redirection strategies
4441 went wrong when messages had multiple recipients, and made Exim's
4442 behaviour dependent on the order in which the addresses were given.
4444 I have moved the duplicate checking until after the routing is complete.
4445 Exim scans the addresses that are assigned to local and remote
4446 transports, and removes any duplicates. This means that more work will be
4447 done, as duplicates will always all be routed, but duplicates are
4448 presumably rare, so I don't expect this is of any significance.
4450 For deliveries to pipes, files, and autoreplies, the duplicate checking
4451 still happens during the routing process, since they are not going to be
4454 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
4455 It corrects a timeout issue with spamd. This is Ian's comment: "The
4456 background is that sometimes spamd either never reads data from a
4457 connection it has accepted, or it never writes response data. The exiscan
4458 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
4459 blindly assumes that writes won't block so it may never time out."
4461 PH/39 Allow G after quota size as well as K and M.
4463 PH/40 The value set for $authenticated_id in an authenticator may not contain
4464 binary zeroes or newlines because the value is written to log lines and
4465 to spool files. There was no check on this. Now the value is run through
4466 the string_printing() function so that such characters are converted to
4467 printable escape sequences.
4469 PH/41 $message_linecount is a new variable that contains the total number of
4470 lines in the message. Compare $body_linecount, which is the count for the
4473 PH/42 Exim no longer gives details of delivery errors for specific addresses in
4474 bounce and delay warning messages, except in certain special cases, which
4477 (a) An SMTP error message from a remote host;
4478 (b) A message specified in a :fail: redirection;
4479 (c) A message specified in a "fail" command in a system filter;
4480 (d) A message specified in a FAIL return from the queryprogram router;
4481 (e) A message specified by the cannot_route_message router option.
4483 In these cases only, Exim does include the error details in bounce and
4484 warning messages. There are also a few cases where bland messages such
4485 as "unrouteable address" or "local delivery error" are given.
4487 PH/43 $value is now also set for the "else" part of a ${run expansion.
4489 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
4490 being worked on, but at least Exim now implements the latest version to
4493 PH/45 In a pipe transport, although a timeout while waiting for the pipe
4494 process to complete was treated as a delivery failure, a timeout while
4495 writing the message to the pipe was logged, but erroneously treated as a
4496 successful delivery. Such timeouts include transport filter timeouts. For
4497 consistency with the overall process timeout, these timeouts are now
4498 treated as errors, giving rise to delivery failures by default. However,
4499 there is now a new Boolean option for the pipe transport called
4500 timeout_defer, which, if set TRUE, converts the failures into defers for
4501 both kinds of timeout. A transport filter timeout is now identified in
4504 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
4505 systems where "make" and "gmake" are different, calling "gmake" at top
4506 level broke things. I've arranged for the value of $(MAKE) to be passed
4507 from the Makefile to this script so that it can call the same version of
4511 A note about Exim versions 4.44 and 4.50
4512 ----------------------------------------
4514 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
4515 changes of various kinds. As a consequence, a big documentation update was
4516 needed. This delayed the release for rather longer than seemed good, especially
4517 in the light of a couple of (minor) security issues. Therefore, the changes
4518 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
4519 release. So 4.44 and 4.50 are in effect two different branches that both start
4522 I have left the 4.50 change log unchanged; it contains all the changes since
4523 4.43. The change log for 4.44 is below; many of its items are identical to
4524 those for 4.50. This seems to be the most sensible way to preserve the
4525 historical information.
4531 1. Minor wording change to the doc/README.SIEVE file.
4533 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
4534 computation of the current number of files was incorrect.
4536 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4537 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4540 4. Give more explanation in the error message when the command for a transport
4541 filter fails to execute.
4543 5. There are several places where Exim runs a non-Exim command in a
4544 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4545 was being done only for the command run by the queryprogram router. It is
4546 now done for all such subprocesses. The other cases are: ${run, transport
4547 filters, and the commands run by the lmtp and pipe transports.
4549 6. Added CONFIGURE_GROUP build-time option.
4551 7. Some older OS have a limit of 256 on the maximum number of file
4552 descriptors. Exim was using setrlimit() to set 1000 as a large value
4553 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4554 systems. I've change it so that if it can't get 1000, it tries for 256.
4556 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4557 was an oversight, and furthermore, ever since the addition of extra
4558 controls (e.g. 4.43/32), the checks on when to allow different forms of
4559 "control" were broken. There should now be diagnostics for all cases when a
4560 control that does not make sense is encountered.
4562 9. Added the /retain_sender option to "control=submission".
4564 10. $recipients is now available in the predata ACL (oversight).
4566 11. Tidy the search cache before the fork to do a delivery from a message
4567 received from the command line. Otherwise the child will trigger a lookup
4568 failure and thereby defer the delivery if it tries to use (for example) a
4569 cached ldap connection that the parent has called unbind on.
4571 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4572 of $address_data from the recipient verification was clobbered by the
4573 sender verification.
4575 13. The value of address_data from a sender verification is now available in
4576 $sender_address_data in subsequent conditions in the ACL statement.
4578 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
4580 15. Added a new option "connect=<time>" to callout options, to set a different
4583 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4584 was its contents. (It was OK if the option was not defined at all.)
4586 17. A "Completed" log line is now written for messages that are removed from
4587 the spool by the -Mrm option.
4589 18. New variables $sender_verify_failure and $recipient_verify_failure contain
4590 information about exactly what failed.
4592 19. Added -dd to debug only the daemon process.
4594 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
4595 handles timeouts, both on the server side and network timeouts. Renamed the
4596 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
4598 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
4599 It is now set to "smtps".
4601 22. $host_address is now set to the target address during the checking of
4602 ignore_target_hosts.
4604 23. When checking ignore_target_hosts for an ipliteral router, no host name was
4605 being passed; this would have caused $sender_host_name to have been used if
4606 matching the list had actually called for a host name (not very likely,
4607 since this list is usually IP addresses). A host name is now passed as
4610 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4611 code that specifies a non-restarting handler (typically sigaction() in
4612 modern systems) in an attempt to fix a rare and obscure crash bug.
4614 25. Narrowed the window for a race in the daemon that could cause it to ignore
4615 SIGCHLD signals. This is not a major problem, because they are used only to
4616 wake it up if nothing else does.
4618 26. A malformed maildirsize file could cause Exim to calculate negative values
4619 for the mailbox size or file count. Odd effects could occur as a result.
4620 The maildirsize information is now recalculated if the size or filecount
4623 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4624 support for a long time. Removed HAVE_SYS_VFS_H.
4626 28. Installed the latest version of exipick from John Jetmore.
4628 29. In an address list, if the pattern was not a regular expression, an empty
4629 subject address (from a bounce message) matched only if the pattern was an
4630 empty string. Non-empty patterns were not even tested. This was the wrong
4631 because it is perfectly reasonable to use an empty address as part of a
4632 database query. An empty address is now tested by patterns that are
4633 lookups. However, all the other forms of pattern expect the subject to
4634 contain a local part and a domain, and therefore, for them, an empty
4635 address still always fails if the pattern is not itself empty.
4637 30. Exim went into a mad DNS loop when attempting to do a callout where the
4638 host was specified on an smtp transport, and looking it up yielded more
4639 than one IP address.
4641 31. Re-factored the code for checking spool and log partition space into a
4642 function that finds that data and another that does the check. The former
4643 is then used to implement four new variables: $spool_space, $log_space,
4644 $spool_inodes, and $log_inodes.
4646 32. The RFC2047 encoding function was originally intended for short strings
4647 such as real names; it was not keeping to the 75-character limit for
4648 encoded words that the RFC imposes. It now respects the limit, and
4649 generates multiple encoded words if necessary. To be on the safe side, I
4650 have increased the buffer size for the ${rfc2047: expansion operator from
4653 33. It is now permitted to omit both strings after an "if" condition; if the
4654 condition is true, the result is "true". As before, when the second string
4655 is omitted, a false condition yields an empty string. This makes it less
4656 cumbersome to write custom ACL and router conditions.
4658 34. Failure to deliver a bounce message always caused it to be frozen, even if
4659 there was an errors_to setting on the router. The errors_to setting is now
4662 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4663 canonical form (fully expanded) before being placed in
4664 $sender_host_address.
4666 36. The table in the code that translates DNS record types into text (T_A to
4667 "A" for instance) was missing entries for NS and CNAME. It is just possible
4668 that this could have caused confusion if both these types were looked up
4669 for the same domain, because the text type is used as part of Exim's
4670 per-process caching. But the chance of anyone hitting this buglet seems
4673 37. The dnsdb lookup has been extended in a number of ways.
4675 (1) There is a new type, "zns", which walks up the domain tree until it
4676 finds some nameserver records. It should be used with care.
4678 (2) There is a new type, "mxh", which is like "mx" except that it returns
4679 just the host names, not the priorities.
4681 (3) It is now possible to give a list of domains (or IP addresses) to be
4682 looked up. The behaviour when one of the lookups defers can be
4683 controlled by a keyword.
4685 (4) It is now possible to specify the separator character for use when
4686 multiple records are returned.
4688 38. The dnslists ACL condition has been extended: it is now possible to supply
4689 a list of IP addresses and/or domains to be looked up in a particular DNS
4692 39. Added log_selector=+queue_time_overall.
4694 40. When running the queue in the test harness, wait just a tad after forking a
4695 delivery process, to get repeatability of debugging output.
4697 41. Include certificate and key file names in error message when GnuTLS fails
4698 to set them up, because the GnuTLS error message doesn't include the name
4699 of the failing file when there is a problem reading it.
4701 42. Allow both -bf and -bF in the same test run.
4703 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
4705 44. The "Exiscan patch" is now merged into the mainline Exim source.
4707 45. Sometimes the final signoff response after QUIT could fail to get
4708 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4709 < 0 before doing a fflush(). This bug looks as though it goes back to the
4710 introduction of TLS in release 3.20, but "sometimes" must have been rare
4711 because the tests only now provoked it.
4713 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
4714 (this can affect the format of dates).
4716 47. exim_tidydb, when checking for the continued existence of a message for
4717 which it has found a message-specific retry record, was not finding
4718 messages that were in split spool directories. Consequently, it was
4719 deleting retry records that should have stayed in existence.
4721 48. Steve fixed some bugs in eximstats.
4723 49. The SPA authentication driver was not abandoning authentication and moving
4724 on to the next authenticator when an expansion was forced to fail,
4725 contradicting the general specification for all authenticators. Instead it
4726 was generating a temporary error. It now behaves as specified.
4728 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
4729 (the order specifies the preference for clients). The order is now AES256,
4730 AES128, 3DES, ARCFOUR128.
4732 51. Small patch to Sieve code - explicitly set From: when generating an
4735 52. Exim crashed if a remote delivery caused a very long error message to be
4736 recorded - for instance if somebody sent an entire SpamAssassin report back
4737 as a large number of 550 error lines. This bug was coincidentally fixed by
4738 increasing the size of one of Exim's internal buffers (big_buffer) that
4739 happened as part of the Exiscan merge. However, to be on the safe side, I
4740 have made the code more robust (and fixed the comments that describe what
4743 53. Now that there can be additional text after "Completed" in log lines (if
4744 the queue_time_overall log selector is set), a one-byte patch to exigrep
4745 was needed to allow it to recognize "Completed" as not the last thing in
4748 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
4749 patch that reportedly fixes this has been added. I am not expert enough to
4750 create a test for it. This is what the patch creator wrote:
4752 "I found a little strange behaviour of ldap code when working with
4753 Windows 2003 AD Domain, where users was placed in more than one
4754 Organization Units. When I tried to give exim partial DN, the exit code
4755 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
4756 But simultaneously result of request was absolutely normal ldap result,
4757 so I produce this patch..."
4759 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
4760 REFERENCE, so I have modified the code to exclude the patch when that macro
4763 55. Some experimental protocols are using DNS PTR records for new purposes. The
4764 keys for these records are domain names, not reversed IP addresses. The
4765 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4766 leaves it alone. Component reversal etc. now happens only for IP addresses.
4769 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
4771 57. Double the size of the debug message buffer (to 2048) so that more of very
4772 long debug lines gets shown.
4774 58. The exicyclog utility now does better if the number of log files to keep
4775 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
4777 59. Two changes related to the smtp_active_hostname option:
4779 (1) $smtp_active_hostname is now available as a variable.
4780 (2) The default for smtp_banner uses $smtp_active_hostname instead
4781 of $primary_hostname.
4783 60. The host_aton() function is supposed to be passed a string that is known
4784 to be a valid IP address. However, in the case of IPv6 addresses, it was
4785 not checking this. This is a hostage to fortune. Exim now panics and dies
4786 if the condition is not met. A case was found where this could be provoked
4787 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4788 components; fortuitously, this particular loophole had already been fixed
4789 by change 4.50/55 above.
4791 If there are any other similar loopholes, the new check in host_aton()
4792 itself should stop them being exploited. The report I received stated that
4793 data on the command line could provoke the exploit when Exim was running as
4794 exim, but did not say which command line option was involved. All I could
4795 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4796 running as the user.
4799 61. There was a buffer overflow vulnerability in the SPA authentication code
4800 (which came originally from the Samba project). I have added a test to the
4801 spa_base64_to_bits() function which I hope fixes it.
4804 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
4805 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
4807 63. The daemon start-up calls getloadavg() while still root for those OS that
4808 need the first call to be done as root, but it missed one case: when
4809 deliver_queue_load_max is set with deliver_drop_privilege. This is
4810 necessary for the benefit of the queue runner, because there is no re-exec
4811 when deliver_drop_privilege is set.
4813 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
4814 This has been fixed.
4816 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4817 was in use, was not putting the data itself into the right store pool;
4818 consequently, it could be overwritten for a subsequent message in the same
4819 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4822 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
4824 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
4825 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
4826 the function were treating the return as a boolean value, which happened to
4827 work because 0=false and not-0=true, but is not correct code.
4829 68. The host_aton() function was not handling scoped IPv6 addresses (those
4830 with, for example, "%eth0" on the end) correctly.
4832 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
4833 negated items (that is, ~something) in unsigned ints. Some compilers
4834 apparently mutter when there is no cast.
4836 70. If an address verification called from an ACL failed, and did not produce a
4837 user-specific message (i.e. there was only a "system" message), nothing was
4838 put in $acl_verify_message. In this situation, it now puts the system
4841 71. Change 4.23/11 added synchronization checking at the start of an SMTP
4842 session; change 4.31/43 added the unwanted input to the log line - except
4843 that it did not do this in the start of session case. It now does.
4845 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
4846 This is wrong and can cause the other end to generate a synchronization
4847 error if it is another Exim or anything else that does the synchronization
4848 check. A QUIT command is no longer sent after a timeout.
4850 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
4851 during host lookups.
4853 74. The defer_ok option of callout verification was not working if it was used
4854 when verifying addresses in header lines, that is, for this case:
4856 verify = header_sender/callout=defer_ok
4858 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
4859 those file descriptors could be used for SMTP connections. If anything
4860 wrote to stderr (the example that came up was "warn" in embedded Perl), it
4861 could be sent to the SMTP client, causing chaos. The daemon now opens
4862 stdin, stdout, and stderr to /dev/null when it puts itself into the
4865 76. Arrange for output from Perl's "warn" command to be written to Exim's main
4866 log by default. The user can override this with suitable Perl magic.
4868 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
4869 the log message when discard triggers, was not working for the DATA ACL or
4870 for the non-SMTP ACL.
4872 78. Error message wording change in sieve.c.
4874 79. If smtp_accept_max_per_host was set, the number of connections could be
4875 restricted to fewer than expected, because the daemon was trying to set up
4876 a new connection before checking whether the processes handling previous
4877 connections had finished. The check for completed processes is now done
4878 earlier. On busy systems, this bug wouldn't be noticed because something
4879 else would have woken the daemon, and it would have reaped the completed
4882 80. If a message was submitted locally by a user whose login name contained one
4883 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
4884 It caused a spool format error. I have fixed the spool reading code. A
4885 related problem was that the "from" clause in the Received: line became
4886 illegal because of the space(s). It is now covered by ${quote_local_part.
4888 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
4891 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
4893 83. Patch from Sieve maintainer to fix unterminated string problem in
4894 "vacation" handling.
4896 84. Some minor changes to the Linux configuration files to help with other
4897 OS variants using glibc.
4899 85. One more patch for Sieve to update vacation handling to latest spec.
4902 ----------------------------------------------------
4903 See the note above about the 4.44 and 4.50 releases.
4904 ----------------------------------------------------
4910 1. Change 4.43/35 introduced a bug that caused file counts to be
4911 incorrectly computed when quota_filecount was set in an appendfile
4914 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4915 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4918 3. Give more explanation in the error message when the command for a transport
4919 filter fails to execute.
4921 4. There are several places where Exim runs a non-Exim command in a
4922 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4923 was being done only for the command run by the queryprogram router. It is
4924 now done for all such subprocesses. The other cases are: ${run, transport
4925 filters, and the commands run by the lmtp and pipe transports.
4927 5. Some older OS have a limit of 256 on the maximum number of file
4928 descriptors. Exim was using setrlimit() to set 1000 as a large value
4929 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4930 systems. I've change it so that if it can't get 1000, it tries for 256.
4932 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4933 was an oversight, and furthermore, ever since the addition of extra
4934 controls (e.g. 4.43/32), the checks on when to allow different forms of
4935 "control" were broken. There should now be diagnostics for all cases when a
4936 control that does not make sense is encountered.
4938 7. $recipients is now available in the predata ACL (oversight).
4940 8. Tidy the search cache before the fork to do a delivery from a message
4941 received from the command line. Otherwise the child will trigger a lookup
4942 failure and thereby defer the delivery if it tries to use (for example) a
4943 cached ldap connection that the parent has called unbind on.
4945 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4946 of $address_data from the recipient verification was clobbered by the
4947 sender verification.
4949 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4950 was its contents. (It was OK if the option was not defined at all.)
4952 11. A "Completed" log line is now written for messages that are removed from
4953 the spool by the -Mrm option.
4955 12. $host_address is now set to the target address during the checking of
4956 ignore_target_hosts.
4958 13. When checking ignore_target_hosts for an ipliteral router, no host name was
4959 being passed; this would have caused $sender_host_name to have been used if
4960 matching the list had actually called for a host name (not very likely,
4961 since this list is usually IP addresses). A host name is now passed as
4964 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4965 code that specifies a non-restarting handler (typically sigaction() in
4966 modern systems) in an attempt to fix a rare and obscure crash bug.
4968 15. Narrowed the window for a race in the daemon that could cause it to ignore
4969 SIGCHLD signals. This is not a major problem, because they are used only to
4970 wake it up if nothing else does.
4972 16. A malformed maildirsize file could cause Exim to calculate negative values
4973 for the mailbox size or file count. Odd effects could occur as a result.
4974 The maildirsize information is now recalculated if the size or filecount
4977 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4978 support for a long time. Removed HAVE_SYS_VFS_H.
4980 18. Updated exipick to current release from John Jetmore.
4982 19. Allow an empty sender to be matched against a lookup in an address list.
4983 Previously the only cases considered were a regular expression, or an
4986 20. Exim went into a mad DNS lookup loop when doing a callout where the
4987 host was specified on the transport, if the DNS lookup yielded more than
4990 21. The RFC2047 encoding function was originally intended for short strings
4991 such as real names; it was not keeping to the 75-character limit for
4992 encoded words that the RFC imposes. It now respects the limit, and
4993 generates multiple encoded words if necessary. To be on the safe side, I
4994 have increased the buffer size for the ${rfc2047: expansion operator from
4997 22. Failure to deliver a bounce message always caused it to be frozen, even if
4998 there was an errors_to setting on the router. The errors_to setting is now
5001 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
5002 canonical form (fully expanded) before being placed in
5003 $sender_host_address.
5005 24. Updated eximstats to version 1.33
5007 25. Include certificate and key file names in error message when GnuTLS fails
5008 to set them up, because the GnuTLS error message doesn't include the name
5009 of the failing file when there is a problem reading it.
5011 26. Expand error message when OpenSSL has problems setting up cert/key files.
5014 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
5015 (this can affect the format of dates).
5017 28. exim_tidydb, when checking for the continued existence of a message for
5018 which it has found a message-specific retry record, was not finding
5019 messages that were in split spool directories. Consequently, it was
5020 deleting retry records that should have stayed in existence.
5022 29. eximstats updated to version 1.35
5023 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
5024 1.35 - bugfix such that pie charts by volume are generated correctly
5026 30. The SPA authentication driver was not abandoning authentication and moving
5027 on to the next authenticator when an expansion was forced to fail,
5028 contradicting the general specification for all authenticators. Instead it
5029 was generating a temporary error. It now behaves as specified.
5031 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
5032 (the order specifies the preference for clients). The order is now AES256,
5033 AES128, 3DES, ARCFOUR128.
5035 31. Small patch to Sieve code - explicitly set From: when generating an
5038 32. Exim crashed if a remote delivery caused a very long error message to be
5039 recorded - for instance if somebody sent an entire SpamAssassin report back
5040 as a large number of 550 error lines. This bug was coincidentally fixed by
5041 increasing the size of one of Exim's internal buffers (big_buffer) that
5042 happened as part of the Exiscan merge. However, to be on the safe side, I
5043 have made the code more robust (and fixed the comments that describe what
5046 33. Some experimental protocols are using DNS PTR records for new purposes. The
5047 keys for these records are domain names, not reversed IP addresses. The
5048 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
5049 leaves it alone. Component reversal etc. now happens only for IP addresses.
5052 34. The host_aton() function is supposed to be passed a string that is known
5053 to be a valid IP address. However, in the case of IPv6 addresses, it was
5054 not checking this. This is a hostage to fortune. Exim now panics and dies
5055 if the condition is not met. A case was found where this could be provoked
5056 from a dnsdb PTR lookup with an IPv6 address that had more than 8
5057 components; fortuitously, this particular loophole had already been fixed
5058 by change 4.50/55 or 4.44/33 above.
5060 If there are any other similar loopholes, the new check in host_aton()
5061 itself should stop them being exploited. The report I received stated that
5062 data on the command line could provoke the exploit when Exim was running as
5063 exim, but did not say which command line option was involved. All I could
5064 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
5065 running as the user.
5068 35. There was a buffer overflow vulnerability in the SPA authentication code
5069 (which came originally from the Samba project). I have added a test to the
5070 spa_base64_to_bits() function which I hope fixes it.
5073 36. The daemon start-up calls getloadavg() while still root for those OS that
5074 need the first call to be done as root, but it missed one case: when
5075 deliver_queue_load_max is set with deliver_drop_privilege. This is
5076 necessary for the benefit of the queue runner, because there is no re-exec
5077 when deliver_drop_privilege is set.
5079 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
5080 was in use, was not putting the data itself into the right store pool;
5081 consequently, it could be overwritten for a subsequent message in the same
5082 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
5085 38. Sometimes the final signoff response after QUIT could fail to get
5086 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
5087 < 0 before doing a fflush(). This bug looks as though it goes back to the
5088 introduction of TLS in release 3.20, but "sometimes" must have been rare
5089 because the tests only now provoked it.
5095 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
5096 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
5097 Now it returns an integer. A number of calls were still expecting a T/F
5098 return. Fortuitously, in all cases, the tests worked in OK situations,
5099 which is the norm. However, things would have gone wrong on any write
5100 failures on the smtp file descriptor. This function is used when sending
5101 messages over SMTP and also when doing verify callouts.
5103 2. When Exim is called to do synchronous delivery of a locally submitted
5104 message (the -odf or -odi options), it no longer closes stderr before doing
5107 3. Implemented the mua_wrapper option.
5109 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
5111 5. Implemented the functions header_remove(), header_testname(),
5112 header_add_at_position(), and receive_remove_recipient(), and exported them
5115 6. If an ACL "warn" statement specified the addition of headers, Exim already
5116 inserted X-ACL-Warn: at the start if there was no header name. However, it
5117 was not making this test for the second and subsequent header lines if
5118 there were newlines in the string. This meant that an invalid header could
5119 be inserted if Exim was badly configured.
5121 7. Allow an ACL "warn" statement to add header lines at the start or after all
5122 the Received: headers, as well as at the end.
5124 8. Added the rcpt_4xx retry error code.
5126 9. Added postmaster_mailfrom=xxx to callout verification option.
5128 10. Added mailfrom=xxxx to the callout verification option, for verify=
5131 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
5132 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
5134 12. Inserted some casts to stop certain compilers warning when using pointer
5135 differences as field lengths or precisions in printf-type calls (mostly
5136 affecting debugging statements).
5138 13. Added optional readline() support for -be (dynamically loaded).
5140 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
5141 same clock tick as a message's arrival, so that its received time was the
5142 same as the "first fail" time on the retry record, and that message
5143 remained on the queue past the ultimate address timeout, every queue runner
5144 would try a delivery (because it was past the ultimate address timeout) but
5145 after another failure, the ultimate address timeout, which should have then
5146 bounced the address, did not kick in. This was a "< instead of <=" error;
5147 in most cases the first failure would have been in the next clock tick
5148 after the received time, and all would be well.
5150 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
5151 being recognized when the domain list was tested by the match_domain
5152 condition in an expansion string.
5154 16. Added the ${str2b64: operator.
5156 17. Exim was always calling setrlimit() to set a large limit for the number of
5157 processes, without checking whether the existing limit was already
5158 adequate. (It did check for the limit on file descriptors.) Furthermore,
5159 errors from getrlimit() and setrlimit() were being ignored. Now they are
5160 logged to the main and panic logs, but Exim does carry on, to try to do its
5161 job under whatever limits there are.
5163 18. Imported PCRE 5.0.
5165 19. Trivial typo in log message " temporarily refused connection" (the leading
5168 20. If the log selector return_path_on_delivery was set and an address was
5169 redirected to /dev/null, the delivery process crashed because it assumed
5170 that a return path would always be set for a "successful" delivery. In this
5171 case, the whole delivery is bypassed as an optimization, and therefore no
5174 21. Internal re-arrangement: the function for sending a challenge and reading
5175 a response while authentication was assuming a zero-terminated challenge
5176 string. It's now changed to take a pointer and a length, to allow for
5177 binary data in such strings.
5179 22. Added the cyrus_sasl authenticator (code supplied by MBM).
5181 23. Exim was not respecting finduser_retries when seeking the login of the
5182 uid under which it was called; it was always trying 10 times. (The default
5183 setting of finduser_retries is zero.) Also, it was sleeping after the final
5184 failure, which is pointless.
5186 24. Implemented tls_on_connect_ports.
5188 25. Implemented acl_smtp_predata.
5190 26. If the domain in control=submission is set empty, Exim assumes that the
5191 authenticated id is a complete email address when it generates From: or
5192 Sender: header lines.
5194 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
5195 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
5196 chown and chgrp in /bin and hostname in /usr/bin.
5198 28. Exim was keeping the "process log" file open after each use, just as it
5199 does for the main log. This opens the possibility of it remaining open for
5200 long periods when the USR1 signal hits a daemon. Occasional processlog
5201 errors were reported, that could have been caused by this. Anyway, it seems
5202 much more sensible not to leave this file open at all, so that is what now
5205 29. The long-running daemon process does not normally write to the log once it
5206 has entered its main loop, and it closes the log before doing so. This is
5207 so that log files can straightforwardly be renamed and moved. However,
5208 there are a couple of unusual error situations where the daemon does write
5209 log entries, and I had neglected to close the log afterwards.
5211 30. The text of an SMTP error response that was received during a remote
5212 delivery was being truncated at 512 bytes. This is too short for some of
5213 the long messages that one sometimes sees. I've increased the limit to
5216 31. It is now possible to make retry rules that apply only when a message has a
5217 specific sender, in particular, an empty sender.
5219 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
5220 it possible to be selective about when SMTP synchronization is enforced.
5222 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
5224 32. Implemented hosts_connection_nolog.
5226 33. Added an ACL for QUIT.
5228 34. Setting "delay_warning=" to disable warnings was not working; it gave a
5231 35. Added mailbox_size and mailbox_filecount to appendfile.
5233 36. Added control = no_multiline_responses to ACLs.
5235 37. There was a bug in the logic of the code that waits for the clock to tick
5236 in the case where the clock went backwards by a substantial amount such
5237 that the microsecond fraction of "now" was more than the microsecond
5238 fraction of "then" (but the whole seconds number was less).
5240 38. Added support for the libradius Radius client library this is found on
5241 FreeBSD (previously only the radiusclient library was supported).
5247 1. When certain lookups returned multiple values in the form name=value, the
5248 quoting of the values was not always being done properly. Specifically:
5249 (a) If the value started with a double quote, but contained no whitespace,
5251 (b) If the value contained whitespace other than a space character (i.e.
5252 tabs or newlines or carriage returns) it was not quoted.
5253 This fix has been applied to the mysql and pgsql lookups by writing a
5254 separate quoting function and calling it from the lookup code. The fix
5255 should probably also be applied to nisplus, ibase and oracle lookups, but
5256 since I cannot test any of those, I have not disturbed their existing code.
5258 2. A hit in the callout cache for a specific address caused a log line with no
5259 reason for rejecting RCPT. Now it says "Previous (cached) callout
5260 verification failure".
5262 3. There was an off-by-one bug in the queryprogram router. An over-long
5263 return line was truncated at 256 instead of 255 characters, thereby
5264 overflowing its buffer with the terminating zero. As well as fixing this, I
5265 have increased the buffer size to 1024 (and made a note to document this).
5267 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
5268 when Exim is waiting for an SMTP response from a remote server, Exim
5269 restarts its select() call on the socket, thereby resetting its timeout.
5270 This is not a problem when such interrupts are rare. Somebody set up a cron
5271 job to run exiwhat every 2 minutes, which is less than the normal select()
5272 timeout (5 or 10 minutes). This meant that the select() timeout never
5273 kicked in because it was always reset. I have fixed this by comparing the
5274 time when an interrupt arrives with the time at the start of the first call
5275 to select(). If more time than the timeout has elapsed, the interrupt is
5276 treated as a timeout.
5278 5. Some internal re-factoring in preparation for the addition of Sieve
5279 extensions (by MH). In particular, the "personal" test is moved to a
5280 separate function, and given an option for scanning Cc: and Bcc: (which is
5281 not set for Exim filters).
5283 6. When Exim created an email address using the login of the caller as the
5284 local part (e.g. when creating a From: or Sender: header line), it was not
5285 quoting the local part when it contained special characters such as @.
5287 7. Installed new OpenBSD configuration files.
5289 8. Reworded some messages for syntax errors in "and" and "or" conditions to
5290 try to make them clearer.
5292 9. Callout options, other than the timeout value, were being ignored when
5293 verifying sender addresses in header lines. For example, when using
5295 verify = header_sender/callout=no_cache
5297 the cache was (incorrectly) being used.
5299 10. Added a missing instance of ${EXE} to the exim_install script; this affects
5300 only the Cygwin environment.
5302 11. When return_path_on_delivery was set as a log selector, if different remote
5303 addresses in the same message used different return paths and parallel
5304 remote delivery occurred, the wrong values would sometimes be logged.
5305 (Whenever a remote delivery process finished, the return path value from
5306 the most recently started remote delivery process was logged.)
5308 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
5309 lines when AUTH and/or TLS are in use. This is the "received protocol"
5310 field. Exim used to use "asmtp" for authenticated SMTP, without any
5311 indication (in the protocol name) for TLS use. Now it follows the RFC and
5312 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
5313 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
5314 names appear in log lines as well as in Received: header lines.
5316 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
5317 extensions, and comparison tests, and to fix some bugs.
5319 14. Changes to the "personal" filter test:
5321 (1) The test was buggy in that it was just doing the equivalent of
5322 "contains" tests on header lines. For example, if a user's address was
5323 anne@some.where, the "personal" test would incorrectly be true for
5325 To: susanne@some.where
5327 This test is now done by extracting each address from the header in turn,
5328 and checking the entire address. Other tests that are part of "personal"
5329 are now done using regular expressions (for example, to check local parts
5330 of addresses in From: header lines).
5332 (2) The list of non-personal local parts in From: addresses has been
5333 extended to include "listserv", "majordomo", "*-request", and "owner-*",
5334 taken from the Sieve specification recommendations.
5336 (3) If the message contains any header line starting with "List-" it is
5337 treated as non-personal.
5339 (4) The test for "circular" in the Subject: header line has been removed
5340 because it now seems ill-conceived.
5342 15. Minor typos in src/EDITME comments corrected.
5344 16. Installed latest exipick from John Jetmore.
5346 17. If headers_add on a router specified a text string that was too long for
5347 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
5348 of string_sprintf() is now avoided.
5350 18. $message_body_size was not set (it was always zero) when running the DATA
5351 ACL and the local_scan() function.
5353 19. For the "mail" command in an Exim filter, no default was being set for
5354 the once_repeat time, causing a random time value to be used if "once" was
5355 specified. (If the value happened to be <= 0, no repeat happened.) The
5356 default is now 0s, meaning "never repeat". The "vacation" command was OK
5357 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
5358 (I found it when inspecting the code).
5360 20. There is now an overall timeout for performing a callout verification. It
5361 defaults to 4 times the callout timeout, which applies to individual SMTP
5362 commands during the callout. The overall timeout applies when there is more
5363 than one host that can be tried. The timeout is checked before trying the
5364 next host. This prevents very long delays if there are a large number of
5365 hosts and all are timing out (e.g. when the network connections are timing
5366 out). The value of the overall timeout can be changed by specifying an
5367 additional sub-option for "callout", called "maxwait". For example:
5369 verify = sender/callout=5s,maxwait=20s
5371 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
5372 to the end before writing, but this should make it even safer).
5374 22. Exim was forgetting that it had advertised PIPELINING for the second and
5375 subsequent messages on an SMTP connection. It was also not resetting its
5376 memory on STARTTLS and an internal HELO.
5378 23. When Exim logs an SMTP synchronization error within a session, it now
5379 records whether PIPELINING has been advertised or not.
5381 24. Added 3 instances of "(long int)" casts to time_t variables that were being
5382 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
5383 rather than long int.
5385 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
5387 26. Added the never_mail option to autoreply.
5393 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
5394 crash if the getsockname() call failed; this can happen if a connection is
5395 closed very soon after it is established. The problem was simply in the
5396 order in which certain operations were done, causing Exim to try to write
5397 to the SMTP stream before it had set up the file descriptor. The bug has
5398 been fixed by making things happen in the correct order.
5404 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
5405 before the connection was closed, thus losing the rejection response.
5407 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
5408 some early Solaris releases, but causes trouble in current releases where
5409 socklen_t is defined.
5411 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
5414 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
5417 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
5418 that releasing the top of it at the end releases what was used for sub-
5419 expansions (unless the block got too big). However, discard this block if
5420 the first thing is a variable or header, so that we can use its block when
5421 it is dynamic (useful for very large $message_headers, for example).
5423 6. Lookups now cache *every* query, not just the most recent. A new, separate
5424 store pool is used for this. It can be recovered when all lookup caches are
5425 flushed. Lookups now release memory at the end of their result strings.
5426 This has involved some general refactoring of the lookup sources.
5428 7. Some code has been added to the store_xxx() functions to reduce the amount
5429 of flapping under certain conditions.
5431 8. log_incoming_interface used to affect only the <= reception log lines. Now
5432 it causes the local interface and port to be added to several more SMTP log
5433 lines, for example "SMTP connection from", and rejection lines.
5435 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
5437 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
5439 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
5440 could be overwritten at the end of the current message (or the start of a
5441 new message if it was set in a HELO ACL). The value is now preserved for
5442 the duration of the SMTP connection.
5444 12. If a transport had a headers_rewrite setting, and a matching header line
5445 contained an unqualified address, that address was qualified, even if it
5446 did not match any rewriting rules. The underlying bug was that the values
5447 of the flags that permit the existence of unqualified sender and recipient
5448 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
5449 non-local messages, and by -bnq for local messages) were not being
5450 preserved with the message after it was received.
5452 13. When Exim was logging an SMTP synchronization error, it could sometimes log
5453 "next input=" as part of the text comprising the host identity instead of
5454 the correct text. The code was using the same buffer for two different
5455 strings. However, depending on which order the printing function evaluated
5456 its arguments, the bug did not always show up. Under Linux, for example, my
5457 test suite worked just fine.
5459 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
5460 doesn't work with some older versions of Perl. It has been changed to "my",
5461 which in any case is probably the better facility to use.
5463 15. A really picky compiler found some instances of statements for creating
5464 error messages that either had too many or two few arguments for the format
5467 16. The size of the buffer for calls to the DNS resolver has been increased
5468 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
5469 for addresses that have a lot of PTR records. This alleviates a problem; it
5470 does not fully solve it.
5472 17. A dnsdb lookup for PTR records that receives more data than will fit in the
5473 buffer now truncates the list and logs the incident, which is the same
5474 action as happens when Exim is looking up a host name and its aliases.
5475 Previously in this situation something unpredictable would happen;
5476 sometimes it was "internal error: store_reset failed".
5478 18. If a server dropped the connection unexpectedly when an Exim client was
5479 using GnuTLS and trying to read a response, the client delivery process
5480 crashed while trying to generate an error log message.
5482 19. If a "warn" verb in an ACL added multiple headers to a message in a single
5483 string, for example:
5485 warn message = H1: something\nH2: something
5487 the text was added as a single header line from Exim's point of view
5488 though it ended up OK in the delivered message. However, searching for the
5489 second and subsequent header lines using $h_h2: did not work. This has been
5490 fixed. Similarly, if a system filter added multiple headers in this way,
5491 the routers could not see them.
5493 20. Expanded the error message when iplsearch is called with an invalid key to
5494 suggest using net-iplsearch in a host list.
5496 21. When running tests using -bh, any delays imposed by "delay" modifiers in
5497 ACLs are no longer actually imposed (and a message to that effect is
5500 22. If a "gecos" field in a passwd entry contained escaped characters, in
5501 particular, if it contained a \" sequence, Exim got it wrong when building
5502 a From: or a Sender: header from that name. A second bug also caused
5503 incorrect handling when an unquoted " was present following a character
5504 that needed quoting.
5506 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
5507 was not being matched caselessly.
5509 24. Arranged for all hyphens in the exim.8 source to be escaped with
5512 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
5513 properly. Recipient callout cache records were still being keyed to include
5514 the sender, even when use_sender was set false. This led to far more
5515 callouts that were necessary. The sender is no longer included in the key
5516 when use_sender is false.
5518 26. Added "control = submission" modifier to ACLs.
5520 27. Added the ${base62d: operator to decode base 62 numbers.
5522 28. dnsdb lookups can now access SRV records.
5524 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
5525 the configuration file.
5527 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
5528 (-v) mode. This makes the output for a verbose queue run more intelligible.
5530 31. Added a use_postmaster feature to recipient callouts.
5532 32. Added the $body_zerocount variable, containing the number of binary zero
5533 bytes in the message body.
5535 33. The time of last modification of the "new" subdirectory is now used as the
5536 "mailbox time last read" when there is a quota error for a maildir
5539 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
5541 35. Added +ignore_unknown as a special item in host lists.
5543 36. Code for decoding IPv6 addresses in host lists is now included, even if
5544 IPv6 support is not being compiled. This fixes a bug in which an IPv6
5545 address was recognized as an IP address, but was then not correctly decoded
5546 into binary, causing unexpected and incorrect effects when compared with
5553 1. Very minor rewording of debugging text in manualroute to say "list of
5554 hosts" instead of "hostlist".
5556 2. If verify=header_syntax was set, and a header line with an unqualified
5557 address (no domain) and a large number of spaces between the end of the
5558 name and the colon was received, the reception process suffered a buffer
5559 overflow, and (when I tested it) crashed. This was caused by some obsolete
5560 code that should have been removed. The fix is to remove it!
5562 3. When running in the test harness, delay a bit after writing a bounce
5563 message to get a bit more predictability in the log output.
5565 4. Added a call to search_tidyup() just before forking a reception process. In
5566 theory, someone could use a lookup in the expansion of smtp_accept_max_
5567 per_host which, without the tidyup, could leave open a database connection.
5569 5. Added the variables $recipient_data and $sender_data which get set from a
5570 lookup success in an ACL "recipients" or "senders" condition, or a router
5571 "senders" option, similar to $domain_data and $local_part_data.
5573 6. Moved the writing of debug_print from before to after the "senders" test
5576 7. Change 4.31/66 (moving the time when the Received: is generated) caused
5577 problems for message scanning, either using a data ACL, or using
5578 local_scan() because the Received: header was not generated till after they
5579 were called (in order to set the time as the time of reception completion).
5580 I have revised the way this works. The header is now generated after the
5581 body is received, but before the ACL or local_scan() are called. After they
5582 are run, the timestamp in the header is updated.
5588 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
5589 before starting a queue runner without re-exec. This happened only when
5590 deliver_drop_privilege was set or when the Exim user was set to root. The
5591 effect of the bug was that timeouts during subsequent deliveries caused
5592 crashes instead of being properly handled. The handler is now left at its
5593 default (and expected) setting.
5595 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
5596 message, again when deliver_drop_privilege is set or Exim is run as root.
5597 The bug described in (1) was not present in this case, but the tidying up
5598 of the other signals was missing. I have made the two cases consistent.
5600 3. The ignore_target_hosts setting on a manualroute router was being ignored
5601 for hosts that were looked up using the /MX notation.
5603 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
5606 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
5607 operated on the sender address. After changing the $sender_address to <>
5608 for the sender address verify, Exim was re-instated it as the original
5609 (before rewriting) address, but remembering that it had rewritten it, so it
5610 wasn't rewriting it again. This bug also had the effect of breaking the
5611 sender address verification caching when the sender address was rewritten.
5613 6. The ignore_target_hosts option was being ignored by the ipliteral router.
5614 This has been changed so that if the ip literal address matches
5615 ignore_target_hosts, the router declines.
5617 7. Added expansion conditions match_domain, match_address, and match_local_
5618 part (NOT match_host).
5620 8. The placeholder for the Received: header didn't have a length field set.
5622 9. Added code to Exim itself and to exim_lock to test for a specific race
5623 condition that could lead to file corruption when using MBX delivery. The
5624 issue is with the lockfile that is created in /tmp. If this file is removed
5625 after a process has opened it but before that process has acquired a lock,
5626 there is the potential for a second process to recreate the file and also
5627 acquire a lock. This could lead to two Exim processes writing to the file
5628 at the same time. The added code performs the same test as UW imapd; it
5629 checks after acquiring the lock that its file descriptor still refers to
5630 the same named file.
5632 10. The buffer for building added header lines was of fixed size, 8192 bytes.
5633 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
5636 11. Added the smtp_active_hostname option. If used, this will typically be made
5637 to depend on the incoming interface address. Because $interface_address is
5638 not set up until the daemon has forked a reception process, error responses
5639 that can happen earlier (such as "too many connections") no longer contain
5642 12. If an expansion in a condition on a "warn" statement fails because a lookup
5643 defers, the "warn" statement is abandoned, and the next ACL statement is
5644 processed. Previously this caused the whole ACL to be aborted.
5646 13. Added the iplsearch lookup type.
5648 14. Added ident_timeout as a log selector.
5650 15. Added tls_certificate_verified as a log selector.
5652 16. Added a global option tls_require_ciphers (compare the smtp transport
5653 option of the same name). This controls incoming TLS connections.
5655 17. I finally figured out how to make tls_require_ciphers do a similar thing
5656 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
5657 before starting the TLS session.
5659 18. Tabs are now shown as \t in -bP output.
5661 19. If the log selector return_path_on_delivery was set, Exim crashed when
5662 bouncing a message because it had too many Received: header lines.
5664 20. If two routers both had headers_remove settings, and the first one included
5665 a superfluous trailing colon, the final name in the first list and the
5666 first name in the second list were incorrectly joined into one item (with a
5667 colon in the middle).
5673 1. Added -C and -D options to the exinext utility, mainly to make it easier
5674 to include in the automated testing, but these could be helpful when
5675 multiple configurations are in use.
5677 2. The exinext utility was not formatting the output nicely when there was
5678 an alternate port involved in the retry record key, nor when there was a
5679 message id as well (for retries that were specific to a specific message
5680 and a specific host). It was also confused by IPv6 addresses, because of
5681 the additional colons they contain. I have fixed the IPv4 problem, and
5682 patched it up to do a reasonable job for IPv6.
5684 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
5685 delivery, the log line now contains "pipelined" if PIPELINING was used.
5687 4. An SMTP transport process used to panic and die if the bind() call to set
5688 an explicit outgoing interface failed. This has been changed; it is now
5689 treated in the same way as a connect() failure.
5691 5. A reference to $sender_host_name in the part of a conditional expansion
5692 that was being skipped was still causing a DNS lookup. This no longer
5695 6. The def: expansion condition was not recognizing references to header lines
5696 that used bh_ and bheader_.
5698 7. Added the _cache feature to named lists.
5700 8. The code for checking quota_filecount in the appendfile transport was
5701 allowing one more file than it should have been.
5703 9. For compatibility with Sendmail, the command line option
5711 and sets the incoming protocol and host name (for trusted callers). The
5712 host name and its colon can be omitted when only the protocol is to be set.
5713 Note the Exim already has two private options, -pd and -ps, that refer to
5714 embedded Perl. It is therefore impossible to set a protocol value of "d" or
5715 "s", but I don't think that's a major issue.
5717 10. A number of refactoring changes to the code, none of which should affect
5720 (a) The number of logging options was getting close to filling up the
5721 32-bit word that was used as a bit map. I have split them into two classes:
5722 those that are passed in the argument to log_write(), and those that are
5723 only ever tested independently outside of that function. These are now in
5724 separate 32-bit words, so there is plenty of room for expansion again.
5725 There is no change in the user interface or the logging behaviour.
5727 (b) When building, for example, log lines, the code previously used a
5728 macro that called string_cat() twice, in order to add two strings. This is
5729 not really sufficiently general. Furthermore, there was one instance where
5730 it was actually wrong because one of the argument was used twice, and in
5731 one call a function was used. (As it happened, calling the function twice
5732 did not affect the overall behaviour.) The macro has been replaced by a
5733 function that can join an arbitrary number of extra strings onto a growing
5736 (c) The code for expansion conditions now uses a table and a binary chop
5737 instead of a serial search (which was left over from when there were very
5738 few conditions). Also, it now recognizes conditions like "pam" even when
5739 the relevant support is not compiled in: a suitably worded error message is
5740 given if an attempt is made to use such a condition.
5742 11. Added ${time_interval:xxxxx}.
5744 12. A bug was causing one of the ddress fields not to be passed back correctly
5745 from remote delivery subprocesses. The field in question was not being
5746 subsequently used, so this caused to problems in practice.
5748 13. Added new log selectors queue_time and deliver_time.
5750 14. Might have fixed a bug in maildirsizefile handling that threw up
5751 "unexpected character" debug warnings, and recalculated the data
5752 unnecessarily. In any case, I expanded the warning message to give more
5755 15. Added the message "Restricted characters in address" to the statements in
5756 the default ACL that block characters like @ and % in local parts.
5758 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
5759 Three changes have been made:
5761 (a) There was a serious bug; a negative response to MAIL caused the whole
5762 recipient domain to be cached as invalid, thereby blocking all messages
5763 to all local parts at the same domain, from all senders. This bug has
5764 been fixed. The domain is no longer cached after a negative response to
5765 MAIL if the sender used is not empty.
5767 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
5770 (c) A new callout option, "use_sender" has been added for people who want
5771 the modified behaviour.
5777 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
5780 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
5781 indeed breaks things for older releases.
5783 3. Added additional logging to the case where there is a problem reading data
5784 from a filter that is running in a subprocess using a pipe, in order to
5785 try to track down a specific problem.
5787 4. Testing facility fudge: when running in the test harness and attempting
5788 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
5789 getting "No route to host". Convert this to a timeout.
5791 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
5794 6. Some OS don't have socklen_t but use size_t instead. This affects the
5795 fifth argument of getsockopt() amongst other things. This is now
5796 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
5797 can be set for individual OS. I have set it for SunOS5, OSF1, and
5798 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
5799 some earlier ones do not.
5801 7. Change 4.30/15 was not doing the test caselessly.
5803 8. The standard form for an IPv6 address literal was being rejected by address
5804 parsing in, for example, MAIL and RCPT commands. An example of this kind of
5805 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5806 this, as well as the form without the "IPv6" on the front (but only when
5807 address literals are enabled, of course).
5809 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
5811 10. Exim crashed if a message with an empty sender address specified by -f
5812 encountered a router with an errors_to setting. This could be provoked only
5813 by a command such as
5817 where an empty string was supplied; "<>" did not hit this bug.
5819 11. Installed PCRE release 4.5.
5821 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
5822 remained set. It is now erased.
5824 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
5825 times from message ids (which are base 36 rather than the normal 62).
5827 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
5828 were being counted as actual protocol errors, and logged if the log
5829 selector +smtp_protocol_error was set. One cannot be perfect in this test,
5830 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
5831 and DATA following a set of rejected RCPTs do not count as protocol errors.
5832 In other words, Exim assumes they were pipelined, though this may not
5833 actually be the case. Of course, in all cases the client gets an
5834 appropriate error code.
5836 15. If a lookup fails in an ACL condition, a message about the failure may
5837 be available; it is used if testing the ACL cannot continue, because most
5838 such messages specify what the cause of the deferral is. However, some
5839 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
5840 that caused an old message to be retained and used if a later statement
5841 caused a defer, replacing the real cause of the deferral.
5843 16. If an IP address had so many PTR records that the DNS lookup buffer
5844 was not large enough to hold them, Exim could crash while trying to process
5845 the truncated data. It now detects and logs this case.
5847 17. Further to 4.21/58, another change has been made: if (and only if) the
5848 first line of a message (the first header line) ends with CRLF, a bare LF
5849 in a subsequent header line has a space inserted after it, so as not to
5850 terminate the header.
5852 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
5853 unnecessarily, used atoi() instead of strtol(), and didn't check the
5854 termination when getting file sizes from file names by regex.
5856 19. Completely re-implemented the support for maildirsize files, in the light
5857 of a number of problems with the previous contributed implementation
5858 (4.30/29). In particular:
5860 . If the quota is zero, the maildirsize file is maintained, but no quota is
5863 . If the maildir directory does not exist, it is created before any attempt
5864 to write a maildirsize file.
5866 . The quota value in the file is just a cache; if the quota is changed in
5867 the transport, the new value overrides.
5869 . A regular expression is available for excluding directories from the
5872 20. The autoreply transport checks the characters in options that define the
5873 message's headers; it allows continued headers, but it was checking with
5874 isspace() after an embedded newline instead of explicitly looking for a
5877 21. If all the "regular" hosts to which an address was routed had passed their
5878 expiry times, and had not reached their retry times, the address was
5879 bounced, even if fallback hosts were defined. Now Exim should go on to try
5882 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
5883 equivalent code in the SMTP transport. Some hosts send humungous responses
5884 to HELO/EHLO, more than 1024 it seems.
5886 23. Refactoring: code in filter.c used (void *) for "any old type" but this
5887 gives compiler warnings in some environments. I've now done it "properly",
5890 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
5891 (because of problems with the built-in one) was declared to return uschar *
5892 instead of char *, causing compiler failure.
5894 25. Fixed a file descriptor leak when processing alias/forward files.
5896 26. Fixed a minor format string issue in dbfn.c.
5898 27. Typo in exim.c: ("dmbnz" for "dbmnz").
5900 28. If a filter file refered to $h_xxx or $message_headers, and the headers
5901 contained RFC 2047 "words", Exim's memory could, under certain conditions,
5904 29. When a sender address is verified, it is cached, to save repeating the test
5905 when there is more than one recipient in a message. However, when the
5906 verification involves a callout, it is possible for different callout
5907 options to be set for different recipients. It is too complicated to keep
5908 track of this in the cache, so now Exim always runs a verification when a
5909 callout is required, relying on the callout cache for the optimization.
5910 The overhead is duplication of the address routing, but this should not be
5913 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
5914 to be verified with callout=postmaster, and the main callout worked but the
5915 postmaster check failed, the verification correctly failed. However, if a
5916 subsequent RCPT command asked for sender verification *without* the
5917 postmaster check, incorrect caching caused this verification also to fail,
5920 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
5921 it was not caching the DNS options (qualify_single, search_parents) that
5922 were used when the lookup failed. A subsequent lookup with different
5923 options therefore always gave the same answer, though there were cases
5924 where it should not have. (Example: a "domains = !$mx_any" option on a
5925 dnslookup router: the "domains" option is always processed without any
5926 widening, but the router might have qualify_single set.) Now Exim uses the
5927 cached value only when the same options are set.
5929 32. Added John Jetmore's "exipick" utility to the distribution.
5931 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
5932 than a timeout (e.g. a certificate is required, and is not provided), an
5933 Exim server now closes the connection immediately. Previously it waited for
5934 the client to close - but if the client is SSL, it seems that they each
5935 wait for each other, leading to a delay before one of them times out.
5937 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
5938 maintained 0.8.x compatibility because I don't think many are using it, and
5939 it is clearly obsolete.
5941 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
5944 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
5945 client certificate was expired. A simple patch fixes this, though I don't
5946 understand the full logic of why the verify callback is called multiple
5949 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
5950 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
5951 which causes problems with some clients (such as the Certicom SSL Plus
5952 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
5953 disables the coutermeasure allowing Eudora to connect."
5955 38. Exim was not checking that a write() to a log file succeeded. This could
5956 lead to Bad Things if a log got too big, in particular if it hit a file
5957 size limit. Exim now panics and dies if it cannot write to a log file, just
5958 as it does if it cannot open a log file.
5960 39. Modified OS/Makefile-Linux so that it now contains
5962 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
5964 The two -D definitions ensure that Exim is compiled with large file
5965 support, which makes it possible to handle log files that are bigger than
5968 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
5969 instance) a domain was checked against a named list that involved a lookup,
5970 causing $domain_data to be set, then another domain was checked against the
5971 same list, then the first domain was re-checked, the value of $domain_data
5972 after the final check could be wrong. In particular, if the second check
5973 failed, it could be set empty. This bug probably also applied to
5976 41. The strip_trailing_dot option was not being applied to the address given
5977 with the -f command-line option.
5979 42. The code for reading a message's header from the spool was incrementing
5980 $received_count, but never initializing it. This meant that the value was
5981 incorrect (doubled) while delivering a message in the same process in which
5982 it was received. In the most common configuration of Exim, this never
5983 happens - a fresh exec is done - but it can happen when
5984 deliver_drop_privilege is set.
5986 43. When Exim logs an SMTP synchronization error - client data sent too soon -
5987 it now includes up to 150 characters of the unexpected data in the log
5990 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
5991 and building data strings. The size of both of these buffers was 10 000
5992 bytes - far larger than anybody would *ever* want, thought I. Needless to
5993 say, somebody hit the limit. I have increased the maximum line length to
5994 20 000 and the maximum data length of concatenated lines to 100 000. I have
5995 also fixed two bugs, because there was no checking on these buffers. Tsk,
5996 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
5997 buffer is too small.
5999 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
6000 lsearch lookups. Now it does.
6002 46. When parsing a route_list item in a manualroute router, a fixed-length
6003 buffer was used for the list of hosts. I made this 1024 bytes long,
6004 thinking that nobody would ever have a list of hosts that long. Wrong.
6005 Somebody had a whole pile of complicated expansion conditions, and the
6006 string was silently truncated, leading to an expansion error. It turns out
6007 that it is easier to change to an unlimited length (owing to other changes
6008 that have happened since this code was originally written) than to build
6009 structure for giving a limitation error. The length of the item that
6010 expands into the list of hosts is now unlimited.
6012 47. The lsearch lookup could not handle data where the length of text line was
6013 more than 4095 characters. Such lines were truncated, leading to shortened
6014 data being returned. It should now handle lines of any length.
6016 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
6017 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
6020 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
6021 replacing strings such as BIN_DIRECTORY in a source file: the replacement
6022 no longer happens in comment lines. A list of replacements is now placed
6023 at the head of all of the source files, except those whose only change is
6024 to replace PERL_COMMAND in the very first #! line.
6026 50. Replaced the slow insertion sort in queue.c, for sorting the list of
6027 messages on the queue, with a bottom-up merge sort, using code contributed
6028 by Michael Haardt. This should make operations like -bp somewhat faster on
6029 large queues. It won't affect queue runners, except when queue_run_in_order
6032 51. Installed eximstats 1.31 in the distribution.
6034 52. Added support for SRV lookups to the dnslookup router.
6036 53. If an ACL referred to $message_body or $message_body_end, the value was not
6037 reset for any messages that followed in the same SMTP session.
6039 54. The store-handling optimization for building very long strings was not
6040 differentiating between the different store pools. I don't think this
6041 actually made any difference in practice, but I've tidied it.
6043 55. While running the routers to verify a sender address, $sender_address
6044 was still set to the sender address. This is wrong, because when routing to
6045 send a bounce to the sender, it would be empty. Therefore, I have changed
6046 it so that, while verifying a sender address, $sender_address is set to <>.
6047 (There is no change to what happens when verifying a recipient address.)
6049 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
6050 target A or AAAA records (if not already returned) without resetting the
6051 qualify_single or search_parents options of the DNS resolver. These are
6052 inappropriate in this case because the targets of MX and SRV records must
6053 be FQDNs. A broken DNS record could cause trouble if it happened to have a
6054 target that, when qualified, matched something in the local domain. These
6055 two options are now turned off when doing these lookups.
6057 57. It seems that at least some releases of Reiserfs (which does not have the
6058 concept of a fixed number of inodes) returns zero and not -1 for the
6059 number of available inodes. This interacted badly with check_spool_inodes,
6060 which assumed that -1 was the "no such thing" setting. What I have done is
6061 to check that the total number of inodes is greater than zero before doing
6062 the test of how many are available.
6064 58. When a "warn" ACL statement has a log_message modifier, the message is
6065 remembered, and not repeated. This is to avoid a lot of repetition when a
6066 message has many recipients that cause the same warning to be written.
6067 However, Exim was preserving the list of already written lines for an
6068 entire SMTP session, which doesn't seem right. The memory is now reset if a
6069 new message is started.
6071 59. The "rewrite" debugging flag was not showing the result of rewriting in the
6072 debugging output unless log_rewrite was also set.
6074 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
6075 of (int)(handle) when we know that handle contains (void *)(-1).
6077 61. The Exim daemon panic-logs an error return when it closes the incoming
6078 connection. However "connection reset by peer" seems to be common, and
6079 isn't really an error worthy of noting specially, so that particular error
6082 62. When Exim is trying to find all the local interfaces, it used to panic and
6083 die if the ioctl to get the interface flags failed. However, it seems that
6084 on at least one OS (Solaris 9) it is possible to have an interface that is
6085 included in the list of interfaces, but for which you get a failure error
6086 for this call. This happens when the interface is not "plumbed" into a
6087 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
6088 failure of the "get flags" call assumes that the interface is down.
6090 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
6091 makes life easier for people who are doing arithmetic on fields extracted
6092 from dates, where you often get leading zeros that should not be
6093 interpreted as octal.
6095 64. Added qualify_domain to the redirect router, to override the global
6098 65. If a pathologically long header line contained very many addresses (the
6099 report of this problem mentioned 10 000) and each of them was rewritten,
6100 Exim could use up a very large amount of memory. (It kept on making new
6101 copies of the header line as it rewrote, and never released the old ones.)
6102 At the expense of a bit more processing, the header rewriting function has
6103 been changed so that it no longer eats memory in this way.
6105 66. The generation of the Received: header has been moved from the time that a
6106 message starts to be received, to the time that it finishes. The timestamp
6107 in the Received: header should now be very close to that of the <= log
6108 line. There are two side-effects of this change:
6110 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
6111 logged header lines no longer include the local Received: line, because
6112 it has not yet been created. The same applies to a copy of the message
6113 that is returned to a non-SMTP sender when a message is rejected.
6115 (b) When a filter file is tested using -bf, no additional Received: header
6116 is added to the test message. After some thought, I decided that this
6119 This change does not affect the value of $received_for. It is still set
6120 after address rewriting, but before local_scan() is called.
6122 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
6124 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
6125 gave an unhelpful panic error message, and a defer error. I have managed to
6126 change this behaviour so that it now rejects any supplied certificate,
6127 which seems right, as the list of acceptable certificates is empty.
6129 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
6130 gave an unhelpful defer error. I have not managed to make this reject any
6131 supplied certificates, but the error message it gives is "no certificate
6132 supplied", which is not helpful.
6134 70. exigrep's output now also includes lines that are not associated with any
6135 message, but which match the given pattern. Implemented by a patch from
6136 Martin Sluka, which also tidied up the Perl a bit.
6138 71. Recipient callout verification, like sender verification, was using <> in
6139 the MAIL FROM command. This isn't really the right thing, since the actual
6140 sender may affect whether the remote host accepts the recipient or not. I
6141 have changed it to use the actual sender in the callout; this means that
6142 the cache record is now keyed on a recipient/sender pair, not just the
6143 recipient address. There doesn't seem to be a real danger of callout loops,
6144 since a callout by the remote host to check the sender would use <>.
6145 [SEE ABOVE: changed after hitting problems.]
6147 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
6148 temporary errors. However, in the case of such a code being given after
6149 the end of a data transmission (i.e. after ".") Exim was failing to write
6150 a retry record for the message. (Yes, there was some broken host that was
6151 actually sending 8xx at this point.)
6153 73. An unknown lookup type in a host list could cause Exim to panic-die when
6154 the list was checked. (An example that provoked this was putting <; in the
6155 middle of a list instead of at the start.) If this happened during a DATA
6156 ACL check, a -D file could be left lying around. This kind of configuration
6157 error no longer causes Exim to die; instead it causes a defer error. The
6158 incident is still logged to the main and panic logs.
6160 74. Buglet left over from Exim 3 conversion. The message "too many messages
6161 in one connection" was written to the rejectlog but not the mainlog, except
6162 when address rewriting (yes!) was being logged.
6164 75. Added write_rejectlog option.
6166 76. When a system filter was run not as root (that is, when system_filter_user
6167 was set), the values of the $n variables were not being returned to the
6168 main process; thus, they were not subsequently available in the $sn
6171 77. Added +return_path_on_delivery log selector.
6173 78. A connection timeout was being treated differently from recipients deferred
6174 when testing hosts_max_try with a message that was older than the host's
6175 retry timeout. (The host should not be counted, thus allowing all hosts to
6176 be tried at least once before bouncing.) This may have been the cause of an
6177 occasionally reported bug whereby a message would remain on the queue
6178 longer than the retry timeout, but would be bounced if a delivery was
6179 forced. I say "may" because I never totally pinned down the problem;
6180 setting up timeout/retry tests is difficult. See also the next item.
6182 79. The ultimate address timeout was not being applied to errors that involved
6183 a combination of host plus message (for example, a timeout on a MAIL
6184 command). When an address resolved to a number of possible hosts, and they
6185 were not all tried for each delivery (e.g. because of hosts_max_try), a
6186 message could remain on the queue longer than the retry timeout.
6188 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
6191 81. Fixed an obscure SMTP outgoing bug which required at least the following
6192 conditions: (a) there was another message waiting for the same server;
6193 (b) the server returned 5xx to all RCPT commands in the first message so
6194 that the message was not completed; (c) the server dropped the connection
6195 or gave a negative response to the RSET that Exim sends to abort the
6196 transaction. The observed case was a dropped connection after DATA that had
6197 been sent in pipelining mode. That is, the server had advertised PIPELINING
6198 but was not implementing it correctly. The effect of the bug was incorrect
6199 behaviour, such as trying another host, and this could lead to a crash.
6205 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
6206 and daemon.c were passed as pointers to ints; they should have been
6207 pointers to socklen_t variables (which are typically unsigned ints).
6209 2. Some signed/unsigned type warnings in the os.c file for Linux have been
6212 3. Fixed a really odd bug that affected only the testing scheme; patching a
6213 certain fixed string in the binary changed the value of another string that
6214 happened to be identical to the end of the original first string.
6216 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
6217 name", it returns that address as the IP address. On some operating
6218 systems (e.g. Solaris), it also passes back the IP address string as the
6219 "host name". However, on others (e.g. Linux), it passes back an empty
6220 string. Exim wasn't checking for this, and was changing the host name to an
6221 empty string, assuming it had been canonicalized.
6223 5. Although rare, it is permitted to have more than one PTR record for a given
6224 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
6225 all the names associated with an address, because they do in Solaris.
6226 However, it seems that they do not in Linux for data that comes from the
6227 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
6228 I found this out when I moved to a new Linux workstation and tried to run
6229 the Exim test suite.
6231 To get round this problem I have changed the code so that it now does its
6232 own call to the DNS to look up PTR records when searching for a host name.
6233 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
6234 addresses that are only in /etc/hosts are still found.
6236 This behaviour is, however, controlled by an option called host_lookup_
6237 order, which defaults to "bydns:byaddr". If people want to use the other
6238 order, or indeed, just use one or the other means of lookup, they can
6239 specify it in this variable.
6241 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
6242 some operating systems, this comes back from gethostbyaddr() as an empty
6243 string, and this is what Exim used to test for. However, it seems that in
6244 other systems, "." is yielded. Exim now tests for this case too.
6246 7. The values of check_spool_space and check_log_space are now held internally
6247 as a number of kilobytes instead of an absolute number of bytes. If a
6248 numbers is specified without 'K' or 'M', it is rounded up to the nearest
6249 kilobyte. This means that much larger values can be stored.
6251 8. Exim monitor: an attempt to get the action menu when not actually pointing
6252 at a message produces an empty menu entitled "No message selected". This
6253 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
6254 no entries in it ("Shell widget menu has zero width and/or height"). So I
6255 have added a single, blank menu entry in this case.
6257 9. Added ${quote_local_part.
6259 10. MIME decoding is now applied to the contents of Subject: header lines when
6262 11. Now that a reference to $sender_host_address automatically causes a reverse
6263 lookup to occur if necessary (4.13/18), there is no need to arrange for a
6264 host lookup before query-style lookups in lists that might use this
6265 variable. This has therefore been abolished, and the "net-" prefix is no
6266 longer necessary for query-style lookups.
6268 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
6269 have been a typo for LFLAGS, so it has been changed.
6271 13. The install script calls Exim with "-C /dev/null" in order to find the
6272 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
6273 to be output. However, since Exim outputs its version number before the
6274 error, it didn't break the script. It just looked ugly. I fixed this by
6275 always allowing "-C /dev/null" if the caller is root.
6277 14. Ignore overlarge ACL variable number when reading spool file - insurance
6278 against a later release with more variables having written the file.
6280 15. The standard form for an IPv6 address literal was being rejected by EHLO.
6281 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
6282 this, as well as the form without the "IPv6" on the front.
6284 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
6285 OS/Makefile-Darwin file.
6287 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
6288 only to LDAP libraries that do not have LDAP_OPT_DEREF.
6290 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
6291 variable. However, some OS use ints for inodes. I've added cast to long int
6292 to get rid of the compiler warning.
6294 19. I had forgotten to lock out "/../" in configuration file names when
6295 ALT_CONFIG_PREFIX was set.
6297 20. Routers used for verification do not need to specify transports. However,
6298 if such a router generated a host list, and callout was configured, Exim
6299 crashed, because it could not find a port number from the (non-existent)
6300 transport. It now assumes port 25 in this circumstance.
6302 21. Added the -t option to exigrep.
6304 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
6305 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
6306 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
6307 them is set, LOOKUP_LSEARCH is forced.
6309 23. "exim -bV" now outputs a list of lookups that are included in the binary.
6311 24. Added sender and host information to the "rejected by local_scan()" log
6312 line; previously there was no indication of these.
6314 25. Added .include_if_exists.
6316 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
6317 Linux. It turns out that not all file systems support this. Apparently some
6318 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
6319 do it.) To cope with this, the error EINVAL, which means that sync-ing is
6320 not supported on the file descriptor, is now ignored when Exim is trying to
6321 sync a directory. This applies only to Linux.
6323 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
6325 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
6326 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
6329 29. Added support for maildirsize files from supplied patch (modified a bit).
6331 30. The use of :fail: followed by an empty string could lead Exim to respond to
6332 sender verification failures with (e.g.):
6334 550 Verification failed for <xxx>
6335 550 Sender verify failed
6337 where the first response line was missing the '-' that indicates it is not
6338 the final line of the response.
6340 31. The loop for finding the name of the user that called Exim had a hardwired
6341 limit of 10; it now uses the value of finduser_retries, which is used for
6342 all other user lookups.
6344 32. Added $received_count variable, available in data and not_smtp ACLs, and at
6347 33. Exim was neglecting to zero errno before one call of strtol() when
6348 expanding a string and expecting an integer value. On some systems this
6349 resulted in spurious "integer overflow" errors. Also, it was casting the
6350 result into an int without checking.
6352 34. Testing for a connection timeout using "timeout_connect" in the retry rules
6353 did not work. The code looks as if it has *never* worked, though it appears
6354 to have been documented since at least release 1.62. I have made it work.
6356 35. The "timeout_DNS" error in retry rules, also documented since at least
6357 1.62, also never worked. As it isn't clear exactly what this means, and
6358 clearly it isn't a major issue, I have abolished the feature by treating it
6359 as "timeout", and writing a warning to the main and panic logs.
6361 36. The display of retry rules for -brt wasn't always showing the error code
6364 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
6365 timeout_connect_A, timeout_connect_MX.
6367 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
6368 to the empty sender.
6370 39. The daemon was not analysing the content of -oX till after it had closed
6371 stderr and disconnected from the controlling terminal. This meant that any
6372 syntax errors were only noted on the panic log, and the return code from
6373 the command was 0. By re-arranging the code a little, I've made the
6374 decoding happen first, so such errors now appear on stderr, and the return
6375 code is 1. However, the actual setting up of the sockets still happens in
6376 the disconnected process, so errors there are still only recorded on the
6379 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
6380 connections (as happens on some IP stacks) was logged at start up time as
6381 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
6382 it from "IPv6 and IPv4", which means that two separate sockets are being
6385 41. The debug output for gethostbyname2() or getipnodebyname() failures now
6386 says whether AF_INET or AF_INET6 was passed as an argument.
6388 42. Exiwhat output was messed up when time zones were included in log
6391 43. Exiwhat now gives more information about the daemon's listening ports,
6392 and whether -tls-on-connect was used.
6394 44. The "port" option of the smtp transport is now expanded.
6396 45. A "message" modifier in a "warn" statement in a non-message ACL was being
6397 silently ignored. Now an error message is written to the main and panic
6400 46. There's a new ACL modifier called "logwrite" which writes to a log file
6401 as soon as it is encountered.
6403 47. Added $local_user_uid and $local_user_gid at routing time.
6405 48. Exim crashed when trying to verify a sender address that was being
6408 49. Exim was recognizing only a space character after ".include". It now also
6409 recognizes a tab character.
6411 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
6412 extracting the relevant information from the specification. The man page no
6413 longer contains scrambled data for the -d option, and I've added a section
6414 at the front about calling Exim under different names.
6416 51. Added "extra_headers" argument to the "mail" command in filter files.
6418 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
6421 53. Installed eximstats 1.29.
6423 54. Added transport_filter_timeout as a generic transport option.
6425 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
6426 Cc: header lines. This was required by RFC 822, but it not required by RFC
6429 56. Exim used to add From:, Date:, and Message-Id: header lines to any
6430 incoming messages that did not have them. Now it does so only if the
6431 message originates locally, that is, if there is no associated remote host
6432 address. When Resent- header lines are present, this applies to the Resent-
6433 lines rather than the non-Resent- lines.
6435 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
6436 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
6438 58. Messages for configuration errors now include the name of the main
6439 configuration file - useful now that there may be more than one file in a
6440 list (.included file names were always shown).
6442 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
6443 for those rare installations that do not start the daemon as root or run it
6444 setuid root. I've cut out the call to initgroups() if the daemon is not
6447 60. The Exim user and group can now be bound into the binary as text strings
6448 that are looked up at the start of Exim's processing.
6450 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
6452 62. Added $mailstore_basename variable.
6454 63. Installed patch to sieve.c from Michael Haardt.
6456 64. When Exim failed to open the panic log after failing to open the main log,
6457 the original message it was trying to log was written to stderr and debug
6458 output, but if they were not available (the usual case in production), it
6459 was lost. Now it is written to syslog before the two lines that record the
6460 failures to open the logs.
6462 65. Users' Exim filters run in subprocesses under the user's uid. It is
6463 possible for a "deliver" command or an alias in a "personal" command to
6464 provoke an address rewrite. If logging of address rewriting is configured,
6465 this fails because the process is not running as root or exim. There may be
6466 a better way of dealing with this, but for the moment (because 4.30 needs
6467 to be released), I have disabled address rewrite logging when running a
6468 filter in a non-root, non-exim process.
6474 1. The buildconfig auxiliary program wasn't quoting the value set for
6475 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
6476 not defined. This bug was masked in 4.22 by the effect that was fixed in
6479 2. Some messages that were rejected after a message id was allocated were
6480 shown as "incomplete" by exigrep. It no longer does this for messages that
6481 are rejected by local_scan() or the DATA or non-SMTP ACLs.
6483 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
6484 have allow_domain_literals set, the ID did not get logged in the <= line.
6485 Domain literals are now always recognized in Message-ID: header lines.
6487 4. The first argument for a ${extract expansion item is the key name or field
6488 number. Leading and trailing spaces in this item were not being ignored,
6489 causing some misleading effects.
6491 5. When deliver_drop_privilege was set, single queue runner processes started
6492 manually (i.e. by the command "exim -q") or by the daemon (which uses the
6493 same command in the process it spins off) were not dropping privilege.
6495 6. When the daemon running as "exim" started a queue runner, it always
6496 re-executed Exim in the spun-off process. This is a waste of effort when
6497 deliver_drop_privilege is set. The new process now just calls the
6498 queue-runner function directly.
6504 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
6507 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
6508 ignored. Though the use of -oP was forcing the writing of a pid file, it
6509 was always written to the default place.
6511 3. If the message "no IP address found for host xxxx" is generated during
6512 incoming verification, it is now followed by identification of the incoming
6513 connection (so you can more easily find what provoked it).
6515 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
6517 5. Added some features to "harden" Exim a bit more against certain attacks:
6519 (a) There is now a build-time option called FIXED_NEVER_USERS that can
6520 be put in Local/Makefile. This is like the never_users runtime option,
6521 but it cannot be overridden. The default setting is "root".
6523 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
6524 prefix string with which any file named in a -C command line option
6527 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
6528 is retained for -C and -D only if the caller of Exim is root. Without
6529 it, the exim user may also use -C and -D and retain privilege.
6531 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
6532 command line option is disabled.
6534 6. Macro names set by the -D option must start with an upper case letter, just
6535 like macro names defined in the configuration file.
6537 7. Added "dereference=" facility to LDAP.
6539 8. Two instances of the typo "uknown" in the source files are fixed.
6541 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
6542 the Configure-Makefile script screwed up while processing it.
6544 10. Incorporated PCRE 4.4.
6546 11. The SMTP synchronization check was not operating right at the start of an
6547 SMTP session. For example, it could not catch a HELO sent before the client
6548 waited for the greeting. There is now a check for outstanding input at the
6549 point when the greeting is written. Because of the duplex, asynchronous
6550 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
6551 way, but not yet received, when the check is performed.
6553 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
6554 on TCP/IP sockets, because this apparently causes some broken clients to
6557 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
6558 unchanged) from the Cygwin maintainer.
6560 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
6561 was supported instead of testing for mbx. Effectively a typo.
6563 15. The spa authenticator server code was not checking that the input it
6564 received was valid base64.
6566 16. The debug output line for the "set" modifier in ACLs was not showing the
6567 name of the variable that was being set.
6569 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
6571 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
6572 lookup on its own. Something else was needed to trigger the lookup. For
6573 example, a match in host_lookup or the need for a host name in a host list.
6574 Now, if $sender_host_name is referenced and the host name has not yet been
6575 looked up, a lookup is performed. If the lookup fails, the variable remains
6576 empty, and $host_lookup_failed is set to "1".
6578 19. Added "eqi" as a case-independent comparison operator.
6580 20. The saslauthd authentication condition could segfault if neither service
6581 nor realm was specified.
6583 21. If an overflowing value such as "2048M" was set for message_size_limit, the
6584 error message that was logged was misleading, and incoming SMTP
6585 connections were dropped. The message is now more accurate, and temporary
6586 errors are given to SMTP connections.
6588 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
6589 (except RSET) with a 421 error, until QUIT is received. However, it was
6590 failing to send a response to QUIT.
6592 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
6593 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
6594 ACL is now run after the helo_try_verify_hosts code.
6596 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
6597 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
6598 case-independent, so other case variants are also recognized). Apparently
6599 some systems use these upper case variants.
6601 25. If more than two messages were waiting for the same host, and a transport
6602 filter was specified for the transport, Exim sent two messages over the
6603 same TCP/IP connection, and then failed with "socket operation on non-
6604 socket" when it tried to send the third.
6606 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
6608 27. The extern definition of crypt16() in expand.c was not being excluded when
6609 the OS had its own crypt16() function.
6611 28. Added bounce_return_body as a new option, and bounce_return_size_limit
6612 as a preferred synonym for return_size_limit, both as an option and as an
6615 29. Added LIBS=-liconv to OS/Makefile-OSF1.
6617 30. Changed the default configuration ACL to relax the local part checking rule
6618 for addresses that are not in any local domains. For these addresses,
6619 slashes and pipe symbols are allowed within local parts, but the sequence
6620 /../ is explicitly forbidden.
6622 31. SPA server authentication was not clearing the challenge buffer before
6625 32. log_message in a "warn" ACL statement was writing to the reject log as
6626 well as to the main log, which contradicts the documentation and doesn't
6627 seem right (because no rejection is happening). So I have stopped it.
6629 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
6630 However, I am unable to do any testing of this.
6632 34. Fixed an infelicity in the appendfile transport. When checking directories
6633 for a mailbox, to see if any needed to be created, it was accidentally
6634 using path names with one or more superfluous leading slashes; tracing
6635 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
6637 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
6638 contents are added to the log line that is written for every discarded
6639 recipient. (Previously a log_message setting was ignored.)
6641 36. The ${quote: operator now quotes the string if it is empty.
6643 37. The install script runs exim in order to find its version number. If for
6644 some reason other than non-existence or emptiness, which it checks, it
6645 could not run './exim', it was installing it with an empty version number,
6646 i.e. as "exim-". This error state is now caught, and the installation is
6649 38. An argument was missing from the function that creates an error message
6650 when Exim fails to connect to the socket for saslauthd authentication.
6651 This could cause Exim to crash, or give a corrupted message.
6653 39. Added isip, isip4, and isip6 to ${if conditions.
6655 40. The ACL variables $acl_xx are now saved with the message, and can be
6656 accessed later in routers, transports, and filters.
6658 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
6659 strings in the file are not string-expanded.
6661 42. If a MAIL command specified a SIZE value that was too large to fit into an
6662 int variable, the check against message_size_limit failed. Such values are
6663 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
6664 day this will have to be increased, but I don't think I want to be around
6665 when emails are that large.
6672 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
6673 iconv() is not standard in FreeBSD.
6675 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
6676 IPv6 enabled. The observed symptom was a segmentation fault on return from
6677 the function os_common_find_running_interfaces() in src/os.c.
6679 3. In the check_special_case() function in daemon.c I had used "errno" as an
6680 argument name, which causes warnings on some systems. This was basically a
6681 typo, since it was named "eno" in the comments!
6683 4. The code that waits for the clock to tick (at a resolution of some fraction
6684 of a second) so as to ensure message-id uniqueness was always waiting for
6685 at least one whole tick, when it could have waited for less. [This is
6686 almost certainly not relevant at current processor speeds, where it is
6687 unlikely to ever wait at all. But we try to future-proof.]
6689 5. The function that sleeps for a time interval that includes fractions of a
6690 second contained a race. It did not block SIGALRM between setting the
6691 timer, and suspending (a couple of lines later). If the interval was short
6692 and the sigsuspend() was delayed until after it had expired, the suspension
6693 never ended. On busy systems this could lead to processes getting stuck for
6696 6. Some uncommon configurations may cause a lookup to happen in a queue runner
6697 process, before it forks any delivery processes. The open lookup caching
6698 mechanism meant that the open file or database connection was passed into
6699 the delivery process. The problem was that delivery processes always tidy
6700 up cached lookup data. This could cause a problem for the next delivery
6701 process started by the queue runner, because the external queue runner
6702 process does not know about the closure. So the next delivery process
6703 still has data in the lookup cache. In the case of a file lookup, there was
6704 no problem because closing a file descriptor in a subprocess doesn't affect
6705 the parent. However, if the lookup was caching a connection to a database,
6706 the connection was closed, and the second delivery process was likely to
6707 see errors such as "PGSQL: query failed: server closed the connection
6708 unexpectedly". The problem has been fixed by closing all cached lookups
6709 in a queue runner before running a delivery process.
6711 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
6712 seem to have the "const" qualifier which it has on other OS. I've
6715 8. Change 4.21/2 was too strict. It is only if there are two authenticators
6716 *of the same type* (client or server) with the same public name that an
6717 error should be diagnosed.
6719 9. When Exim looked up a host name for an IP address, but failed to find the
6720 original IP address when looking up the host name (a safety check), it
6721 output the message "<ip address> does not match any IP for NULL", which was
6722 confusing, to say the least. The bug was that the host name should have
6723 appeared instead of "NULL".
6725 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
6726 user that is built into the binary, and the -C or -D options is used, root
6727 privilege is dropped before the configuration file is read. In addition,
6728 logging is switched to stderr instead of the normal log files. If the
6729 configuration then re-defines the Exim user, the unprivileged environment
6730 is probably not what is expected, so Exim logs a panic warning message (but
6733 However, if deliver_drop_privilege is set, the unprivileged state may well
6734 be exactly what is intended, so the warning has been cut out in that case,
6735 and Exim is allowed to try to write to its normal log files.
6741 1. smtp_return_error_details was not giving details for temporary sender
6742 or receiver verification errors.
6744 2. Diagnose a configuration error if two authenticators have the same public
6747 3. Exim used not to create the message log file for a message until the first
6748 delivery attempt. This could be confusing when incoming messages were held
6749 for policy or load reasons. The message log file is now created at the time
6750 the message is received, and an initial "Received" line is written to it.
6752 4. The automatically generated man page for command line options had a minor
6753 bug that caused no ill effects; however, a more serious problem was that
6754 the procedure for building the man page automatically didn't always
6755 operate. Consequently, release 4.20 contains an out-of-date version. This
6756 shouldn't happen again.
6758 5. When building Exim with embedded Perl support, the script that builds the
6759 Makefile was calling 'perl' to find its compile-time parameters, ignoring
6760 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
6762 6. The freeze_tell option was not being used for messages that were frozen on
6763 arrival, either by an ACL or by local_scan().
6765 7. Added the smtp_incomplete_transaction log selector.
6767 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
6768 was accepting AUTH without a new EHLO.
6770 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
6771 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
6774 10. Logging of TCP/IP connections (when configured) now happens in the main
6775 daemon process instead of the child process, so that the TCP/IP connection
6776 count is more accurate (but it can never be perfect).
6778 11. The use of "drop" in a nested ACL was not being handled correctly in the
6779 outer ACL. Now, if condition failure induced by the nested "drop" causes
6780 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
6781 or "require"), the connection is dropped.
6783 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
6784 that yield "discard" can now be used with an "accept" or a "discard" verb,
6785 but an error is generated for any others (because I can't see a useful way
6786 to define what should happen).
6788 13. When an ACL is read dynamically from a file (or anywhere else), the lines
6789 are now processed in the same way as lines in the Exim configuration file.
6790 In particular, continuation lines are supported.
6792 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
6794 15. Added -ti meaning -t -i.
6796 16. Check for letters, digits, hyphens, and dots in the names of dnslist
6797 domains, and warn by logging if others are found.
6799 17. At least on BSD, alignment is not guaranteed for the array of ifreq's
6800 returned from GIFCONF when Exim is trying to find the list of interfaces on
6801 a host. The code in os.c has been modified to copy each ifreq to an aligned
6802 structure in all cases.
6804 Also, in some cases, the returned ifreq's were being copied to a 'struct
6805 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
6806 means the last couple of bytes of an IPv6 address could be chopped if the
6807 ifreq contained only a normal sockaddr (14 bytes storage).
6809 18. Named domain lists were not supported in the hosts_treat_as_local option.
6810 An entry such as +xxxx was not recognized, and was treated as a literal
6813 19. Ensure that header lines added by a DATA ACL are included in the reject log
6814 if the ACL subsequently rejects the message.
6816 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
6817 MD5 (which is deprecated).
6819 21. When testing a filter file using -bf, Exim was writing a message when it
6820 took the sender from a "From " line in the message, but it was not doing so
6821 when it took $return_path from a Return-Path: header line. It now does.
6823 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
6824 with a valid header line field name (a series of printing characters
6825 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
6827 23. Changed "disc" in the source to "disk" to conform to the documentation and
6828 the book and for uniformity.
6830 24. Ignore Sendmail's -Ooption=value command line item.
6832 25. When execve() failed while trying to run a command in a pipe transport,
6833 Exim was returning EX_UNAVAILABLE (69) from the subprocess. However, this
6834 could be confused with a return value of 69 from the command itself. This
6835 has been changed to 127, the value the shell returns if it is asked to run
6836 a non-existent command. The wording for the related log line suggests a
6837 non-existent command as the problem.
6839 26. If received_header_text expands to an empty string, do not add a Received:
6840 header line to the message. (Well, it adds a token one on the spool, but
6841 marks it "old" so that it doesn't get used or transmitted.)
6843 27. Installed eximstats 1.28 (addition of -nt option).
6845 28. There was no check for failure on the call to getsockname() in the daemon
6846 code. This can fail if there is a shortage of resources on the system, with
6847 ENOMEM, for example. A temporary error is now given on failure.
6849 29. Contrary to the C standard, it seems that in some environments, the
6850 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
6851 program. Exim now does this explicitly; it affects the formatting of
6852 timestamps using strftime().
6854 30. If exiqsumm was given junk data, it threw up some uninitialized variable
6855 complaints. I've now initialized all the variables, to avoid this.
6857 32. Header lines added by a system filter were not being "seen" during
6858 transport-time rewrites.
6860 33. The info_callback() function passed to OpenSSL is set up with type void
6861 (*)(SSL *, int, int), as described somewhere. However, when calling the
6862 function (actually a macro) that sets it up, the type void(*)() is
6863 expected. I've put in a cast to prevent warnings from picky compilers.
6865 34. If a DNS black list lookup found a CNAME record, but there were no A
6866 records associated with the domain it pointed at, Exim crashed.
6868 35. If a DNS black list lookup returned more than one A record, Exim ignored
6869 all but the first. It now scans all returned addresses if a particular IP
6870 value is being sought. In this situation, the contents of the
6871 $dnslist_value variable are a list of all the addresses, separated by a
6874 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
6875 to accept a host name and all its aliases if the forward lookup for any of
6876 them yielded the IP address of the incoming connection. Now it accepts only
6877 those names whose forward lookup yields the correct IP address. Any other
6878 names are discarded. This closes a loophole whereby a rogue DNS
6879 administrator could create reverse DNS records to break through a
6880 wildcarded host restriction in an ACL.
6882 37. If a user filter or a system filter that ran in a subprocess used any of
6883 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
6884 the wrong values were passed to the pipe command ($thisaddress had the
6885 value of $0, $0 had the value of $1, etc). This bug was introduced by
6886 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
6888 38. Improved the line breaking for long SMTP error messages from ACLs.
6889 Previously, if there was no break point between 40 and 75 characters, Exim
6890 left the rest of the message alone. Two changes have been made: (a) I've
6891 reduced the minimum length to 35 characters; (b) if it can't find a break
6892 point between 35 and 75 characters, it looks ahead and uses the first one
6893 that it finds. This may give the occasional overlong line, but at least the
6894 remaining text gets split now.
6896 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
6897 file descriptors might be low, and that setting 1000 would always raise it.
6898 It turns out that in some environments, the limit is already over 1000 and
6899 that lowering it causes trouble. So now Exim takes care not to decrease it.
6901 40. When delivering a message, the value of $return_path is set to $sender_
6902 address at the start of routing (routers may change the value). By an
6903 oversight, this default was not being set up when an address was tested by
6904 -bt or -bv, which affected the outcome if any router or filter referred to
6907 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
6908 log only when "message" or "log_message" are set. However, if one of the
6909 conditions was an address verification, or a call to a nested ACL, the
6910 messages generated by the underlying test were being passed through. This
6911 no longer happens. The underlying message is available in $acl_verify_
6912 message for both "message" and "log_message" expansions, so it can be
6913 passed through if needed.
6915 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
6916 new expansion $bh_ to give the encoded byte string without charset
6917 translation. Translation happens only if iconv() is available; HAVE_ICONV
6918 indicates this at build time. HEADERS_CHARSET gives the charset to
6919 translate to; headers_charset can change it in the configuration, and
6920 "headers charset" can change it in an individual filter file.
6922 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
6923 that creates RFC 2047 encoded "words" labels them as that charset instead
6924 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
6925 expansion operator; (ii) when Exim creates a From: line for a local
6926 message; (iii) when a header line is rewritten to include a "phrase" part.
6928 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
6929 buggy, causing it to skip the first lines of messages whose message ID
6930 ended in 'D'. This would not have bitten before Exim release 4.14, because
6931 message IDs were unlikely to end in 'D' before then. The effect was to have
6932 incorrect size information for certain domains.
6934 45. #include "config.h" was missing at the start of the crypt16.c module. This
6935 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
6938 46. If there was a timeout during a "random" callout check, Exim treated it as
6939 a failure of the random address, and carried on sending RSET and the real
6940 address. If the delay was just some slowness somewhere, the response to the
6941 original RCPT would be taken as a response to RSET and so on, causing
6942 mayhem of various kinds.
6944 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
6945 when I implemented it. It didn't allow for the fact that some option values
6946 may legitimately be negative (e.g. size_addition), and it didn't even do
6947 the right test for positive values.
6949 48. Domain names in DNS records are case-independent. Exim always looks them up
6950 in lower case. Some resolvers return domain names in exactly the case they
6951 appear in the zone file, that is, they may contain uppercase letters. Not
6952 all resolvers do this - some return always lower case. Exim was treating a
6953 change of case by a resolver as a change of domain, similar to a widening
6954 of a domain abbreviation. This triggered its re-routing code and so it was
6955 trying to route what was effectively the same domain again. This normally
6956 caused routing to fail (because the router wouldn't handle the domain
6957 twice). Now Exim checks for this case specially, and just changes the
6958 casing of the domain that it ultimately uses when it transmits the message
6961 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
6964 50. If a filter generated a file delivery with a non-absolute name (possible if
6965 no home directory exists for the router), the forbid_file option was not
6968 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
6969 the existing equality matching.
6971 52. Exim was using ints instead of ino_t variables in some places where it was
6972 dealing with inode numbers.
6974 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
6975 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
6976 called TMPDIR, and if it finds it is different, it changes its value.
6978 54. The smtp_printf() function is now made available to local_scan() so
6979 additional output lines can be written before returning. There is also an
6980 smtp_fflush() function to enable the detection of a dropped connection.
6981 The variables smtp_input and smtp_batched_input are exported to
6984 55. Changed the default runtime configuration: the message "Unknown user"
6985 has been removed from the ACL, and instead placed on the localuser router,
6986 using the cannot_route_message feature. This means that any verification
6987 failures that generate their own messages won't get overridden. Similarly,
6988 the "Unrouteable address" message that was in the ACL for unverifiable
6989 relay addresses has also been removed.
6991 56. Added hosts_avoid_esmtp to the smtp transport.
6993 57. The exicyclog script was not checking for the esoteric option
6994 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
6995 will work only if exicyclog is run under the appropriate euid.
6997 58. Following a discussion on the list, the rules by which Exim recognises line
6998 endings on incoming messages have been changed. The -dropcr and drop_cr
6999 options are now no-ops, retained only for backwards compatibility. The
7000 following line terminators are recognized: LF CRLF CR. However, special
7001 processing applies to CR:
7003 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
7004 nor a local message in the state where . is a terminator.
7006 (ii) If a bare CR is encountered in a header line, an extra space is added
7007 after the line terminator so as not to end the header. The reasoning
7008 behind this is that bare CRs in header lines are most likely either
7009 to be mistakes, or people trying to play silly games.
7011 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
7012 was being incorrectly given as 18 bytes larger than it should have been.
7013 This is a VOB (very old bug).
7015 60. This may never have affected anything current, but just in case it has:
7016 When the local host is found other than at the start of a list of hosts,
7017 the local host, those with the same MX, and any that follow, are discarded.
7018 When the list in question was part of a longer list of hosts, the following
7019 hosts (not currently being processed) were also being discarded. This no
7020 longer happens. I'm not sure if this situation could ever has previously
7023 61. Added the "/MX" feature to lists of hosts in the manualroute and query
7026 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
7027 header. This is something that is recommended in a new Internet Draft, and
7028 is something that is documented as being done by Sendmail. There are two
7029 possible values. For messages generated by the autoreply transport, Exim
7032 Auto-Submitted: auto-replied
7034 whereas for all other generated messages (e.g. bounces) it adds
7036 Auto-Submitted: auto-generated
7038 63. The "personal" condition in filters now includes a test for the
7039 Auto-Submitted: header. If it contains the string "auto-" the message it
7040 not considered personal.
7042 64. Added rcpt_include_affixes as a generic transport option.
7044 65. Added queue_only_override (default true).
7046 66. Added the syslog_duplication option.
7048 67. If what should have been the first header line of a message consisted of
7049 a space followed by a colon, Exim was mis-interpreting it as a header line.
7050 It isn't of course - it is syntactically invalid and should therefore be
7051 treated as the start of the message body. The misbehaviour could have
7052 caused a number of strange effects, including loss of data in subsequent
7053 header lines, and spool format errors.
7055 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
7056 client host had authenticated. This control can now be exercised by an ACL
7057 for more flexibility.
7059 69. By default, callouts do not happen when testing with -bh. There is now a
7060 variant, -bhc, which does actually run the callout code, including
7061 consulting and updating the callout cache.
7063 70. Added support for saslauthd authentication, courtesy of Alexander
7066 71. If statvfs() failed on the spool or log directories while checking their
7067 size for availability, Exim confusingly gave the error "space shortage".
7068 Furthermore, in debugging mode it crashed with a floating point exception.
7069 These checks are done if check_{spool,log}_{space,inodes} are set, and when
7070 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
7071 serious problem, Exim now writes to the main and panic logs when this
7072 happens, with details of the failure. It then refuses to accept the
7073 incoming message, giving the message "spool directory problem" or "log
7074 directory problem" with a 421 code for SMTP messages.
7076 72. When Exim is about to re-exec itself, it ensures that the file descriptors
7077 0, 1, and 2 exist, because some OS complain for execs without them (see
7078 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
7079 descriptors. However, the code omitted to check that the open succeeded,
7080 causing mysterious errors if for some reason the permissions on /dev/null
7081 got screwed. Now Exim writes a message to the main and panic logs, and
7082 bombs out if it can't open /dev/null.
7084 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
7085 interact so that it is all more flexible. It is supposed to remain
7086 backwards compatible. Also added extra_local_interfaces.
7088 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
7089 to bomb out with an assertion failure - to the client this appears as a
7090 connection drop. This problem occurs in the part of the code that was taken
7091 from the Samba project. Fortunately, the assertion is in a very simple
7092 function, so I have fixed this by reproducing the function inline in the
7093 one place where it is called, and arranging for authentication to fail
7094 instead of killing the process with assert().
7096 75. The SPA client code was not working when the server requested OEM rather
7097 than Unicode encoding.
7099 76. Added code to make require_files with a specific uid setting more usable in
7100 the case where statting the file as root fails - usually a non-root-mounted
7101 NFS file system. When this happens and the failure is EACCES, Exim now
7102 forks a subprocess and does the per-uid checking as the relevant uid.
7104 77. Added process_log_path.
7106 78. If log_file_path was not explicitly set, a setting of check_log_space or
7107 check_log_inodes was ignored.
7109 79. If a space check for the spool or log partitions fails, the incident is now
7110 logged. Of course, in the latter case the data may get lost...
7112 80. Added the %p formatting code to string_format() so that it can be used to
7113 print addresses in debug_print(). Adjusted all the address printing in the
7114 debugging in store.c to use %p rather than %d.
7116 81. There was a concern that a line of code in smtp_in.c could overflow a
7117 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
7118 initially expressed, the concern was not well-founded, because trailing
7119 spaces are removed early. However, if the trailing spaces were followed by
7120 a NULL, they did not get removed, so the overflow was possible. Two fixes
7123 (a) I re-wrote the offending code in a cleaner fashion.
7124 (b) If an incoming SMTP command contains a NULL character, it is rejected
7127 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
7128 runs initgroups(), so that if the Exim user is in any additional groups,
7129 they will be used during message reception.
7135 The change log for 4.20 and earlier releases has been archived.
projects / exim.git / blob