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 JH/01 Replace the store_release() internal interface with store_newblock(),
19 which internalises the check required to safely use the old one, plus
20 the allocate and data copy operations duplicated in both (!) of the
23 JH/02 Disallow '/' characters in queue names specified for the "queue=" ACL
24 modifier. This matches the restriction on the commandline.
26 JH/03 Fix pgsql lookup for multiple result-tuples with a single column.
27 Previously only the last row was returned.
29 JH/04 Bug 2217: Tighten up the parsing of DKIM signature headers. Previously
30 we assumed that tags in the header were well-formed, and parsed the
31 element content after inspecting only the first char of the tag.
32 Assumptions at that stage could crash the receive process on malformed
35 JH/05 Bug 2215: Fix crash associated with dnsdb lookup done from DKIM ACL.
36 While running the DKIM ACL we operate on the Permanent memory pool so that
37 variables created with "set" persist to the DATA ACL. Also (at any time)
38 DNS lookups that fail create cache records using the Permanent pool. But
39 expansions release any allocations made on the current pool - so a dnsdb
40 lookup expansion done in the DKIM ACL releases the memory used for the
41 DNS negative-cache, and bad things result. Solution is to switch to the
42 Main pool for expansions.
43 While we're in that code, add checks on the DNS cache during store_reset,
44 active in the testsuite.
45 Problem spotted, and debugging aided, by Wolfgang Breyha.
47 JH/06 Fix issue with continued-connections when the DNS shifts unreliably.
48 When none of the hosts presented to a transport match an already-open
49 connection, close it and proceed with the list. Previously we would
50 queue the message. Spotted by Lena with Yahoo, probably involving
53 JH/07 Bug 2214: Fix SMTP responses resulting from non-accept result of MIME ACL.
54 Previously a spurious "250 OK id=" response was appended to the proper
57 JH/08 The "support for" informational output now, which built with Content
58 Scanning support, has a line for the malware scanner interfaces compiled
59 in. Interface can be individually included or not at build time.
61 JH/09 The "aveserver", "kavdaemon" and "mksd" interfaces are now not included
62 by the template makefile "src/EDITME". The "STREAM" support for an older
63 ClamAV interface method is removed.
65 JH/10 Bug 2223: Fix mysql lookup returns for the no-data case (when the number of
66 rows affected is given instead).
68 JH/11 The runtime Berkeley DB library version is now additionally output by
69 "exim -d -bV". Previously only the compile-time version was shown.
71 JH/12 Bug 2230: Fix cutthrough routing for nonfirst messages in an initiating
72 SMTP connection. Previously, when one had more receipients than the
73 first, an abortive onward connection was made. Move to full support for
74 multiple onward connections in sequence, handling cutthrough connection
75 for all multi-message initiating connections.
77 JH/13 Bug 2229: Fix cutthrough routing for nonstandard port numbers defined by
78 routers. Previously, a multi-recipient message would fail to match the
79 onward-connection opened for the first recipient, and cause its closure.
81 JH/14 Bug 2174: A timeout on connect for a callout was also erroneously seen as
82 a timeout on read on a GnuTLS initiating connection, resulting in the
83 initiating connection being dropped. This mattered most when the callout
84 was marked defer_ok. Fix to keep the two timeout-detection methods
87 JH/15 Relax results from ACL control request to enable cutthrough, in
88 unsupported situations, from error to silently (except under debug)
89 ignoring. This covers use with PRDR, frozen messages, queue-only and
92 HS/01 Fix Buffer overflow in base64d() (CVE-2018-6789)
94 JH/16 Fix bug in DKIM verify: a buffer overflow could corrupt the malloc
95 metadata, resulting in a crash in free().
97 PP/01 Fix broken Heimdal GSSAPI authenticator integration.
98 Broken in f2ed27cf5, missing an equals sign for specified-initialisers.
99 Broken also in d185889f4, with init system revamp.
101 JH/17 Bug 2113: Fix conversation closedown with the Avast malware scanner.
102 Previously we abruptly closed the connection after reading a malware-
103 found indication; now we go on to read the "scan ok" response line,
106 JH/18 Bug 2239: Enforce non-usability of control=utf8_downconvert in the mail
107 ACL. Previously, a crash would result.
109 JH/19 Speed up macro lookups during configuration file read, by skipping non-
110 macro text after a replacement (previously it was only once per line) and
111 by skipping builtin macros when searching for an uppercase lead character.
113 JH/20 DANE support moved from Experimental to mainline. The Makefile control
114 for the build is renamed.
116 JH/21 Fix memory leak during multi-message connections using STARTTLS. A buffer
117 was allocated for every new TLS startup, meaning one per message. Fix
118 by only allocating once (OpenSSL) or freeing on TLS-close (GnuTLS).
120 JH/22 Bug 2236: When a DKIM verification result is overridden by ACL, DMARC
121 reported the original. Fix to report (as far as possible) the ACL
122 result replacing the original.
124 JH/23 Fix memory leak during multi-message connections using STARTTLS under
125 OpenSSL. Certificate information is loaded for every new TLS startup,
126 and the resources needed to be freed.
128 JH/24 Bug 2242: Fix exim_dbmbuild to permit directoryless filenames.
130 JH/25 Fix utf8_downconvert propagation through a redirect router. Previously it
133 JH/26 Bug 2253: For logging delivery lines under PRDR, append the overall
134 DATA response info to the (existing) per-recipient response info for
135 the "C=" log element. It can have useful tracking info from the
136 destination system. Patch from Simon Arlott.
138 JH/27 Bug 2251: Fix ldap lookups that return a single attribute having zero-
139 length value. Previously this would segfault.
141 HS/02 Support Avast multiline protoocol, this allows passing flags to
142 newer versions of the scanner.
144 JH/28 Ensure that variables possibly set during message acceptance are marked
145 dead before release of memory in the daemon loop. This stops complaints
146 about them when the debug_store option is enabled. Discovered specifically
147 for sender_rate_period, but applies to a whole set of variables.
148 Do the same for the queue-runner loop, for variables set from spool
149 message files. Do the same for the SMTP per-message loop, for certain
150 variables indirectly set in ACL operations.
152 JH/29 Bug 2250: Fix a longstanding bug in heavily-pipelined SMTP input (such
153 as a multi-recipient message from a mailinglist manager). The coding had
154 an arbitrary cutoff number of characters while checking for more input;
155 enforced by writing a NUL into the buffer. This corrupted long / fast
156 input. The problem was exposed more widely when more pipelineing of SMTP
157 responses was introduced, and one Exim system was feeding another.
158 The symptom is log complaints of SMTP syntax error (NUL chars) on the
159 receiving system, and refused recipients seen by the sending system
160 (propating to people being dropped from mailing lists).
161 Discovered and pinpointed by David Carter.
167 JH/01 Rework error string handling in TLS interface so that the caller in
168 more cases is responsible for logging. This permits library-sourced
169 string to be attached to addresses during delivery, and collapses
170 pairs of long lines into single ones.
172 PP/01 Allow PKG_CONFIG_PATH to be set in Local/Makefile and use it correctly
173 during configuration. Wildcards are allowed and expanded.
175 JH/02 Rework error string handling in DKIM to pass more info back to callers.
176 This permits better logging.
178 JH/03 Rework the transport continued-connection mechanism: when TLS is active,
179 do not close it down and have the child transport start it up again on
180 the passed-on TCP connection. Instead, proxy the child (and any
181 subsequent ones) for TLS via a unix-domain socket channel. Logging is
182 affected: the continued delivery log lines do not have any DNSSEC, TLS
183 Certificate or OCSP information. TLS cipher information is still logged.
185 JH/04 Shorten the log line for daemon startup by collapsing adjacent sets of
186 identical IP addresses on different listening ports. Will also affect
189 PP/02 Bug 2070: uClibc defines __GLIBC__ without providing glibc headers;
190 add noisy ifdef guards to special-case this sillyness.
191 Patch from Bernd Kuhls.
193 JH/05 Tighten up the checking in isip4 (et al): dotted-quad components larger
194 than 255 are no longer allowed.
196 JH/06 Default openssl_options to include +no_ticket, to reduce load on peers.
197 Disable the session-cache too, which might reduce our load. Since we
198 currrectly use a new context for every connection, both as server and
199 client, there is no benefit for these.
200 GnuTLS appears to not support tickets server-side by default (we don't
201 call gnutls_session_ticket_enable_server()) but client side is enabled
202 by default on recent versions (3.1.3 +) unless the PFS priority string
205 PP/03 Add $SOURCE_DATE_EPOCH support for reproducible builds, per spec at
206 <https://reproducible-builds.org/specs/source-date-epoch/>.
208 JH/07 Fix smtp transport use of limited max_rcpt under mua_wrapper. Previously
209 the check for any unsuccessful recipients did not notice the limit, and
210 erroneously found still-pending ones.
212 JH/08 Pipeline CHUNKING command and data together, on kernels that support
213 MSG_MORE. Only in-clear (not on TLS connections).
215 JH/09 Avoid using a temporary file during transport using dkim. Unless a
216 transport-filter is involved we can buffer the headers in memory for
217 creating the signature, and read the spool data file once for the
218 signature and again for transmission.
220 JH/10 Enable use of sendfile in Linux builds as default. It was disabled in
221 4.77 as the kernel support then wasn't solid, having issues in 64bit
222 mode. Now, it's been long enough. Add support for FreeBSD also.
224 JH/11 Bug 2104: Fix continued use of a transport connection with TLS. In the
225 case where the routing stage had gathered several addresses to send to
226 a host before calling the transport for the first, we previously failed
227 to close down TLS in the old transport process before passing the TCP
228 connection to the new process. The new one sent a STARTTLS command
229 which naturally failed, giving a failed delivery and bloating the retry
230 database. Investigation and fix prototype from Wolfgang Breyha.
232 JH/12 Fix check on SMTP command input synchronisation. Previously there were
233 false-negatives in the check that the sender had not preempted a response
234 or prompt from Exim (running as a server), due to that code's lack of
235 awareness of the SMTP input buffering.
237 PP/04 Add commandline_checks_require_admin option.
238 Exim drops privileges sanely, various checks such as -be aren't a
239 security problem, as long as you trust local users with access to their
240 own account. When invoked by services which pass untrusted data to
241 Exim, this might be an issue. Set this option in main configuration
242 AND make fixes to the calling application, such as using `--` to stop
245 JH/13 Do pipelining under TLS. Previously, although safe, no advantage was
246 taken. Now take care to pack both (client) MAIL,RCPT,DATA, and (server)
247 responses to those, into a single TLS record each way (this usually means
248 a single packet). As a side issue, smtp_enforce_sync now works on TLS
251 PP/05 OpenSSL/1.1: use DH_bits() for more accurate DH param sizes. This
252 affects you only if you're dancing at the edge of the param size limits.
253 If you are, and this message makes sense to you, then: raise the
254 configured limit or use OpenSSL 1.1. Nothing we can do for older
257 JH/14 For the "sock" variant of the malware scanner interface, accept an empty
258 cmdline element to get the documented default one. Previously it was
261 JH/15 Fix a crash in the smtp transport caused when two hosts in succession
262 are unsuable for non-message-specific reasons - eg. connection timeout,
263 banner-time rejection.
265 JH/16 Fix logging of delivery remote port, when specified by router, under
268 PP/06 Repair manualroute's ability to take options in any order, even if one
269 is the name of a transport.
272 HS/01 Cleanup, prevent repeated use of -p/-oMr (CVE-2017-1000369)
274 JH/17 Change the list-building routines interface to use the expanding-string
275 triplet model, for better allocation and copying behaviour.
277 JH/18 Prebuild the data-structure for "builtin" macros, for faster startup.
278 Previously it was constructed the first time a possibly-matching string
279 was met in the configuration file input during startup; now it is done
282 JH/19 Bug 2141: Use the full-complex API for Berkeley DB rather than the legacy-
283 compatible one, to avoid the (poorly documented) possibility of a config
284 file in the working directory redirecting the DB files, possibly correpting
285 some existing file. CVE-2017-10140 assigned for BDB.
287 JH/20 Bug 2147: Do not defer for a verify-with-callout-and-random which is not
288 cache-hot. Previously, although the result was properly cached, the
289 initial verify call returned a defer.
291 JH/21 Bug 2151: Avoid using SIZE on the MAIL for a callout verify, on any but
292 the main verify for receipient in uncached-mode.
294 JH/22 Retire historical build files to an "unsupported" subdir. These are
295 defined as "ones for which we have no current evidence of testing".
297 JH/23 DKIM: enforce the DNS pubkey record "h" permitted-hashes optional field,
298 if present. Previously it was ignored.
300 JH/24 Start using specified-initialisers in C structure init coding. This is
301 a C99 feature (it's 2017, so now considered safe).
303 JH/25 Use one-bit bitfields for flags in the "addr" data structure. Previously
304 if was a fixed-sized field and bitmask ops via macros; it is now more
307 PP/07 GitHub PR 56: Apply MariaDB build fix.
308 Patch provided by Jaroslav Škarvada.
310 PP/08 Bug 2161: Fix regression in sieve quoted-printable handling introduced
311 during Coverity cleanups [4.87 JH/47]
312 Diagnosis and fix provided by Michael Fischer v. Mollard.
314 JH/26 Fix DKIM bug: when the pseudoheader generated for signing was exactly
315 the right size to place the terminating semicolon on its own folded
316 line, the header hash was calculated to an incorrect value thanks to
317 the (relaxed) space the fold became.
319 HS/02 Fix Bug 2130: large writes from the transport subprocess where chunked
320 and confused the parent.
322 JH/27 Fix SOCKS bug: an unitialized pointer was deref'd by the transport process
323 which could crash as a result. This could lead to undeliverable messages.
325 JH/28 Logging: "next input sent too soon" now shows where input was truncated
328 JH/29 Fix queue_run_in_order to ignore the PID portion of the message ID. This
329 matters on fast-turnover and PID-randomising systems, which were getting
330 out-of-order delivery.
332 JH/30 Fix a logging bug on aarch64: an unsafe routine was previously used for
333 a possibly-overlapping copy. The symptom was that "Remote host closed
334 connection in response to HELO" was logged instead of the actual 4xx
337 JH/31 Fix CHUNKING code to properly flush the unwanted chunk after an error.
338 Previously only that bufferd was discarded, resulting in SYMTP command
341 JH/32 DKIM: when a message has multiple signatures matching an identity given
342 in dkim_verify_signers, run the dkim acl once for each. Previously only
343 one run was done. Bug 2189.
345 JH/33 Downgrade an unfound-list name (usually a typo in the config file) from
346 "panic the current process" to "deliberately defer". The panic log is
347 still written with the problem list name; the mail and reject logs now
348 get a temp-reject line for the message that was being handled, saying
349 something like "domains check lookup or other defer". The SMTP 451
350 message is still "Temporary local problem".
352 JH/34 Bug 2199: Fix a use-after-free while reading smtp input for header lines.
353 A crafted sequence of BDAT commands could result in in-use memory beeing
354 freed. CVE-2017-16943.
356 HS/03 Bug 2201: Fix checking for leading-dot on a line during headers reading
357 from SMTP input. Previously it was always done; now only done for DATA
358 and not BDAT commands. CVE-2017-16944.
360 JH/35 Bug 2201: Flush received data in BDAT mode after detecting an error fatal
361 to the message (such as an overlong header line). Previously this was
362 not done and we did not exit BDAT mode. Followon from the previous item
363 though a different problem.
369 JH/01 Bug 1922: Support IDNA2008. This has slightly different conversion rules
370 than -2003 did; needs libidn2 in addition to libidn.
372 JH/02 The path option on a pipe transport is now expanded before use.
374 PP/01 GitHub PR 50: Do not call ldap_start_tls_s on ldapi:// connections.
375 Patch provided by "Björn", documentation fix added too.
377 JH/03 Bug 2003: fix Proxy Protocol v2 handling: the address size field was
378 missing a wire-to-host endian conversion.
380 JH/04 Bug 2004: fix CHUNKING in non-PIPELINEING mode. Chunk data following
381 close after a BDAT command line could be taken as a following command,
382 giving a synch failure. Fix by only checking for synch immediately
383 before acknowledging the chunk.
385 PP/02 GitHub PR 52: many spelling fixes, which include fixing parsing of
386 no_require_dnssec option and creation of _HAVE_TRANSPORT_APPEND_MAILDIR
387 macro. Patches provided by Josh Soref.
389 JH/05 Have the EHLO response advertise VRFY, if there is a vrfy ACL defined.
390 Previously we did not; the RFC seems ambiguous and VRFY is not listed
391 by IANA as a service extension. However, John Klensin suggests that we
394 JH/06 Bug 2017: Fix DKIM verification in -bh test mode. The data feed into
395 the dkim code may be unix-mode line endings rather than smtp wire-format
396 CRLF, so prepend a CR to any bare LF.
398 JH/07 Rationalise the coding for callout smtp conversations and transport ones.
399 As a side-benfit, callouts can now use PIPELINING hence fewer round-trips.
401 JH/08 Bug 2016: Fix DKIM verification vs. CHUNKING. Any BDAT commands after
402 the first were themselves being wrongly included in the feed into dkim
403 processing; with most chunk sizes in use this resulted in an incorrect
404 body hash calculated value.
406 JH/09 Bug 2014: permit inclusion of a DKIM-Signature header in a received
407 DKIM signature block, for verification. Although advised against by
408 standards it is specifically not ruled illegal.
410 JH/10 Bug 2025: Fix reception of (quoted) local-parts with embedded spaces.
412 JH/11 Bug 2029: Fix crash in DKIM verification when a message signature block is
413 missing a body hash (the bh= tag).
415 JH/12 Bug 2018: Re-order Proxy Protocol startup versus TLS-on-connect startup.
416 It seems that HAProxy sends the Proxy Protocol information in clear and
417 only then does a TLS startup, so do the same.
419 JH/13 Bug 2027: Avoid attempting to use TCP Fast Open for non-transport client
420 TCP connections (such as for Spamd) unless the daemon successfully set
421 Fast Open mode on its listening sockets. This fixes breakage seen on
422 too-old kernels or those not configured for Fast Open, at the cost of
423 requiring both directions being enabled for TFO, and TFO never being used
424 by non-daemon-related Exim processes.
426 JH/14 Bug 2000: Reject messages recieved with CHUNKING but with malformed line
427 endings, at least on the first header line. Try to canonify any that get
428 past that check, despite the cost.
430 JH/15 Angle-bracket nesting (an error inserted by broken sendmails) levels are
431 now limited to an arbitrary five deep, while parsing addresses with the
432 strip_excess_angle_brackets option enabled.
434 PP/03 Bug 2018: For Proxy Protocol and TLS-on-connect, do not over-read and
435 instead leave the unprompted TLS handshake in socket buffer for the
436 TLS library to consume.
438 PP/04 Bug 2018: Also handle Proxy Protocol v2 safely.
440 PP/05 FreeBSD compat: handle that Ports no longer create /usr/bin/perl
442 JH/16 Drop variables when they go out of scope. Memory management drops a whole
443 region in one operation, for speed, and this leaves assigned pointers
444 dangling. Add checks run only under the testsuite which checks all
445 variables at a store-reset and panics on a dangling pointer; add code
446 explicitly nulling out all the variables discovered. Fixes one known
447 bug: a transport crash, where a dangling pointer for $sending_ip_address
448 originally assigned in a verify callout, is re-used.
450 PP/06 Drop '.' from @INC in various Perl scripts.
452 PP/07 Switch FreeBSD iconv to always use the base-system libc functions.
454 PP/08 Reduce a number of compilation warnings under clang; building with
455 CC=clang CFLAGS+=-Wno-dangling-else -Wno-logical-op-parentheses
456 should be warning-free.
458 JH/17 Fix inbound CHUNKING when DKIM disabled at runtime.
460 HS/01 Fix portability problems introduced by PP/08 for platforms where
461 realloc(NULL) is not equivalent to malloc() [SunOS et al].
463 HS/02 Bug 1974: Fix missing line terminator on the last received BDAT
464 chunk. This allows us to accept broken chunked messages. We need a more
465 general solution here.
467 PP/09 Wrote util/chunking_fixqueue_finalnewlines.pl to help recover
468 already-broken messages in the queue.
470 JH/18 Bug 2061: Fix ${extract } corrupting an enclosing ${reduce } $value.
472 JH/19 Fix reference counting bug in routing-generated-address tracking.
478 JH/01 Use SIZE on MAIL FROM in a cutthrough connection, if the destination
479 supports it and a size is available (ie. the sending peer gave us one).
481 JH/02 The obsolete acl condition "demime" is removed (finally, after ten
482 years of being deprecated). The replacements are the ACLs
483 acl_smtp_mime and acl_not_smtp_mime.
485 JH/03 Upgrade security requirements imposed for hosts_try_dane: previously
486 a downgraded non-dane trust-anchor for the TLS connection (CA-style)
487 or even an in-clear connection were permitted. Now, if the host lookup
488 was dnssec and dane was requested then the host is only used if the
489 TLSA lookup succeeds and is dnssec. Further hosts (eg. lower priority
490 MXs) will be tried (for hosts_try_dane though not for hosts_require_dane)
491 if one fails this test.
492 This means that a poorly-configured remote DNS will make it incommunicado;
493 but it protects against a DNS-interception attack on it.
495 JH/04 Bug 1810: make continued-use of an open smtp transport connection
496 non-noisy when a race steals the message being considered.
498 JH/05 If main configuration option tls_certificate is unset, generate a
499 self-signed certificate for inbound TLS connections.
501 JH/06 Bug 165: hide more cases of password exposure - this time in expansions
502 in rewrites and routers.
504 JH/07 Retire gnutls_require_mac et.al. These were nonfunctional since 4.80
505 and logged a warning sing 4.83; now they are a configuration file error.
507 JH/08 Bug 1836: Fix crash in VRFY handling when handed an unqualified name
508 (lacking @domain). Apply the same qualification processing as RCPT.
510 JH/09 Bug 1804: Avoid writing msglog files when in -bh or -bhc mode.
512 JH/10 Support ${sha256:} applied to a string (as well as the previous
515 JH/11 Cutthrough: avoid using the callout hints db on a verify callout when
516 a cutthrough deliver is pending, as we always want to make a connection.
517 This also avoids re-routing the message when later placing the cutthrough
518 connection after a verify cache hit.
519 Do not update it with the verify result either.
521 JH/12 Cutthrough: disable when verify option success_on_redirect is used, and
522 when routing results in more than one destination address.
524 JH/13 Cutthrough: expand transport dkim_domain option when testing for dkim
525 signing (which inhibits the cutthrough capability). Previously only
526 the presence of an option was tested; now an expansion evaluating as
527 empty is permissible (obviously it should depend only on data available
528 when the cutthrough connection is made).
530 JH/14 Fix logging of errors under PIPELINING. Previously the log line giving
531 the relevant preceding SMTP command did not note the pipelining mode.
533 JH/15 Fix counting of empty lines in $body_linecount and $message_linecount.
534 Previously they were not counted.
536 JH/16 DANE: treat a TLSA lookup response having all non-TLSA RRs, the same
537 as one having no matching records. Previously we deferred the message
538 that needed the lookup.
540 JH/17 Fakereject: previously logged as a normal message arrival "<="; now
541 distinguished as "(=".
543 JH/18 Bug 1867: make the fail_defer_domains option on a dnslookup router work
544 for missing MX records. Previously it only worked for missing A records.
546 JH/19 Bug 1850: support Radius libraries that return REJECT_RC.
548 JH/20 Bug 1872: Ensure that acl_smtp_notquit is run when the connection drops
549 after the data-go-ahead and data-ack. Patch from Jason Betts.
551 JH/21 Bug 1846: Send DMARC forensic reports for reject and quarantine results,
552 even for a "none" policy. Patch from Tony Meyer.
554 JH/22 Fix continued use of a connection for further deliveries. If a port was
555 specified by a router, it must also match for the delivery to be
558 JH/23 Bug 1874: fix continued use of a connection for further deliveries.
559 When one of the recipients of a message was unsuitable for the connection
560 (has no matching addresses), we lost track of needing to mark it
561 deferred. As a result mail would be lost.
563 JH/24 Bug 1832: Log EHLO response on getting conn-close response for HELO.
565 JH/25 Decoding ACL controls is now done using a binary search; the source code
566 takes up less space and should be simpler to maintain. Merge the ACL
567 condition decode tables also, with similar effect.
569 JH/26 Fix problem with one_time used on a redirect router which returned the
570 parent address unchanged. A retry would see the parent address marked as
571 delivered, so not attempt the (identical) child. As a result mail would
574 JH/27 Fix a possible security hole, wherein a process operating with the Exim
575 UID can gain a root shell. Credit to http://www.halfdog.net/ for
576 discovery and writeup. Ubuntu bug 1580454; no bug raised against Exim
579 JH/28 Enable {spool,log} filesystem space and inode checks as default.
580 Main config options check_{log,spool}_{inodes,space} are now
581 100 inodes, 10MB unless set otherwise in the configuration.
583 JH/29 Fix the connection_reject log selector to apply to the connect ACL.
584 Previously it only applied to the main-section connection policy
587 JH/30 Bug 1897: fix callouts connection fallback from TLS to cleartext.
589 PP/01 Changed default Diffie-Hellman parameters to be Exim-specific, created
590 by me. Added RFC7919 DH primes as an alternative.
592 PP/02 Unbreak build via pkg-config with new hash support when crypto headers
593 are not in the system include path.
595 JH/31 Fix longstanding bug with aborted TLS server connection handling. Under
596 GnuTLS, when a session startup failed (eg because the client disconnected)
597 Exim did stdio operations after fclose. This was exposed by a recent
598 change which nulled out the file handle after the fclose.
600 JH/32 Bug 1909: Fix OCSP proof verification for cases where the proof is
601 signed directly by the cert-signing cert, rather than an intermediate
602 OCSP-signing cert. This is the model used by LetsEncrypt.
604 JH/33 Bug 1914: Ensure socket is nonblocking before draining after SMTP QUIT.
606 HS/01 Fix leak in verify callout under GnuTLS, about 3MB per recipient on
607 an incoming connection.
609 HS/02 Bug 1802: Do not half-close the connection after sending a request
612 HS/03 Use "auto" as the default EC curve parameter. For OpenSSL < 1.0.2
613 fallback to "prime256v1".
615 JH/34 SECURITY: Use proper copy of DATA command in error message.
616 Could leak key material. Remotely exploitable. CVE-2016-9963.
622 JH/01 Bug 1664: Disable OCSP for GnuTLS library versions at/before 3.3.16
623 and 3.4.4 - once the server is enabled to respond to an OCSP request
624 it does even when not requested, resulting in a stapling non-aware
625 client dropping the TLS connection.
627 TF/01 Code cleanup: Overhaul the debug_selector and log_selector machinery to
628 support variable-length bit vectors. No functional change.
630 TF/02 Improve the consistency of logging incoming and outgoing interfaces.
631 The I= interface field on outgoing lines is now after the H= remote
632 host field, same as incoming lines. There is a separate
633 outgoing_interface log selector which allows you to disable the
636 JH/02 Bug 728: Close logfiles after a daemon-process "exceptional" log write.
637 If not running log_selector +smtp_connection the mainlog would be held
638 open indefinitely after a "too many connections" event, including to a
639 deleted file after a log rotate. Leave the per net connection logging
640 leaving it open for efficiency as that will be quickly detected by the
641 check on the next write.
643 HS/01 Bug 1671: Fix post transport crash.
644 Processing the wait-<transport> messages could crash the delivery
645 process if the message IDs didn't exist for some reason. When
646 using 'split_spool_directory=yes' the construction of the spool
647 file name failed already, exposing the same netto behaviour.
649 JH/03 Bug 425: Capture substrings in $regex1, $regex2 etc from regex &
650 mime_regex ACL conditions.
652 JH/04 Bug 1686: When compiled with EXPERIMENTAL_DSN_INFO: Add extra information
653 to DSN fail messages (bounces): remote IP, remote greeting, remote response
654 to HELO, local diagnostic string.
656 JH/05 Downgrade message for a TLS-certificate-based authentication fail from
657 log line to debug. Even when configured with a tls authenticator many
658 client connections are expected to not authenticate in this way, so
659 an authenticate fail is not an error.
661 HS/02 Add the Exim version string to the process info. This way exiwhat
662 gives some more detail about the running daemon.
664 JH/06 Bug 1395: time-limit caching of DNS lookups, to the TTL value. This may
665 matter for fast-change records such as DNSBLs.
667 JH/07 Bug 1678: Always record an interface option value, if set, as part of a
668 retry record, even if constant. There may be multiple transports with
669 different interface settings and the retry behaviour needs to be kept
672 JH/08 Bug 1586: exiqgrep now refuses to run if there are unexpected arguments.
674 JH/09 Bug 1700: ignore space & tab embedded in base64 during decode.
676 JH/10 Bug 840: fix log_defer_output option of pipe transport
678 JH/11 Bug 830: use same host for all RCPTS of a message, even under
679 hosts_randomize. This matters a lot when combined with mua_wrapper.
681 JH/12 Bug 1706: percent and underbar characters are no longer escaped by the
682 ${quote_pgsql:<string>} operator.
684 JH/13 Bug 1708: avoid misaligned access in cached lookup.
686 JH/14 Change header file name for freeradius-client. Relevant if compiling
687 with Radius support; from the Gentoo tree and checked under Fedora.
689 JH/15 Bug 1712: Introduce $prdr_requested flag variable
691 JH/16 Bug 1714: Permit an empty string as expansion result for transport
692 option transport_filter, meaning no filtering.
694 JH/17 Bug 1713: Fix non-PDKIM_DEBUG build. Patch from Jasen Betts.
696 JH/18 Bug 1709: When built with TLS support, the tls_advertise_hosts option now
697 defaults to "*" (all hosts). The variable is now available when not built
698 with TLS, default unset, mainly to enable keeping the testsuite sane.
699 If a server certificate is not supplied (via tls_certificate) an error is
700 logged, and clients will find TLS connections fail on startup. Presumably
701 they will retry in-clear.
702 Packagers of Exim are strongly encouraged to create a server certificate
703 at installation time.
705 HS/03 Add -bP config_file as a synonym for -bP configure_file, for consistency
706 with the $config_file variable.
708 JH/19 Two additional event types: msg:rcpt:defer and msg:rcpt:host:defer. Both
709 in transport context, after the attempt, and per-recipient. The latter type
710 is per host attempted. The event data is the error message, and the errno
711 information encodes the lookup type (A vs. MX) used for the (first) host,
712 and the trailing two digits of the smtp 4xx response.
714 GF/01 Bug 1715: Fix for race condition in exicyclog, where exim could attempt
715 to write to mainlog (or rejectlog, paniclog) in the window between file
716 creation and permissions/ownership being changed. Particularly affects
717 installations where exicyclog is run as root, rather than exim user;
718 result is that the running daemon panics and dies.
720 JH/20 Bug 1701: For MySQL lookups, support MySQL config file option group names.
722 JH/21 Bug 1720: Add support for priority groups and weighted-random proxy
723 selection for the EXPERIMENTAL_SOCKS feature, via new per-proxy options
724 "pri" and "weight". Note that the previous implicit priority given by the
725 list order is no longer honoured.
727 JH/22 Bugs 963, 1721: Fix some corner cases in message body canonicalization
730 JH/23 Move SOCKS5 support from Experimental to mainline, enabled for a build
731 by defining SUPPORT_SOCKS.
733 JH/26 Move PROXY support from Experimental to mainline, enabled for a build
734 by defining SUPPORT_PROXY. Note that the proxy_required_hosts option
735 is renamed to hosts_proxy, and the proxy_{host,target}_{address,port}.
736 variables are renamed to proxy_{local,external}_{address,port}.
738 JH/27 Move Internationalisation support from Experimental to mainline, enabled
739 for a build by defining SUPPORT_I18N
741 JH/28 Bug 1745: Fix redis lookups to handle (quoted) spaces embedded in parts
742 of the query string, and make ${quote_redis:} do that quoting.
744 JH/29 Move Events support from Experimental to mainline, enabled by default
745 and removable for a build by defining DISABLE_EVENT.
747 JH/30 Updated DANE implementation code to current from Viktor Dukhovni.
749 JH/31 Fix bug with hosts_connection_nolog and named-lists which were wrongly
750 cached by the daemon.
752 JH/32 Move Redis support from Experimental to mainline, enabled for a build
753 by defining LOOKUP_REDIS. The libhiredis library is required.
755 JH/33 Bug 1748: Permit ACL dnslists= condition in non-smtp ACLs if explicit
756 keys are given for lookup.
758 JH/34 Bug 1192: replace the embedded copy of PolarSSL RSA routines in the DKIM
759 support, by using OpenSSL or GnuTLS library ones. This means DKIM is
760 only supported when built with TLS support. The PolarSSL SHA routines
761 are still used when the TLS library is too old for convenient support.
763 JH/35 Require SINGLE_DH_USE by default in OpenSSL (main config option
764 openssl_options), for security. OpenSSL forces this from version 1.1.0
765 server-side so match that on older versions.
767 JH/36 Bug 1778: longstanding bug in memory use by the ${run } expansion: A fresh
768 allocation for $value could be released as the expansion processing
769 concluded, but leaving the global pointer active for it.
771 JH/37 Bug 1769: Permit a VRFY ACL to override the default 252 response,
772 and to use the domains and local_parts ACL conditions.
774 JH/38 Fix cutthrough bug with body lines having a single dot. The dot was
775 incorrectly not doubled on cutthrough transmission, hence seen as a
776 body-termination at the receiving system - resulting in truncated mails.
777 Commonly the sender saw a TCP-level error, and retransmitted the message
778 via the normal store-and-forward channel. This could result in duplicates
779 received - but deduplicating mailstores were liable to retain only the
780 initial truncated version.
782 JH/39 Bug 1781: Fix use of DKIM private-keys having trailing '=' in the base-64.
784 JH/40 Fix crash in queryprogram router when compiled with EXPERIMENTAL_SRS.
786 JH/41 Bug 1792: Fix selection of headers to sign for DKIM: bottom-up. While
787 we're in there, support oversigning also; bug 1309.
789 JH/42 Bug 1796: Fix error logged on a malware scanner connection failure.
791 HS/04 Add support for keep_environment and add_environment options.
793 JH/43 Tidy coding issues detected by gcc --fsanitize=undefined. Some remain;
794 either intentional arithmetic overflow during PRNG, or testing config-
797 JH/44 Bug 1800: The combination of a -bhc commandline option and cutthrough
798 delivery resulted in actual delivery. Cancel cutthrough before DATA
801 JH/45 Fix cutthrough, when connection not opened by verify and target hard-
802 rejects a recipient: pass the reject to the originator.
804 JH/46 Multiple issues raised by Coverity. Some were obvious or plausible bugs.
805 Many were false-positives and ignorable, but it's worth fixing the
808 JH/47 Fix build on HP-UX and older Solaris, which need (un)setenv now also
809 for the new environment-manipulation done at startup. Move the routines
810 from being local to tls.c to being global via the os.c file.
812 JH/48 Bug 1807: Fix ${extract } for the numeric/3-string case. While preparsing
813 an extract embedded as result-arg for a map, the first arg for extract
814 is unavailable so we cannot tell if this is a numbered or keyed
815 extraction. Accept either.
821 JH/01 Bug 1545: The smtp transport option "retry_include_ip_address" is now
824 JH/02 The smtp transport option "multi_domain" is now expanded.
826 JH/03 The smtp transport now requests PRDR by default, if the server offers
829 JH/04 Certificate name checking on server certificates, when exim is a client,
830 is now done by default. The transport option tls_verify_cert_hostnames
831 can be used to disable this per-host. The build option
832 EXPERIMENTAL_CERTNAMES is withdrawn.
834 JH/05 The value of the tls_verify_certificates smtp transport and main options
835 default to the word "system" to access the system default CA bundle.
836 For GnuTLS, only version 3.0.20 or later.
838 JH/06 Verification of the server certificate for a TLS connection is now tried
839 (but not required) by default. The verification status is now logged by
840 default, for both outbound TLS and client-certificate supplying inbound
843 JH/07 Changed the default rfc1413 lookup settings to disable calls. Few
846 JH/08 The EXPERIMENTAL_DSN compile option is no longer needed; all Delivery
847 Status Notification (bounce) messages are now MIME format per RFC 3464.
848 Support for RFC 3461 DSN options NOTIFY,ENVID,RET,ORCPT can be advertised
849 under the control of the dsn_advertise_hosts option, and routers may
850 have a dsn_lasthop option.
852 JH/09 A timeout of 2 minutes is now applied to all malware scanner types by
853 default, modifiable by a malware= option. The list separator for
854 the options can now be changed in the usual way. Bug 68.
856 JH/10 The smtp_receive_timeout main option is now expanded before use.
858 JH/11 The incoming_interface log option now also enables logging of the
859 local interface on delivery outgoing connections.
861 JH/12 The cutthrough-routing facility now supports multi-recipient mails,
862 if the interface and destination host and port all match.
864 JH/13 Bug 344: The verify = reverse_host_lookup ACL condition now accepts a
867 JH/14 Bug 1573: The spam= ACL condition now additionally supports Rspamd.
868 Patch from Andrew Lewis.
870 JH/15 Bug 670: The spamd_address main option (for the spam= ACL condition)
871 now supports optional time-restrictions, weighting, and priority
872 modifiers per server. Patch originally by <rommer@active.by>.
874 JH/16 The spamd_address main option now supports a mixed list of local
875 and remote servers. Remote servers can be IPv6 addresses, and
876 specify a port-range.
878 JH/17 Bug 68: The spamd_address main option now supports an optional
879 timeout value per server.
881 JH/18 Bug 1581: Router and transport options headers_add/remove can
882 now have the list separator specified.
884 JH/19 Bug 392: spamd_address, and clamd av_scanner, now support retry
887 JH/20 Bug 1571: Ensure that $tls_in_peerdn is set, when verification fails
890 JH/21 Support for the A6 type of dns record is withdrawn.
892 JH/22 Bug 608: The result of a QUIT or not-QUIT toplevel ACL now matters
893 rather than the verbs used.
895 JH/23 Bug 1572: Increase limit on SMTP confirmation message copy size
896 from 255 to 1024 chars.
898 JH/24 Verification callouts now attempt to use TLS by default.
900 HS/01 DNSSEC options (dnssec_require_domains, dnssec_request_domains)
901 are generic router options now. The defaults didn't change.
903 JH/25 Bug 466: Add RFC2322 support for MIME attachment filenames.
904 Original patch from Alexander Shikoff, worked over by JH.
906 HS/02 Bug 1575: exigrep falls back to autodetection of compressed
907 files if ZCAT_COMMAND is not executable.
909 JH/26 Bug 1539: Add timeout/retry options on dnsdb lookups.
911 JH/27 Bug 286: Support SOA lookup in dnsdb lookups.
913 JH/28 Bug 1588: Do not use the A lookup following an AAAA for setting the FQDN.
914 Normally benign, it bites when the pair was led to by a CNAME;
915 modern usage is to not canonicalize the domain to a CNAME target
916 (and we were inconsistent anyway for A-only vs AAAA+A).
918 JH/29 Bug 1632: Removed the word "rejected" from line logged for ACL discards.
920 JH/30 Check the forward DNS lookup for DNSSEC, in addition to the reverse,
921 when evaluating $sender_host_dnssec.
923 JH/31 Check the HELO verification lookup for DNSSEC, adding new
924 $sender_helo_dnssec variable.
926 JH/32 Bug 1397: Enable ECDHE on OpenSSL, just the NIST P-256 curve.
928 JH/33 Bug 1346: Note MAIL cmd seen in -bS batch, to avoid smtp_no_mail log.
930 JH/34 Bug 1648: Fix a memory leak seen with "mailq" and large queues.
932 JH/35 Bug 1642: Fix support of $spam_ variables at delivery time. Was
933 documented as working, but never had. Support all but $spam_report.
935 JH/36 Bug 1659: Guard checking of input smtp commands again pseudo-command
936 added for tls authenticator.
938 HS/03 Add perl_taintmode main config option
944 TL/01 When running the test suite, the README says that variables such as
945 no_msglog_check are global and can be placed anywhere in a specific
946 test's script, however it was observed that placement needed to be near
947 the beginning for it to behave that way. Changed the runtest perl
948 script to read through the entire script once to detect and set these
949 variables, reset to the beginning of the script, and then run through
950 the script parsing/test process like normal.
952 TL/02 The BSD's have an arc4random API. One of the functions to induce
953 adding randomness was arc4random_stir(), but it has been removed in
954 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
955 function when detected.
957 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
958 cause callback expansion.
960 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
961 syntax errors in an expansion can be treated as a string instead of
962 logging or causing an error, due to the internal use of bool_lax
963 instead of bool when processing it.
965 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
966 server certificates when making smtp deliveries.
968 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
970 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
972 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
974 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
975 Merged patch from Sebastian Wiedenroth.
977 JH/05 Fix results-pipe from transport process. Several recipients, combined
978 with certificate use, exposed issues where response data items split
979 over buffer boundaries were not parsed properly. This eventually
980 resulted in duplicates being sent. This issue only became common enough
981 to notice due to the introduction of connection certificate information,
982 the item size being so much larger. Found and fixed by Wolfgang Breyha.
984 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
985 size buffer was used, resulting in syntax errors when an expansion
988 JH/07 Add support for directories of certificates when compiled with a GnuTLS
989 version 3.3.6 or later.
991 JH/08 Rename the TPDA experimental facility to Event Actions. The #ifdef
992 is EXPERIMENTAL_EVENT, the main-configuration and transport options
993 both become "event_action", the variables become $event_name, $event_data
994 and $event_defer_errno. There is a new variable $verify_mode, usable in
995 routers, transports and related events. The tls:cert event is now also
996 raised for inbound connections, if the main configuration event_action
999 TL/06 In test suite, disable OCSP for old versions of openssl which contained
1000 early OCSP support, but no stapling (appears to be less than 1.0.0).
1002 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
1003 server certificate names available under the smtp transport option
1004 "tls_verify_cert_hostname" now do not permit multi-component wildcard
1007 JH/10 Time-related extraction expansions from certificates now use the main
1008 option "timezone" setting for output formatting, and are consistent
1009 between OpenSSL and GnuTLS compilations. Bug 1541.
1011 JH/11 Fix a crash in mime ACL when meeting a zero-length, quoted or RFC2047-
1012 encoded parameter in the incoming message. Bug 1558.
1014 JH/12 Bug 1527: Autogrow buffer used in reading spool files. Since they now
1015 include certificate info, eximon was claiming there were spoolfile
1018 JH/13 Bug 1521: Fix ldap lookup for single-attr request, multiple-attr return.
1020 JH/14 Log delivery-related information more consistently, using the sequence
1021 "H=<name> [<ip>]" wherever possible.
1023 TL/07 Bug 1547: Omit RFCs from release. Draft and RFCs have licenses which
1024 are problematic for Debian distribution, omit them from the release
1027 JH/15 Updates and fixes to the EXPERIMENTAL_DSN feature.
1029 JH/16 Fix string representation of time values on 64bit time_t architectures.
1032 JH/17 Fix a null-indirection in certextract expansions when a nondefault
1033 output list separator was used.
1038 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
1039 checkers that were complaining about end of non-void function with no
1042 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
1043 This was a regression introduced in 4.83 by another bugfix.
1045 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
1047 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
1048 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
1054 TF/01 Correctly close the server side of TLS when forking for delivery.
1056 When a message was received over SMTP with TLS, Exim failed to clear up
1057 the incoming connection properly after forking off the child process to
1058 deliver the message. In some situations the subsequent outgoing
1059 delivery connection happened to have the same fd number as the incoming
1060 connection previously had. Exim would try to use TLS and fail, logging
1061 a "Bad file descriptor" error.
1063 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
1064 utilities have not been installed.
1066 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
1067 temporary space as the ACL may create new global variables.
1069 TL/01 LDAP support uses per connection or global context settings, depending
1070 upon the detected version of the libraries at build time.
1072 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
1073 to extract and use the src ip:port in logging and expansions as if it
1074 were a direct connection from the outside internet. PPv2 support was
1075 updated based on HAProxy spec change in May 2014.
1077 JH/02 Add ${listextract {number}{list}{success}{fail}}.
1079 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
1080 Properly escape header and check for NULL return.
1082 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
1085 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
1087 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
1088 characters in header names, implemented as a verify condition.
1089 Contributed by Michael Fischer v. Mollard.
1091 TL/05 Rename SPF condition results err_perm and err_temp to standardized
1092 results permerror and temperror. Previous values are deprecated but
1093 still accepted. In a future release, err_perm and err_temp will be
1094 completely removed, which will be a backward incompatibility if the
1095 ACL tests for either of these two old results. Patch contributed by
1096 user bes-internal on the mailing list.
1098 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
1100 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
1101 selectors, in both main and reject logs.
1103 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
1106 JH/07 Add malware type "sock" for talking to simple daemon.
1108 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
1110 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
1111 routers/transports under cutthrough routing.
1113 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
1114 numbers. Touch up "bool" conditional to keep the same definition.
1116 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
1118 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
1120 JH/12 Expand items in router/transport headers_add or headers_remove lists
1121 individually rather than the list as a whole. Bug 1452.
1123 Required for reasonable handling of multiple headers_ options when
1124 they may be empty; requires that headers_remove items with embedded
1125 colons must have them doubled (or the list-separator changed).
1127 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
1128 view the policy declared in the DMARC record. Currently, $dmarc_status
1129 is a combined value of both the record presence and the result of the
1132 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
1134 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
1135 dnslookup router and the smtp transport (applying to the forward
1138 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
1139 of ldap servers used for a specific lookup. Patch provided by Heiko
1142 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
1143 New variable $lookup_dnssec_authenticated for observability.
1145 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
1146 Patch submitted by Lars Timman.
1148 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
1150 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
1151 Requires trusted mode and valid format message id, aborts otherwise.
1152 Patch contributed by Heiko Schlichting.
1154 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
1155 certextract with support for various fields. Bug 1358.
1157 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
1158 is requested by default, modifiable by smtp transport option
1161 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
1162 operate on certificate variables to give certificate fingerprints
1163 Also new ${sha256:cert_variable}.
1165 JH/23 The PRDR feature is moved from being Experimental into the mainline.
1167 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
1168 Christian Aistleitner.
1170 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
1172 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
1173 file. Patch from Wolfgang Breyha.
1175 JH/25 Expand the coverage of the delivery $host and $host_address to
1176 client authenticators run in verify callout. Bug 1476.
1178 JH/26 Port service names are now accepted for tls_on_connect_ports, to
1179 align with daemon_smtp_ports. Bug 72.
1181 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
1182 support and error reporting did not work properly.
1184 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
1185 and is readable. Patch from Andrew Colin Kissa.
1187 TL/14 Enhance documentation of ${run expansion and how it parses the
1188 commandline after expansion, particularly in the case when an
1189 unquoted variable expansion results in an empty value.
1191 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
1193 PP/02 Fix internal collision of T_APL on systems which support RFC3123
1194 by renaming away from it. Addresses GH issue 15, reported by
1197 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
1199 TL/15 SECURITY: prevent double expansion in math comparison functions
1200 (can expand unsanitized data). Not remotely exploitable.
1207 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
1209 PP/02 Make -n do something, by making it not do something.
1210 When combined with -bP, the name of an option is not output.
1212 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
1215 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
1216 $sender_host_name and config options to manage this, and basic check
1219 PP/05 DSCP support for outbound connections and control modifier for inbound.
1221 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
1222 (Only plugin which currently uses this is kerberos4, which nobody should
1223 be using, but we should make it available and other future plugins might
1224 conceivably use it, even though it would break NAT; stuff *should* be
1225 using channel bindings instead).
1227 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
1228 name; added for Sendmail compatibility; requires admin caller.
1229 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
1230 just ignore it); requires trusted caller.
1231 Also parse but ignore: -Ac -Am -X<logfile>
1234 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
1236 TL/02 Add +smtp_confirmation as a default logging option.
1238 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
1239 Patch by Magnus Holmgren from 2007-02-20.
1241 TL/04 Bugzilla 1281 - Spec typo.
1242 Bugzilla 1283 - Spec typo.
1243 Bugzilla 1290 - Spec grammar fixes.
1245 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
1247 TL/06 Add Experimental DMARC support using libopendmarc libraries.
1249 TL/07 Fix an out of order global option causing a segfault. Reported to dev
1250 mailing list by by Dmitry Isaikin.
1252 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
1254 JH/02 Support "G" suffix to numbers in ${if comparisons.
1256 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
1258 NM/01 Bugzilla 1197 - Spec typo
1259 Bugzilla 1196 - Spec examples corrections
1261 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
1263 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
1264 gnutls_enable_pkcs11, but renamed to more accurately indicate its
1267 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
1268 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
1270 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
1271 "acl {{name}{arg}...}", and optional args on acl condition
1274 JH/05 Permit multiple router/transport headers_add/remove lines.
1276 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
1278 JH/07 Avoid using a waiting database for a single-message-only transport.
1279 Performance patch from Paul Fisher. Bugzilla 1262.
1281 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
1284 JH/09 Add $headers_added variable, with content from use of ACL modifier
1285 add_header (but not yet added to the message). Bugzilla 199.
1287 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
1288 Pulled from Bugzilla 817 by Wolfgang Breyha.
1290 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
1292 (nb: this is the same fix as in Exim 4.80.1)
1294 JH/11 Add A= logging on delivery lines, and a client_set_id option on
1297 JH/12 Add optional authenticated_sender logging to A= and a log_selector
1300 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
1302 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
1303 advertise SMTP AUTH mechanism to us, instead of a generic
1304 protocol violation error. Also, make Exim more robust to bad
1305 data from the Dovecot auth socket.
1307 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
1309 When a queue runner is handling a message, Exim first routes the
1310 recipient addresses, during which it prunes them based on the retry
1311 hints database. After that it attempts to deliver the message to
1312 any remaining recipients. It then updates the hints database using
1315 So if a recipient address works intermittently, it can get repeatedly
1316 deferred at routing time. The retry hints record remains fresh so the
1317 address never reaches the final cutoff time.
1319 This is a fairly common occurrence when a user is bumping up against
1320 their storage quota. Exim had some logic in its local delivery code
1321 to deal with this. However it did not apply to per-recipient defers
1322 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
1324 This change adds a proper retry rule check during routing so that the
1325 final cutoff time is checked against the message's age. We only do
1326 this check if there is an address retry record and there is not a
1327 domain retry record; this implies that previous attempts to handle
1328 the address had the retry_use_local_parts option turned on. We use
1329 this as an approximation for the destination being like a local
1330 delivery, as in LMTP.
1332 I suspect this new check makes the old local delivery cutoff check
1333 redundant, but I have not verified this so I left the code in place.
1335 TF/02 Correct gecos expansion when From: is a prefix of the username.
1337 Test 0254 submits a message to Exim with the header
1341 When I ran the test suite under the user fanf2, Exim expanded
1342 the header to contain my full name, whereas it should have added
1343 a Resent-Sender: header. It erroneously treats any prefix of the
1344 username as equal to the username.
1346 This change corrects that bug.
1348 GF/01 DCC debug and logging tidyup
1349 Error conditions log to paniclog rather than rejectlog.
1350 Debug lines prefixed by "DCC: " to remove any ambiguity.
1352 TF/03 Avoid unnecessary rebuilds of lookup-related code.
1354 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
1355 Bug spotted by Jeremy Harris; was flawed since initial commit.
1356 Would have resulted in OCSP responses post-SNI triggering an Exim
1357 NULL dereference and crash.
1359 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
1361 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
1362 Bug detection, analysis and fix by Samuel Thibault.
1363 Bugzilla 1331, Debian bug #698092.
1365 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
1367 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
1368 Server implementation by Todd Lyons, client by JH.
1369 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
1370 config variable "prdr_enable" controls whether the server
1371 advertises the facility. If the client requests PRDR a new
1372 acl_data_smtp_prdr ACL is called once for each recipient, after
1373 the body content is received and before the acl_smtp_data ACL.
1374 The client is controlled by both of: a hosts_try_prdr option
1375 on the smtp transport, and the server advertisement.
1376 Default client logging of deliveries and rejections involving
1377 PRDR are flagged with the string "PRDR".
1379 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
1380 fclose(). Diagnosis by Todd Lyons.
1382 PP/17 Update configure.default to handle IPv6 localhost better.
1383 Patch by Alain Williams (plus minor tweaks).
1386 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
1387 This is now consistent with GnuTLS, and is now documented: the
1388 previous undocumented portable approach to treating the option as
1389 unset was to force an expansion failure. That still works, and
1390 an empty string is now equivalent.
1392 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
1393 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
1394 not performing validation itself.
1396 PP/20 Added force_command boolean option to pipe transport.
1397 Patch from Nick Koston, of cPanel Inc.
1399 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
1402 TF/04 Added udpsend ACL modifier and hexquote expansion operator
1404 PP/21 Fix eximon continuous updating with timestamped log-files.
1405 Broken in a format-string cleanup in 4.80, missed when I repaired the
1406 other false fix of the same issue.
1407 Report and fix from Heiko Schlichting.
1410 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
1411 Report from Prashanth Katuri.
1413 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
1414 It's SecureTransport, so affects any MacOS clients which use the
1415 system-integrated TLS libraries, including email clients.
1417 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
1418 using a MIME ACL for non-SMTP local injection.
1419 Report and assistance in diagnosis by Warren Baker.
1421 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
1423 JH/16 Fix comparisons for 64b. Bugzilla 1385.
1425 TL/09 Add expansion variable $authenticated_fail_id to keep track of
1426 last id that failed so it may be referenced in subsequent ACL's.
1428 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
1431 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
1432 ldap library initialization, allowing self-signed CA's to be
1433 used. Also properly sets require_cert option later in code by
1434 using NULL (global ldap config) instead of ldap handle (per
1435 session). Bug diagnosis and testing by alxgomz.
1437 TL/12 Enhanced documentation in the ratelimit.pl script provided in
1438 the src/util/ subdirectory.
1440 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
1441 renamed to Transport Post Delivery Action by Jeremy Harris, as
1444 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
1445 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
1446 redis_servers = needs to be configured which will be used by the redis
1447 lookup. Patch from Warren Baker, of The Packet Hub.
1449 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
1451 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
1452 hostname or reverse DNS when processing a host list. Used suggestions
1453 from multiple comments on this bug.
1455 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
1457 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
1458 Missed a few lines, added it to make the runtest require no keyboard
1461 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
1462 contains upper case chars. Make router use caseful_local_part.
1464 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
1465 support when GnuTLS has been built with p11-kit.
1471 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
1473 This, or similar/improved, will also be change PP/11 of 4.82.
1479 PP/01 Handle short writes when writing local log-files.
1480 In practice, only affects FreeBSD (8 onwards).
1481 Bugzilla 1053, with thanks to Dmitry Isaikin.
1483 NM/01 Bugzilla 949 - Documentation tweak
1485 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
1488 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
1490 PP/02 Implemented gsasl authenticator.
1492 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
1494 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
1495 `pkg-config foo` for cflags/libs.
1497 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
1498 with rest of GSASL and with heimdal_gssapi.
1500 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
1501 `pkg-config foo` for cflags/libs for the TLS implementation.
1503 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
1504 properties get this fed in as external SSF. A number of robustness
1505 and debugging improvements to the cyrus_sasl authenticator.
1507 PP/08 cyrus_sasl server now expands the server_realm option.
1509 PP/09 Bugzilla 1214 - Log authentication information in reject log.
1510 Patch by Jeremy Harris.
1512 PP/10 Added dbmjz lookup type.
1514 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
1516 PP/12 MAIL args handles TAB as well as SP, for better interop with
1517 non-compliant senders.
1518 Analysis and variant patch by Todd Lyons.
1520 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
1521 Bug report from Lars Müller <lars@samba.org> (via SUSE),
1522 Patch from Dirk Mueller <dmueller@suse.com>
1524 PP/13 tls_peerdn now print-escaped for spool files.
1525 Observed some $tls_peerdn in wild which contained \n, which resulted
1526 in spool file corruption.
1528 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
1529 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
1530 or write after TLS renegotiation, which otherwise led to messages
1533 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
1534 as a tracking header (ie: a signed header comes before the signature).
1535 Patch from Wolfgang Breyha.
1537 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
1538 comma-sep list; embedded commas doubled.
1540 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
1542 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
1544 Report and patch from Dmitry Banschikov.
1546 PP/16 Removed "dont_insert_empty_fragments" from "openssl_options".
1547 Removed SSL_clear() after SSL_new() which led to protocol negotiation
1548 failures. We appear to now support TLS1.1+ with Exim.
1550 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
1551 lets Exim select keys and certificates based upon TLS SNI from client.
1552 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
1553 before an outbound SMTP session. New log_selector, +tls_sni.
1555 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
1556 NULL dereference. Report and patch from Alun Jones.
1558 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
1559 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
1560 Not seeing resolver debug output on NetBSD, but suspect this is a
1561 resolver implementation change.
1563 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
1564 Left warnings. Added "eximon gdb" invocation mode.
1566 PP/21 Defaulting "accept_8bitmime" to true, not false.
1568 PP/22 Added -bw for inetd wait mode support.
1570 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
1571 locate the relevant includes and libraries. Made this the default.
1573 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
1574 Bugzilla 1246, report and most of solution from Tomasz Kusy.
1576 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
1577 This may cause build issues on older platforms.
1579 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
1580 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
1581 gnutls_require_mac & gnutls_require_protocols (no longer supported).
1582 Added SNI support via GnuTLS too.
1583 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
1585 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
1587 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
1588 Applied second patch from Janne, implementing suggestion to default
1589 multiple-strings-in-record handling to match SPF spec.
1591 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
1593 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
1594 read-only, out of scope).
1595 Patch from Wolfgang Breyha, report from Stuart Northfield.
1597 PP/29 Fix three issues highlighted by clang analyser static analysis.
1598 Only crash-plausible issue would require the Cambridge-specific
1599 iplookup router and a misconfiguration.
1600 Report from Marcin Mirosław.
1602 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
1604 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
1605 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
1606 As part of this, removing so much warning spew let me fix some minor
1607 real issues in debug logging.
1609 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
1610 assignment on my part. Fixed.
1612 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
1613 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
1614 Janne Snabb (who went above and beyond: thank you).
1616 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
1617 string otherwise requires a connection and a bunch more work and it's
1618 relatively easy to get wrong. Should also expose TLS library linkage
1621 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
1622 64-bit ${eval} (JH/03).
1624 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
1625 GNU libc to support some of the 64-bit stuff, should not lead to
1626 conflicts. Defined before os.h is pulled in, so if a given platform
1627 needs to override this, it can.
1629 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
1630 protection layer was required, which is not implemented.
1631 Bugzilla 1254, patch from Wolfgang Breyha.
1633 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
1634 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
1635 tls_dhparam take prime identifiers. Also unbreak combination of
1636 OpenSSL+DH_params+TLSSNI.
1638 PP/39 Disable SSLv2 by default in OpenSSL support.
1644 PP/01 Solaris build fix for Oracle's LDAP libraries.
1645 Bugzilla 1109, patch from Stephen Usher.
1647 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
1649 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
1652 TF/02 Fix a couple more cases where we did not log the error message
1653 when unlink() failed. See also change 4.74-TF/03.
1655 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
1656 lock up or crash if it happened to be inside a call to libc when it
1657 got a SIGUSR1 from exiwhat.
1659 The SIGUSR1 handler appends the current process status to the process
1660 log which is later printed by exiwhat. It used to use the general
1661 purpose logging code to do this, but several functions it calls are
1662 not safe for signals.
1664 The new output code in the SIGUSR1 handler is specific to the process
1665 log, and simple enough that it's easy to inspect for signal safety.
1666 Removing some special cases also simplifies the general logging code.
1667 Removing the spurious timestamps from the process log simplifies
1670 TF/04 Improved ratelimit ACL condition.
1672 The /noupdate option has been deprecated in favour of /readonly which
1673 has clearer semantics. The /leaky, /strict, and /readonly update modes
1674 are mutually exclusive. The update mode is no longer included in the
1675 database key; it just determines when the database is updated. (This
1676 means that when you upgrade Exim will forget old rate measurements.)
1678 Exim now checks that the per_* options are used with an update mode that
1679 makes sense for the current ACL. For example, when Exim is processing a
1680 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
1681 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
1682 must specify per_mail/readonly. If you omit the update mode it defaults to
1683 /leaky where that makes sense (as before) or /readonly where required.
1685 The /noupdate option is now undocumented but still supported for
1686 backwards compatibility. It is equivalent to /readonly except that in
1687 ACLs where /readonly is required you may specify /leaky/noupdate or
1688 /strict/noupdate which are treated the same as /readonly.
1690 A useful new feature is the /count= option. This is a generalization
1691 of the per_byte option, so that you can measure the throughput of other
1692 aggregate values. For example, the per_byte option is now equivalent
1693 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
1695 The per_rcpt option has been generalized using the /count= mechanism
1696 (though it's more complicated than the per_byte equivalence). When it is
1697 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
1698 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
1699 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
1700 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
1701 Note that using per_rcpt with a non-readonly update mode in more than
1702 one ACL will cause the recipients to be double-counted. (The per_mail
1703 and per_byte options don't have this problem.)
1705 The handling of very low rates has changed slightly. If the computed rate
1706 is less than the event's count (usually one) then this event is the first
1707 after a long gap. In this case the rate is set to the same as this event's
1708 count, so that the first message of a spam run is counted properly.
1710 The major new feature is a mechanism for counting the rate of unique
1711 events. The new per_addr option counts the number of different
1712 recipients that someone has sent messages to in the last time period. It
1713 behaves like per_rcpt if all the recipient addresses are different, but
1714 duplicate recipient addresses do not increase the measured rate. Like
1715 the /count= option this is a general mechanism, so the per_addr option
1716 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
1717 example, measure the rate that a client uses different sender addresses
1718 with the options per_mail/unique=$sender_address. There are further
1719 details in the main documentation.
1721 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
1723 TF/06 Removed a few PCRE remnants.
1725 TF/07 Automatically extract Exim's version number from tags in the git
1726 repository when doing development or release builds.
1728 PP/02 Raise smtp_cmd_buffer_size to 16kB.
1729 Bugzilla 879. Patch from Paul Fisher.
1731 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
1732 Heavily based on revision 40f9a89a from Simon Arlott's tree.
1735 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
1737 PP/05 Variable $av_failed, true if the AV scanner deferred.
1738 Bugzilla 1078. Patch from John Horne.
1740 PP/06 Stop make process more reliably on build failure.
1741 Bugzilla 1087. Patch from Heiko Schlittermann.
1743 PP/07 Make maildir_use_size_file an _expandable_ boolean.
1744 Bugzilla 1089. Patch from Heiko Schlittermann.
1746 PP/08 Handle ${run} returning more data than OS pipe buffer size.
1747 Bugzilla 1131. Patch from Holger Weiß.
1749 PP/09 Handle IPv6 addresses with SPF.
1750 Bugzilla 860. Patch from Wolfgang Breyha.
1752 PP/10 GnuTLS: support TLS 1.2 & 1.1.
1754 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
1757 PP/11 match_* no longer expand right-hand-side by default.
1758 New compile-time build option, EXPAND_LISTMATCH_RHS.
1759 New expansion conditions, "inlist", "inlisti".
1761 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
1763 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
1765 PP/14 fix log_write() format string regression from TF/03.
1766 Bugzilla 1152. Patch from Dmitry Isaikin.
1772 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
1774 PP/02 Harmonised TLS library version reporting; only show if debugging.
1775 Layout now matches that introduced for other libraries in 4.74 PP/03.
1777 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
1779 PP/04 New "dns_use_edns0" global option.
1781 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
1784 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
1785 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
1787 TK/01 Updated PolarSSL code to 0.14.2.
1788 Bugzilla 1097. Patch from Andreas Metzler.
1790 PP/07 Catch divide-by-zero in ${eval:...}.
1791 Fixes bugzilla 1102.
1793 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
1796 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
1797 format-string attack -- SECURITY: remote arbitrary code execution.
1799 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
1800 time unintentionally subject to list matching rules, letting the header
1801 cause arbitrary Exim lookups (of items which can occur in lists, *not*
1802 arbitrary string expansion). This allowed for information disclosure.
1804 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
1805 INT_MIN/-1 -- value coerced to INT_MAX.
1811 NM/01 Workaround for PCRE version dependency in version reporting
1814 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
1815 This fixes portability to compilers other than gcc, notably
1816 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
1818 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
1819 makefiles for portability to HP-UX and POSIX correctness.
1821 PP/01 Permit LOOKUP_foo enabling on the make command-line.
1822 Also via indented variable definition in the Makefile.
1823 (Debugging by Oliver Heesakkers).
1825 PP/02 Restore caching of spamd results with expanded spamd_address.
1826 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
1828 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
1829 Improves build reliability. Fix from: Frank Elsner
1831 NM/02 Fix wide character breakage in the rfc2047 coding
1832 Fixes bug 1064. Patch from Andrey N. Oktyabrski
1834 NM/03 Allow underscore in dnslist lookups
1835 Fixes bug 1026. Patch from Graeme Fowler
1837 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
1838 Code patches from Adam Ciarcinski of NetBSD.
1840 NM/04 Fixed exiqgrep to cope with mailq missing size issue
1843 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
1844 is logged, to avoid truncation. Patch from John Horne.
1846 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
1847 Patch from Jakob Hirsch.
1849 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
1850 SQL string expansion failure details.
1851 Patch from Andrey Oktyabrski.
1853 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
1854 Patch from Simon Arlott.
1856 PP/09 New lookups functionality failed to compile on old gcc which rejects
1857 extern declarations in function scope.
1858 Patch from Oliver Fleischmann
1860 PP/10 Use sig_atomic_t for flags set from signal handlers.
1861 Check getgroups() return and improve debugging.
1862 Fixed developed for diagnosis in bug 927 (which turned out to be
1865 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
1866 Patch from Mark Zealey.
1868 PP/12 Bugzilla 1056: Improved spamd server selection.
1869 Patch from Mark Zealey.
1871 PP/13 Bugzilla 1086: Deal with maildir quota file races.
1872 Based on patch from Heiko Schlittermann.
1874 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
1875 Patch from Uwe Doering, sign-off by Michael Haardt.
1877 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
1878 variable declaration deep within a block. Bug and patch from
1881 PP/15 lookups-Makefile IRIX compatibility coercion.
1883 PP/16 Make DISABLE_DKIM build knob functional.
1885 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
1886 Patch by Simon Arlott
1888 TF/03 Fix valgrind.h portability to C89 compilers that do not support
1889 variable argument macros. Our copy now differs from upstream.
1895 TF/01 Failure to get a lock on a hints database can have serious
1896 consequences so log it to the panic log.
1898 TF/02 Log LMTP confirmation messages in the same way as SMTP,
1899 controlled using the smtp_confirmation log selector.
1901 TF/03 Include the error message when we fail to unlink a spool file.
1903 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
1904 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
1905 for maintaining out-of-tree patches for some time.
1907 PP/01 Bugzilla 139: Documentation and portability issues.
1908 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
1909 Handle per-OS dynamic-module compilation flags.
1911 PP/02 Let /dev/null have normal permissions.
1912 The 4.73 fixes were a little too stringent and complained about the
1913 permissions on /dev/null. Exempt it from some checks.
1914 Reported by Andreas M. Kirchwitz.
1916 PP/03 Report version information for many libraries, including
1917 Exim version information for dynamically loaded libraries. Created
1918 version.h, now support a version extension string for distributors
1919 who patch heavily. Dynamic module ABI change.
1921 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
1922 privilege escalation vulnerability whereby the Exim run-time user
1923 can cause root to append content of the attacker's choosing to
1926 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
1929 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
1930 If dropping privileges for untrusted macros, we disabled normal logging
1931 on the basis that it would fail; for the Exim run-time user, this is not
1932 the case, and it resulted in successful deliveries going unlogged.
1933 Fixed. Reported by Andreas Metzler.
1939 PP/01 Date: & Message-Id: revert to normally being appended to a message,
1940 only prepend for the Resent-* case. Fixes regression introduced in
1941 Exim 4.70 by NM/22 for Bugzilla 607.
1943 PP/02 Include check_rfc2047_length in configure.default because we're seeing
1944 increasing numbers of administrators be bitten by this.
1946 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
1948 PP/03 Bugzilla 994: added openssl_options main configuration option.
1950 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
1952 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
1954 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
1956 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
1957 without a peer certificate, leading to a segfault because of an
1958 assumption that peers always have certificates. Be a little more
1959 paranoid. Problem reported by Martin Tscholak.
1961 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
1962 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
1963 NB: ClamAV planning to remove STREAM in "middle of 2010".
1964 CL also introduces -bmalware, various -d+acl logging additions and
1965 more caution in buffer sizes.
1967 PP/09 Implemented reverse_ip expansion operator.
1969 PP/10 Bugzilla 937: provide a "debug" ACL control.
1971 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
1973 PP/12 Bugzilla 973: Implement --version.
1975 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
1977 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
1979 PP/15 Bugzilla 816: support multiple condition rules on Routers.
1981 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
1982 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
1983 ignore trailing whitespace.
1985 JJ/02 prevent non-panic DKIM error from being sent to paniclog
1987 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
1990 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
1991 Notification from Dr Andrew Aitchison.
1993 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
1994 ExtendedDetectionInfo response format.
1995 Notification from John Horne.
1997 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
2000 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
2001 XSL and documented dependency on system catalogs, with examples of how
2004 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
2007 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
2008 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
2009 configuration file which is writeable by the Exim user or group.
2011 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
2012 of configuration files to cover files specified with the -C option if
2013 they are going to be used with root privileges, not just the default
2016 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
2017 option (effectively making it always true).
2019 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
2020 files to be used while preserving root privileges.
2022 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
2023 that rogue child processes cannot use them.
2025 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
2026 run-time user, instead of root.
2028 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
2029 Exim run-time user without dropping privileges.
2031 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
2032 result string, instead of calling string_vformat() twice with the same
2035 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
2036 for other users. Others should always drop root privileges if they use
2037 -C on the command line, even for a whitelisted configure file.
2039 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
2041 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
2047 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
2048 $data_path, and $header_path variables; fixed documentation bugs and
2051 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
2052 exipick to access non-standard spools, including the "frozen" queue
2055 NM/01 Bugzilla 965: Support mysql stored procedures.
2056 Patch from Alain Williams
2058 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
2060 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
2061 Patch from Andreas Metzler
2063 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
2064 Patch from Kirill Miazine
2066 NM/05 Bugzilla 671: Added umask to procmail example.
2068 JJ/03 installed exipick 20100323.0, fixing doc bug
2070 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
2071 directory. Notification and patch from Dan Rosenberg.
2073 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
2075 TK/02 Improve log output when DKIM signing operation fails.
2077 MH/01 Treat the transport option dkim_domain as a colon separated
2078 list, not as a single string, and sign the message with each element,
2079 omitting multiple occurences of the same signer.
2081 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
2082 Bugzilla 985, 986. Patch by Simon Arlott
2084 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
2085 Patch by Simon Arlott
2087 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
2088 MBX locking. Notification from Dan Rosenberg.
2094 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
2096 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
2098 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
2100 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
2102 NM/04 Bugzilla 915: Flag broken perl installation during build.
2108 TK/01 Added patch by Johannes Berg that expands the main option
2109 "spamd_address" if it starts with a dollar sign.
2111 TK/02 Write list of recipients to X-Envelope-Sender header when building
2112 the mbox-format spool file for content scanning (suggested by Jakob
2115 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
2116 (http://www.dcc-servers.net/) support via dccifd. Activated by
2117 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
2119 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
2120 by Mark Daniel Reidel <mr@df.eu>.
2122 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
2123 When building exim an external PCRE library is now needed -
2124 PCRE is a system library on the majority of modern systems.
2125 See entry on PCRE_LIBS in EDITME file.
2127 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
2128 conversation. Added nologin parameter to request.
2129 Patch contributed by Kirill Miazine.
2131 TF/01 Do not log submission mode rewrites if they do not change the address.
2133 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
2135 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
2136 log files in place. Contributed by Roberto Lima.
2138 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
2140 TF/03 Bugzilla 615: When checking the local_parts router precondition
2141 after a local_part_suffix or local_part_prefix option, Exim now
2142 does not use the address's named list lookup cache, since this
2143 contains cached lookups for the whole local part.
2145 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
2146 Robert Millan. Documentation is in experimental-spec.txt.
2148 TF/04 Bugzilla 668: Fix parallel build (make -j).
2150 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
2152 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
2153 Patch provided by Jan Srzednicki.
2155 TF/05 Leading white space used to be stripped from $spam_report which
2156 wrecked the formatting. Now it is preserved.
2158 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
2159 that they are available at delivery time.
2161 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
2163 TF/08 TLS error reporting now respects the incoming_interface and
2164 incoming_port log selectors.
2166 TF/09 Produce a more useful error message if an SMTP transport's hosts
2167 setting expands to an empty string.
2169 NM/06 Bugzilla 744: EXPN did not work under TLS.
2170 Patch provided by Phil Pennock.
2172 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
2173 Patch provided by Richard Godbee.
2175 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
2176 acl_smtp_notquit, added index entry.
2178 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
2179 Patch provided by Eugene Bujak.
2181 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
2182 accept(). Patch provided by Maxim Dounin.
2184 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
2185 Patch provided by Phil Pennock.
2187 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
2189 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
2190 Patch provided by Brad "anomie" Jorsch.
2192 NM/14 Bugzilla 622: Added timeout setting to transport filter.
2193 Patch provided by Dean Brooks.
2195 TK/05 Add native DKIM support (does not depend on external libraries).
2197 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
2198 Patch provided by Graeme Fowler.
2200 NM/16 Bugzilla 851: Documentation example syntax fix.
2202 NM/17 Changed NOTICE file to remove references to embedded PCRE.
2204 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
2207 NM/19 Bugzilla 745: TLS version reporting.
2208 Patch provided by Phil Pennock.
2210 NM/20 Bugzilla 167: bool: condition support.
2211 Patch provided by Phil Pennock.
2213 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
2214 clients. Patch provided by Phil Pennock.
2216 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
2217 Patch provided by Brad "anomie" Jorsch.
2219 NM/23 Bugzilla 687: Fix misparses in eximstats.
2220 Patch provided by Heiko Schlittermann.
2222 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
2223 Patch provided by Heiko Schlittermann.
2225 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
2226 plus update to original patch.
2228 NM/26 Bugzilla 799: Documentation correction for ratelimit.
2230 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
2231 Patch provided by David Brownlee.
2233 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
2235 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
2237 NM/30 Bugzilla 888: TLS documentation bugfixes.
2239 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
2241 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
2242 Unlike the original bugzilla I have changed all shell scripts in src tree.
2244 NM/33 Bugzilla 898: Transport filter timeout fix.
2245 Patch by Todd Rinaldo.
2247 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
2248 Patch by Serge Demonchaux.
2250 NM/35 Bugzilla 39: Base64 decode bug fixes.
2251 Patch by Jakob Hirsch.
2253 NM/36 Bugzilla 909: Correct connect() call in dcc code.
2255 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
2257 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
2259 NM/39 Bugzilla 911: Fixed MakeLinks build script.
2265 TK/01 Add preliminary DKIM support. Currently requires a forked version of
2266 ALT-N's libdkim that I have put here:
2267 http://duncanthrax.net/exim-experimental/
2269 Note to Michael Haardt: I had to rename some vars in sieve.c. They
2270 were called 'true' and it seems that C99 defines that as a reserved
2271 keyword to be used with 'bool' variable types. That means you could
2272 not include C99-style headers which use bools without triggering
2273 build errors in sieve.c.
2275 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
2276 as mailq or other aliases. Changed the --help handling significantly
2277 to do whats expected. exim_usage() emits usage/help information.
2279 SC/01 Added the -bylocaldomain option to eximstats.
2281 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
2283 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
2285 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
2291 PH/01 Another patch from the Sieve maintainer.
2293 PH/02 When an IPv6 address is converted to a string for single-key lookup
2294 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
2295 dots are used instead of colons so that keys in lsearch files need not
2296 contain colons. This was done some time before quoting was made available
2297 in lsearch files. However, iplsearch files do require colons in IPv6 keys
2298 (notated using the quote facility) so as to distinguish them from IPv4
2299 keys. This meant that lookups for IP addresses in host lists did not work
2300 for iplsearch lookups.
2302 This has been fixed by arranging for IPv6 addresses to be expressed with
2303 colons if the lookup type is iplsearch. This is not incompatible, because
2304 previously such lookups could never work.
2306 The situation is now rather anomalous, since one *can* have colons in
2307 ordinary lsearch keys. However, making the change in all cases is
2308 incompatible and would probably break a number of configurations.
2310 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
2313 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
2314 conversion specification without a maximum field width, thereby enabling
2315 a rogue spamd server to cause a buffer overflow. While nobody in their
2316 right mind would setup Exim to query an untrusted spamd server, an
2317 attacker that gains access to a server running spamd could potentially
2318 exploit this vulnerability to run arbitrary code as the Exim user.
2320 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
2321 $primary_hostname instead of what libspf2 thinks the hosts name is.
2323 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
2324 a directory entry by the name of the lookup key. Previously, if a
2325 symlink pointed to a non-existing file or a file in a directory that
2326 Exim lacked permissions to read, a lookup for a key matching that
2327 symlink would fail. Now it is enough that a matching directory entry
2328 exists, symlink or not. (Bugzilla 503.)
2330 PH/03 The body_linecount and body_zerocount variables are now exported in the
2333 PH/04 Added the $dnslist_matched variable.
2335 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
2336 This means they are set thereafter only if the connection becomes
2339 PH/06 Added the client_condition to authenticators so that some can be skipped
2340 by clients under certain conditions.
2342 PH/07 The error message for a badly-placed control=no_multiline_responses left
2343 "_responses" off the end of the name.
2345 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
2347 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
2348 (without spaces) instead of just copying the configuration text.
2350 PH/10 Added the /noupdate option to the ratelimit ACL condition.
2352 PH/11 Added $max_received_linelength.
2354 PH/12 Added +ignore_defer and +include_defer to host lists.
2356 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
2357 way in which PCRE > 7.0 is built.
2359 PH/14 Implemented queue_only_load_latch.
2361 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
2362 MAIL command. The effect was to mangle the value on 64-bit systems.
2364 PH/16 Another patch from the Sieve maintainer.
2366 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
2368 PH/18 If a system quota error occurred while trying to create the file for
2369 a maildir delivery, the message "Mailbox is full" was not appended to the
2370 bounce if the delivery eventually timed out. Change 4.67/27 below applied
2371 only to a quota excession during the actual writing of the file.
2373 PH/19 It seems that peer DN values may contain newlines (and other non-printing
2374 characters?) which causes problems in log lines. The DN values are now
2375 passed through string_printing() before being added to log lines.
2377 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
2378 and InterBase are left for another time.)
2380 PH/21 Added message_body_newlines option.
2382 PH/22 Guard against possible overflow in moan_check_errorcopy().
2384 PH/23 POSIX allows open() to be a macro; guard against that.
2386 PH/24 If the recipient of an error message contained an @ in the local part
2387 (suitably quoted, of course), incorrect values were put in $domain and
2388 $local_part during the evaluation of errors_copy.
2394 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
2395 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
2398 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
2399 issue a MAIL command.
2401 PH/02 In an ACL statement such as
2403 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
2405 if a client was not listed at all, or was listed with a value other than
2406 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
2407 the condition was not true (as it should be), so access was not denied.
2408 The bug was that the ! inversion was incorrectly passed on to the second
2409 item. This has been fixed.
2411 PH/03 Added additional dnslists conditions == and =& which are different from
2412 = and & when the dns lookup returns more than one IP address.
2414 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
2415 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
2417 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
2418 FSYNC, which compiles an option called disable_fsync that allows for
2419 bypassing fsync(). The documentation is heavily laced with warnings.
2421 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
2423 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
2424 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
2425 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
2426 including adding "make clean"; (3) Added -fPIC when compiling the test
2427 dynamically loaded module, to get rid of a warning.
2429 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
2430 message fails, move_frozen_messages = true and ignore_bounce_errors_after
2431 = 0s. The bug is otherwise harmless.
2433 PH/07 There was a bug in the dovecot authenticator such that the value of
2434 $auth1 could be overwritten, and so not correctly preserved, after a
2435 successful authentication. This usually meant that the value preserved by
2436 the server_setid option was incorrect.
2438 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
2440 PH/09 Installed PCRE release 7.0.
2442 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
2443 run for batched SMTP input. It is now run at the start of every message
2444 in the batch. While fixing this I discovered that the process information
2445 (output by running exiwhat) was not always getting set for -bs and -bS
2446 input. This is fixed, and it now also says "batched" for BSMTP.
2448 PH/11 Added control=no_pipelining.
2450 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
2451 patch, slightly modified), and move the expansion of helo_data till after
2452 the connection is made in the smtp transport (so it can use these
2455 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
2457 PH/14 Added log_selector = +pid.
2459 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
2461 PH/16 Add ${if forany and ${if forall.
2463 PH/17 Added dsn_from option to vary the From: line in DSNs.
2465 PH/18 Flush SMTP output before performing a callout, unless control =
2466 no_callout_flush is set.
2468 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
2469 was true (the default) a successful delivery failed to delete the retry
2470 item, thus causing premature timeout of the address. The bug is now
2473 PH/20 Added hosts_avoid_pipelining to the smtp transport.
2475 PH/21 Long custom messages for fakedefer and fakereject are now split up
2476 into multiline responses in the same way that messages for "deny" and
2477 other ACL rejections are.
2479 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
2480 with slight modification.
2482 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
2483 draft, changing the syntax and factoring some duplicate code".
2485 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
2486 for deliveries of the second and subsequent messages over the same SMTP
2489 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
2490 ${reduce, with only minor "tidies".
2492 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
2494 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
2495 expansion side effects.
2497 PH/27 When a message times out after an over-quota error from an Exim-imposed
2498 quota, the bounce message says "mailbox is full". This message was not
2499 being given when it was a system quota that was exceeded. It now should
2502 MH/03 Made $recipients available in local_scan(). local_scan() already has
2503 better access to the recipient list through recipients_list[], but
2504 $recipients can be useful in postmaster-provided expansion strings.
2506 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
2507 in the case of a MAIL command with additional options following the
2508 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
2509 were accidentally chopped off.
2511 PH/29 SMTP synchronization checks are implemented when a command is read -
2512 there is a check that no more input is waiting when there shouldn't be
2513 any. However, for some commands, a delay in an ACL can mean that it is
2514 some time before the response is written. In this time, more input might
2515 arrive, invalidly. So now there are extra checks after an ACL has run for
2516 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
2517 pipelining has not been advertised.
2519 PH/30 MH's patch to allow iscntrl() characters to be list separators.
2521 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
2522 returned in the SMTP response when smtp_return_error_details was false.
2523 This has been fixed.
2525 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
2526 instead of the C I/O that was originally supplied, because problems were
2527 reported on Solaris.
2529 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
2530 Exim which did not show up earlier: it was assuming that a call to
2531 SSL_CTX_set_info_callback() might give an error value. In fact, there is
2532 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
2533 was a macro that became an assignment, so it seemed to work. This has
2534 changed to a proper function call with a void return, hence the compile
2535 error. Exim's code has been fixed.
2537 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
2540 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
2542 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
2544 PH/37 If a message is not accepted after it has had an id assigned (e.g.
2545 because it turns out to be too big or there is a timeout) there is no
2546 "Completed" line in the log. When some messages of this type were
2547 selected by exigrep, they were listed as "not completed". Others were
2548 picked up by some special patterns. I have improved the selection
2549 criteria to be more general.
2551 PH/38 The host_find_failed option in the manualroute router can now be set
2552 to "ignore", to completely ignore a host whose IP address cannot be
2553 found. If all hosts are ignored, the behaviour is controlled by the new
2554 host_all_ignored option.
2556 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
2557 homing or because of multiple MX records with /mx) generated more than
2558 one IP address, and the following item turned out to be the local host,
2559 all the secondary addresses of the first item were incorrectly removed
2560 from the list, along with the local host and any following hosts (which
2561 is what is supposed to happen).
2563 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
2564 whoever called Exim into the -H file. In the case of the daemon it was
2565 behaving confusingly. When first started, it used values for whoever
2566 started the daemon, but after a SIGHUP it used the Exim user (because it
2567 calls itself on a restart). I have changed the code so that it now always
2570 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
2571 message are rejected with the same error (e.g. no authentication or bad
2572 sender address), and a DATA command is nevertheless sent (as can happen
2573 with PIPELINING or a stupid MUA), the error message that was given to the
2574 RCPT commands is included in the rejection of the DATA command. This is
2575 intended to be helpful for MUAs that show only the final error to their
2578 PH/42 Another patch from the Sieve maintainer.
2580 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
2581 Eximstats - Fixed some broken HTML links and added missing column headers
2583 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
2584 columns for Rejects, Temp Rejects, Ham, and Spam rows.
2586 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
2588 PH/43 Yet another patch from the Sieve maintainer.
2590 PH/44 I found a way to check for a TCP/IP connection going away before sending
2591 the response to the final '.' that terminates a message, but only in the
2592 case where the client has not sent further data following the '.'
2593 (unfortunately, this is allowed). However, in many cases there won't be
2594 any further data because there won't be any more messages to send. A call
2595 to select() can be used: if it shows that the input is "ready", there is
2596 either input waiting, or the socket has been closed. An attempt to read
2597 the next input character can distinguish the two cases. Previously, Exim
2598 would have sent an OK response which the client would never have see.
2599 This could lead to message repetition. This fix should cure that, at
2600 least in a lot of common cases.
2602 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
2603 advertised in response to EHLO.
2609 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
2610 fixed by 4.65/MH/01 (is this a record?) are fixed:
2612 (i) An empty string was always treated as zero by the numeric comparison
2613 operators. This behaviour has been restored.
2615 (ii) It is documented that the numeric comparison operators always treat
2616 their arguments as decimal numbers. This was broken in that numbers
2617 starting with 0 were being interpreted as octal.
2619 While fixing these problems I realized that there was another issue that
2620 hadn't been noticed. Values of message_size_limit (both the global option
2621 and the transport option) were treated as octal if they started with 0.
2622 The documentation was vague. These values are now always treated as
2623 decimal, and I will make that clear in the documentation.
2629 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
2630 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
2633 MH/01 Don't check that the operands of numeric comparison operators are
2634 integers when their expansion is in "skipping" mode (fixes bug
2635 introduced by 4.64-PH/07).
2637 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
2638 child addresses, Exim now panics and dies. Previously, because the count
2639 is held in a short int, deliveries were likely to be lost. As such a
2640 large number of recipients for a single message is ridiculous
2641 (performance will be very, very poor), I have chosen to impose a limit
2642 rather than extend the field.
2648 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
2649 leftover -K file (the existence of which was triggered by #402).
2650 While we were at it, introduced process PID as part of the -K
2651 filename. This should rule out race conditions when creating
2654 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
2655 processing considerably. Previous code took too long for large mails,
2656 triggering a timeout which in turn triggers #401.
2658 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
2659 in the DK code in transports.c. sendfile() is not really portable,
2660 hence the _LINUX specificness.
2662 TF/01 In the add_headers option to the mail command in an Exim filter,
2663 there was a bug that Exim would claim a syntax error in any
2664 header after the first one which had an odd number of characters
2667 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
2668 callout verification, Exim cached a "reject" for the entire domain. This
2669 is correct for most verifications, but it is not correct for a recipient
2670 verification with use_sender or use_postmaster set, because in that case
2671 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
2672 case of MAIL FROM:<> rejection from other early rejections (e.g.
2673 rejection of HELO). When verifying a recipient using a non-null MAIL
2674 address, the cache is ignored if it shows MAIL FROM:<> rejection.
2675 Whatever the result of the callout, the value of the domain cache is
2676 left unchanged (for any other kind of callout, getting as far as trying
2677 RCPT means that the domain itself is ok).
2679 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
2682 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
2683 manifest itself as EPIPE rather than ECONNECT. When tidying away a
2684 session, the daemon ignores ECONNECT errors and logs others; it now
2685 ignores EPIPE as well.
2687 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
2688 (quoted-printable decoding).
2690 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
2691 later the small subsequent patch to fix an introduced bug.
2693 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
2695 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
2697 PH/08 An error is now given if message_size_limit is specified negative.
2699 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
2700 to be given (somewhat) arbitrary names.
2702 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
2705 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
2706 miscellaneous code fixes
2708 PH/10 Added the log_reject_target ACL modifier to specify where to log
2711 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
2712 hostname. This is wrong, because it relates to the incoming message (and
2713 probably the interface on which it is arriving) and not to the outgoing
2714 callout (which could be using a different interface). This has been
2715 changed to use the value of the helo_data option from the smtp transport
2716 instead - this is what is used when a message is actually being sent. If
2717 there is no remote transport (possible with a router that sets up host
2718 addresses), $smtp_active_hostname is used.
2720 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
2721 tweaks were necessary in order to get it to work (see also 21 below):
2722 (a) The code assumed that strncpy() returns a negative number on buffer
2723 overflow, which isn't the case. Replaced with Exim's string_format()
2725 (b) There were several signed/unsigned issues. I just did the minimum
2726 hacking in of casts. There is scope for a larger refactoring.
2727 (c) The code used strcasecmp() which is not a standard C function.
2728 Replaced with Exim's strcmpic() function.
2729 (d) The code set only $1; it now sets $auth1 as well.
2730 (e) A simple test gave the error "authentication client didn't specify
2731 service in request". It would seem that Dovecot has changed its
2732 interface. Fortunately there's a specification; I followed it and
2733 changed what the client sends and it appears to be working now.
2735 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
2738 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
2739 address A is aliased to B and C, where B exists and C does not. Without
2740 -v the output is "A verified" because verification stops after a
2741 successful redirection if more than one address is generated. However,
2742 with -v the child addresses are also verified. Exim was outputting "A
2743 failed to verify" and then showing the successful verification for C,
2744 with its parentage. It now outputs "B failed to verify", showing B's
2745 parentage before showing the successful verification of C.
2747 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
2748 look up a TXT record in a specific list after matching in a combined
2751 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
2752 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
2753 they consult the DNS. I had assumed they would set it the way they
2754 wanted; and indeed my experiments on Linux seem to show that in some
2755 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
2756 To be on the safe side, however, I have now made the interface to
2757 host_find_byname() similar to host_find_bydns(), with an argument
2758 containing the DNS resolver options. The host_find_byname() function now
2759 sets these options at its start, just as host_find_bydns() does. The smtp
2760 transport options dns_qualify_single and dns_search_parents are passed to
2761 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
2762 of host_find_byname() use the default settings of RES_DEFNAMES
2763 (qualify_single) but not RES_DNSRCH (search_parents).
2765 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
2766 spool_read_header() do less string testing, by means of a preliminary
2767 switch on the second character of optional "-foo" lines. (This is
2768 overdue, caused by the large number of possibilities that now exist.
2769 Originally there were few.) While I was there, I also converted the
2770 str(n)cmp tests so they don't re-test the leading "-" and the first
2771 character, in the hope this might squeeze out yet more improvement.
2773 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
2774 flag allowing group syntax was set by the header_syntax check but not
2775 turned off, possible causing trouble later; (2) The flag was not being
2776 set at all for the header_verify test, causing "group"-style headers to
2777 be rejected. I have now set it in this case, and also caused header_
2778 verify to ignore an empty address taken from a group. While doing this, I
2779 came across some other cases where the code for allowing group syntax
2780 while scanning a header line wasn't quite right (mostly, not resetting
2781 the flag correctly in the right place). These bugs could have caused
2782 trouble for malformed header lines. I hope it is now all correct.
2784 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
2785 with the "reply" argument non-NULL. The code, however (which originally
2786 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
2787 but it didn't always do it. This confused somebody who was copying the
2788 code for some other use. I have removed all the tests.
2790 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
2791 feature that was used to support insecure browsers during the U.S. crypto
2792 embargo. It requires special client support, and Exim is probably the
2793 only MTA that supported it -- and would never use it because real RSA is
2794 always available. This code has been removed, because it had the bad
2795 effect of slowing Exim down by computing (never used) parameters for the
2796 RSA_EXPORT functionality.
2798 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
2799 authenticator to fail if there's a tab character in the incoming data
2800 (there should never be unless someone is messing about, as it's supposed
2801 to be base64-encoded). Also added, on Timo's advice, the "secured" option
2802 if the connection is using TLS or if the remote IP is the same as the
2803 local IP, and the "valid-client-cert option" if a client certificate has
2806 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
2807 authenticators. This can be used for authorization after authentication
2808 succeeds. (In the case of plaintext, it servers for both authentication
2811 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
2812 if any retry times were supplied.
2814 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
2815 connection, where there is no client IP address to check. In this
2816 situation, the verify now always succeeds.
2818 PH/25 Applied John Jetmore's -Mset patch.
2820 PH/26 Added -bem to be like -Mset, but loading a message from a file.
2822 PH/27 In a string expansion for a processed (not raw) header when multiple
2823 headers of the same name were present, leading whitespace was being
2824 removed from all of them, but trailing whitespace was being removed only
2825 from the last one. Now trailing whitespace is removed from each header
2826 before concatenation. Completely empty headers in a concatenation (as
2827 before) are ignored.
2829 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
2830 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
2832 PH/29 [Removed. This was a change that I later backed out, and forgot to
2833 correct the ChangeLog entry (that I had efficiently created) before
2834 committing the later change.]
2836 PH/30 Exim was sometimes attempting to deliver messages that had suffered
2837 address errors (4xx response to RCPT) over the same connection as other
2838 messages routed to the same hosts. Such deliveries are always "forced",
2839 so retry times are not inspected. This resulted in far too many retries
2840 for the affected addresses. The effect occurred only when there were more
2841 hosts than the hosts_max_try setting in the smtp transport when it had
2842 the 4xx errors. Those hosts that it had tried were not added to the list
2843 of hosts for which the message was waiting, so if all were tried, there
2844 was no problem. Two fixes have been applied:
2846 (i) If there are any address or message errors in an SMTP delivery, none
2847 of the hosts (tried or untried) are now added to the list of hosts
2848 for which the message is waiting, so the message should not be a
2849 candidate for sending over the same connection that was used for a
2850 successful delivery of some other message. This seems entirely
2851 reasonable: after all the message is NOT "waiting for some host".
2852 This is so "obvious" that I'm not sure why it wasn't done
2853 previously. Hope I haven't missed anything, but it can't do any
2854 harm, as the worst effect is to miss an optimization.
2856 (ii) If, despite (i), such a delivery is accidentally attempted, the
2857 routing retry time is respected, so at least it doesn't keep
2858 hammering the server.
2860 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
2861 in ${readsocket because some servers need this prod.
2863 PH/32 Added some extra debug output when updating a wait-xxx database.
2865 PH/33 The hint "could be header name not terminated by colon", which has been
2866 given for certain expansion errors for a long time, was not being given
2867 for the ${if def:h_colon_omitted{... case.
2869 PH/34 The spec says: "With one important exception, whenever a domain list is
2870 being scanned, $domain contains the subject domain." There was at least
2871 one case where this was not true.
2873 PH/35 The error "getsockname() failed: connection reset by peer" was being
2874 written to the panic log as well as the main log, but it isn't really
2875 panic-worthy as it just means the connection died rather early on. I have
2876 removed the panic log writing for the ECONNRESET error when getsockname()
2879 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
2880 runs only) independently of the message's sender address. This meant
2881 that, if the 4xx error was in fact related to the sender, a different
2882 message to the same recipient with a different sender could confuse
2883 things. In particular, this can happen when sending to a greylisting
2884 server, but other circumstances could also provoke similar problems.
2885 I have changed the default so that the retry time for these errors is now
2886 based a combination of the sender and recipient addresses. This change
2887 can be overridden by setting address_retry_include_sender=false in the
2890 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
2891 remote server are returned as part of bounce messages. This was not
2892 happening for LMTP over a pipe (the lmtp transport), but now it is the
2893 same for both kinds of LMTP.
2895 PH/38 Despite being documented as not happening, Exim was rewriting addresses
2896 in header lines that were in fact CNAMEs. This is no longer the case.
2898 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
2899 and queue runs started by the daemon processed all messages. This has
2900 been fixed so that -R and -S can now usefully be given with -q<time>.
2902 PH/40 Import PCRE release 6.7 (fixes some bugs).
2904 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
2906 PH/42 Give an error if -q is specified more than once.
2908 PH/43 Renamed the variables $interface_address and $interface_port as
2909 $received_ip_address and $received_port, to make it clear that these
2910 values apply to message reception, and not to the outgoing interface when
2911 a message is delivered. (The old names remain recognized, of course.)
2913 PH/44 There was no timeout on the connect() call when using a Unix domain
2914 socket in the ${readsocket expansion. There now is.
2916 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
2917 be meaningful with "accept".
2919 SC/01 Eximstats V1.43
2920 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
2922 SC/02 Eximstats V1.44
2923 Use a glob alias rather than an array ref in the generated
2924 parser. This improves both readability and performance.
2926 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
2927 Collect SpamAssassin and rejection statistics.
2928 Don't display local sender or destination tables unless
2929 there is data to show.
2930 Added average volumes into the top table text output.
2932 SC/04 Eximstats V1.46
2933 Collect data on the number of addresses (recipients)
2934 as well as the number of messages.
2936 SC/05 Eximstats V1.47
2937 Added 'Message too big' to the list of mail rejection
2938 reasons (thanks to Marco Gaiarin).
2940 SC/06 Eximstats V1.48
2941 Mainlog lines which have GMT offsets and are too short to
2942 have a flag are now skipped.
2944 SC/07 Eximstats V1.49 (Alain Williams)
2945 Added the -emptyok flag.
2947 SC/08 Eximstats V1.50
2948 Fixes for obtaining the IP address from reject messages.
2950 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
2951 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
2952 whitespace changes from 4.64-PH/27
2954 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
2957 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
2958 are found, allow negative numbers in numeric criteria)
2960 JJ/06 exipick.20061117.2, added new $message_body_missing variable
2962 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
2963 to match changes made in 4.64-PH/43
2965 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
2967 PH/47 Put in an explicit test for a DNS lookup of an address record where the
2968 "domain" is actually an IP address, and force a failure. This locks out
2969 those revolvers/nameservers that support "A-for-A" lookups, in
2970 contravention of the specifications.
2972 PH/48 When a host name was looked up from an IP address, and the subsequent
2973 forward lookup of the name timed out, the host name was left in
2974 $sender_host_name, contrary to the specification.
2976 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
2977 restricted to single-key lookups, Exim was not diagnosing an error if
2978 * or *@ was used with a query-style lookup.
2980 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
2982 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
2983 long ago, but noone interested enough thought of it. Let's just say that
2984 the "1.1" means that there are some new functions that weren't there at
2985 some point in the past.
2987 PH/51 Error processing for expansion failure of helo_data from an smtp
2988 transport during callout processing was broken.
2990 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
2991 tested/used via the -bh/-bhc/-bs options.
2993 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
2994 bug, fixed in subsequent PCRE releases).
2996 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
2997 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
2999 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
3005 SC/01 Use a glob alias rather than an array ref in eximstats generated
3006 parser. This improves both readability and performance.
3008 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
3009 Don't display local sender or destination tables in eximstats unless
3010 there is data to show.
3011 Added average volumes into the eximstats top table text output.
3013 SC/03 Collect data on the number of addresses (recipients) as well
3014 as the number of messages in eximstats.
3016 TF/01 Correct an error in the documentation for the redirect router. Exim
3017 does (usually) call initgroups() when daemonizing.
3019 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
3020 with consistent privilege compared to when running as a daemon.
3022 TF/03 Note in the spec that $authenticated_id is not set for local
3023 submissions from trusted users.
3025 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
3026 Thanks to Dean Brooks <dean@iglou.com> for the patch.
3028 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
3029 by adding some example configuration directives to the default
3030 configuration file. A little bit of work is required to uncomment the
3031 directives and define how usernames and passwords are checked, but
3032 there is now a framework to start from.
3034 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
3035 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
3036 without this. I don't know how relevant this is to other LDAP libraries.
3038 PH/02 Add the verb name to the "unknown ACL verb" error.
3040 PH/03 Magnus Holmgren's patch for filter_prepend_home.
3042 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
3044 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
3045 directory not expanded when it should be if an expanded home directory
3046 was set for the address (which is overridden by the transport).
3048 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
3051 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
3052 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
3053 because it is too late at that time, and has no effect.
3055 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
3056 security issue with \' (bugzilla #107). I could not use the
3057 PQescapeStringConn() function, because it needs a PGconn value as one of
3060 PH/08 When testing addresses using -bt, indicate those final addresses that
3061 are duplicates that would not cause an additional delivery. At least one
3062 person was confused, thinking that -bt output corresponded to deliveries.
3063 (Suppressing duplicates isn't a good idea as you lose the information
3064 about possibly different redirections that led to the duplicates.)
3066 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
3067 systems where poll() doesn't work, in particular OS X.
3069 PH/10 Added more information to debugging output for retry time not reached.
3071 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
3072 operations in malware.c.
3074 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
3077 PH/13 If write_rejectlog was set false when logging was sent to syslog with
3078 syslog_duplication set false, log lines that would normally be written
3079 both the the main log and to the reject log were not written to syslog at
3082 PH/14 In the default configuration, change the use of "message" in ACL warn
3083 statements to "add_header".
3085 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
3086 not followed by a command (e.g. "seen endif").
3088 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
3089 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
3092 PH/17 Added extra conditions to the default value of delay_warning_condition
3096 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
3097 { match{$h_precedence:}{(?i)bulk|list|junk} } \
3098 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
3101 The Auto-Submitted: and various List- headers are standardised, whereas I
3102 don't think Precedence: ever was.
3104 PH/18 Refactored debugging code in route_finduser() to show more information,
3105 in particular, the error code if getpwnam() issues one.
3107 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
3108 This is apparently needed in addition to the PH/07 change above to avoid
3109 any possible encoding problems.
3111 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
3112 but not after initializing Perl.
3114 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
3115 output them only if debugging. By default they are written stderr,
3116 apparently, which is not desirable.
3118 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
3121 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
3124 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
3126 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
3127 authenticated or an ident call has been made. Suppress the default
3128 values for $authenticated_id and $authenticated_sender (but permit -oMai
3129 and -oMas) when testing with -bh.
3131 PH/24 Re-jigged the order of the tests in the default configuration so that the
3132 tests for valid domains and recipients precede the DNS black list and CSA
3133 tests, on the grounds that those ones are more expensive.
3135 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
3136 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
3137 This bug exists in every version of Exim that I still have, right back to
3140 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
3141 However, an attempt to turn on case-sensitivity in a regex key by
3142 including (?-i) didn't work because the subject string was already
3143 lowercased, and the effects were non-intuitive. It turns out that a
3144 one-line patch can be used to allow (?-i) to work as expected.
3150 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
3151 other effects) broke the use of negated acl sub-conditions.
3153 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
3156 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
3157 "Deny" causes Exim to reject the incoming connection with a 554 error.
3158 Unfortunately, if there is a major crisis, such as a disk failure,
3159 tcp-wrappers gives "deny", whereas what one would like would be some
3160 kind of temporary error. A kludge has been added to help with this.
3161 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
3162 554 error is used if errno is still zero or contains ENOENT (which occurs
3163 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
3166 PH/03 Add -lutil to the default FreeBSD LIBS setting.
3168 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
3169 errors. Otherwise a message that provokes a temporary error (when other
3170 messages do not) can cause a whole host to time out.
3172 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
3173 addresses were routed directly to files or pipes from a redirect router.
3174 File deliveries just didn't batch; pipe deliveries might have suffered
3177 PH/06 A failure to get a lock for a hints database would erroneously always say
3178 "Failed to get write lock", even when it was really a read lock.
3180 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
3181 of 0600. This has been changed to use the value of the lockfile_mode
3182 option (which defaults to 0600).
3184 PH/08 Applied small patch from the Sieve maintainer.
3186 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
3187 folder from quota calculations, a direct delivery into this folder messed
3188 up the contents of the maildirsize file. This was because the regex was
3189 used only to exclude .Trash (or whatever) when the size of the mailbox
3190 was calculated. There was no check that a delivery was happening into an
3191 excluded directory. This bug has been fixed by ignoring all quota
3192 processing for deliveries into excluded directories.
3194 PH/10 Added the maildirfolder_create_regex option to appendfile.
3200 PH/01 The code for finding all the local interface addresses on a FreeBSD
3201 system running IPv6 was broken. This may well have applied to all BSD
3202 systems, as well as to others that have similar system calls. The broken
3203 code found IPv4 interfaces correctly, but gave incorrect values for the
3204 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
3205 that it would not match correctly against @[] and not recognize the IPv6
3208 PH/02 The ipliteral router was not recognizing addresses of the form user@
3209 [ipv6:....] because it didn't know about the "ipv6:" prefix.
3211 PH/03 Added disable_ipv6.
3213 PH/04 Changed $reply_address to use the raw form of the headers instead of the
3214 decoded form, because it is most often used to construct To: headers
3215 lines in autoreplies, and the decoded form may well be syntactically
3216 invalid. However, $reply_address has leading white space removed, and all
3217 newlines turned into spaces so that the autoreply transport does not
3220 PH/05 If group was specified without a user on a router, and no group or user
3221 was specified on a transport, the group from the router was ignored.
3223 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
3224 for visible compile-time settings that can be used to change these
3225 numbers, for those that want even more. Backwards compatibility with old
3226 spool files has been maintained. However, going back to a previous Exim
3227 release will lost any variables that are in spool files.
3229 PH/07 Two small changes when running in the test harness: increase delay when
3230 passing a TCP/IP connection to a new process, in case the original
3231 process has to generate a bounce, and remove special handling of
3232 127.0.0.2 (sic), which is no longer necessary.
3234 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
3235 be the same on different OS.
3237 PH/09 Moved a debug statement in filter processing to avoid a race problem when
3240 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
3241 whether --show-vars was specified or not
3243 JJ/02 exipick: Added support for new ACL variable spool format introduced
3246 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
3247 syntactically invalid From: or Reply-to: line, and a filter used this to
3248 generate an autoreply, and therefore failed to obtain an address for the
3249 autoreply, Exim could try to deliver to a non-existent relative file
3250 name, causing unrelated and misleading errors. What now happens is that
3251 it logs this as a hard delivery error, but does not attempt to create a
3254 PH/11 The exinext utility has a -C option for testing purposes, but although
3255 the given file was scanned by exinext itself; it wasn't being passed on
3256 when Exim was called.
3258 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
3259 an end-of-file indication when reading a command response.
3261 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
3262 compiled. In many other places in Exim, IPv6 addresses are always
3263 recognized, so I have changed this. It also means that IPv4 domain
3264 literals of the form [IPV4:n.n.n.n] are now always recognized.
3266 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
3267 used if the router is not running as root, for example, when verifying at
3268 ACL time, or when using -bh. The debugging output from this situation was
3269 non-existent - all you got was a failure to exec. I have made two
3272 (a) Failures to set uid/gid, the current directory, or a process leader
3273 in a subprocess such as that created by queryprogram now generate
3274 suitable debugging output when -d is set.
3276 (b) The queryprogram router detects when it is not running as root,
3277 outputs suitable debugging information if -d is set, and then runs
3278 the subprocess without attempting to change uid/gid.
3280 PH/15 Minor change to Makefile for building test_host (undocumented testing
3283 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
3284 additional section of a DNS packet that returns MX or SRV records.
3285 Instead, it always explicitly searches for A/AAAA records. This avoids
3286 major problems that occur when a DNS server includes only records of one
3287 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
3288 fixed another bug: if SRV records were looked up and the corresponding
3289 address records were *not* found in the additional section, the port
3290 values from the SRV records were lost.
3292 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
3293 using the correct key (the original address) when searching the retry
3294 rules in order to find which one to use for generating the retry hint.
3296 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
3297 adding the default one. Similarly, if it contains a Reply-To: header, the
3298 errors_reply_to option, if set, is not used.
3300 PH/19 When calculating a retry time, Exim used to measure the "time since
3301 failure" by looking at the "first failed" field in the retry record. Now
3302 it does not use this if it is later than than the arrival time of the
3303 message. Instead it uses the arrival time. This makes for better
3304 behaviour in cases where some deliveries succeed, thus re-setting the
3305 "first failed" field. An example is a quota failure for a huge message
3306 when small messages continue to be delivered. Without this change, the
3307 "time since failure" will always be short, possible causing more frequent
3308 delivery attempts for the huge message than are intended.
3309 [Note: This change was subsequently modified - see PH/04 for 4.62.]
3311 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
3312 $1, $2, $3) because the numerical variables can be reset during some
3313 expansion items (e.g. "match"), thereby losing the authentication data.
3315 PH/21 Make -bV show the size of off_t variables so that the test suite can
3316 decide whether to run tests for quotas > 2G.
3318 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
3319 mailbox_size, and mailbox_filecount in the appendfile transport. If a
3320 filecount value is greater than 2G or if a quota value is greater than 2G
3321 on a system where the size of off_t is not greater than 4, a panic error
3324 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
3325 never match. The debug and -bh output now contains an explicit error
3326 message indicating a malformed IPv4 address or mask.
3328 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
3329 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
3330 PH/23 above applies.
3332 PH/25 Do not write to syslog when running in the test harness. The only
3333 occasion when this arises is a failure to open the main or panic logs
3334 (for which there is an explicit test).
3336 PH/26 Added the /no_tell option to "control=freeze".
3338 PH/27 If a host name lookup failed very early in a connection, for example, if
3339 the IP address matched host_lookup and the reverse lookup yielded a name
3340 that did not have a forward lookup, an error message of the form "no IP
3341 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
3342 could be logged. Now it outputs the IP address instead of "NULL".
3344 PH/28 An enabling patch from MH: add new function child_open_exim2() which
3345 allows the sender and the authenticated sender to be set when
3346 submitting a message from within Exim. Since child_open_exim() is
3347 documented for local_scan(), the new function should be too.
3349 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
3350 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
3351 results in an empty string is now treated as unset.
3353 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
3355 PH/31 Added sender_verify_fail logging option.
3357 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
3358 needed by RFC 822 but not by RFC 2822 was commented out. I have now
3359 tidied the source and removed it altogether.
3361 PH/33 When a queue run was abandoned because the load average was too high, a
3362 log line was always written; now it is written only if the queue_run log
3363 selector is set. In addition, the log line for abandonment now contains
3364 information about the queue run such as the pid. This is always present
3365 in "start" and "stop" lines but was omitted from the "abandon" line.
3367 PH/34 Omit spaces between a header name and the colon in the error message that
3368 is given when verify = headers_syntax fails (if there are lots of them,
3369 the message gets confusing).
3371 PH/35 Change the default for dns_check_names_pattern to allow slashes within
3372 names, as there are now some PTR records that contain slashes. This check
3373 is only to protect against broken name servers that fall over on strange
3374 characters, so the fact that it applies to all lookups doesn't matter.
3376 PH/36 Now that the new test suite is complete, we can remove some of the
3377 special code in Exim that was needed for the old test suite. For example,
3378 sorting DNS records because real resolvers return them in an arbitrary
3379 order. The new test suite's fake resolver always returns records in the
3382 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
3383 bounces) except when queue_only is set, to avoid logging races between
3384 the different processes.
3386 PH/38 Panic-die if .include specifies a non-absolute path.
3388 PH/39 A tweak to the "H" retry rule from its user.
3390 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
3391 a label. They prevented compilation on older perls.
3393 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
3394 a warning to be raised on newish perls.
3396 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
3397 on queue. Changes to match documented behaviour of showing count of
3398 messages matching specified criteria.
3400 PH/40 Changed the default ident timeout from 30s to 5s.
3402 PH/41 Added support for the use of login_cap features, on those BSD systems
3403 that have them, for controlling the resources used by pipe deliveries.
3405 PH/42 The content-scanning code uses fopen() to create files in which to put
3406 message data. Previously it was not paying any attention to the mode of
3407 the files. Exim runs with umask(0) because the rest of the code creates
3408 files with open(), and sets the required mode explicitly. Thus, these
3409 files were ending up world-writeable. This was not a big issue, because,
3410 being within the spool directory, they were not world-accessible. I have
3411 created a function called modefopen, which takes an additional mode
3412 argument. It sets umask(777), creates the file, chmods it to the required
3413 mode, then resets the umask. All the relevant calls to fopen() in the
3414 content scanning code have been changed to use this function.
3416 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
3417 to 24 hours. This avoids potential overflow problems when processing G
3418 and H retry rules. I suspect nobody ever tinkers with this value.
3420 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
3422 PH/45 When the plaintext authenticator is running as a client, the server's
3423 challenges are checked to ensure they are valid base64 strings. By
3424 default, the authentication attempt is cancelled if an invalid string is
3425 received. Setting client_ignore_invalid_base64 true ignores these errors.
3426 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
3427 they are received. Thus, the responses can be made to depend on the
3428 challenges. If an invalid string is ignored, an empty string is placed in
3431 PH/46 Messages that are created by the autoreply transport now contains a
3432 References: header, in accordance with RFCs 2822 and 3834.
3434 PH/47 Added authenticated_sender_force to the smtp transport.
3436 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
3438 PH/49 Installed latest patch from the Sieve maintainer.
3440 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
3441 was also set, the appendfile transport was unnecessarily scanning a
3442 directory of message files (e.g. for maildir delivery) to find the count
3443 of files (along with the size), even though it did not need this
3444 information. It now does the scan only if it needs to find either the
3445 size of the count of files.
3447 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
3449 PH/52 Two bugs concerned with error handling when the smtp transport is
3452 (i) Exim was not creating retry information for temporary errors given
3453 for individual recipients after the DATA command when the smtp transport
3454 was used in LMTP mode. This meant that they could be retried too
3455 frequently, and not timed out correctly.
3457 (ii) Exim was setting the flag that allows error details to be returned
3458 for LMTP errors on RCPT commands, but not for LMTP errors for individual
3459 recipients that were returned after the DATA command.
3461 PH/53 This is related to PH/52, but is more general: for any failing address,
3462 when detailed error information was permitted to be returned to the
3463 sender, but the error was temporary, then after the final timeout, only
3464 "retry timeout exceeded" was returned. Now it returns the full error as
3465 well as "retry timeout exceeded".
3467 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
3468 do this, and (what is worse) MTAs that accept it.
3470 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
3471 will now be deprecated.
3473 PH/56 New os.c-cygwin from the Cygwin maintainer.
3475 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
3476 formats (previously only available in exim formats via -bpr, -bpru,
3477 and -bpra. Now also available in native and exiqgrep formats)
3479 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
3480 with very large, slow to parse queues
3482 JJ/08 exipick: added ! as generic prefix to negate any criteria format
3484 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
3486 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
3487 responses to authentication challenges, though it was showing the
3488 challenges; (ii) I've removed the CR characters from the debug output for
3491 PH/58 Allow for the insertion of a newline as well as a space when a string
3492 is turned into more than one encoded-word during RFC 2047 encoding. The
3493 Sieve code now uses this.
3495 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
3496 data_4xx, lost_connection, tls_required.
3498 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
3499 message was being sent as an SMTP response.
3501 PH/61 Add -l and -k options to exicyclog.
3503 PH/62 When verifying, if an address was redirected to one new address, so that
3504 verification continued, and the new address failed or deferred after
3505 having set something in $address_data, the value of $address_data was not
3506 passed back to the ACL. This was different to the case when no
3507 redirection occurred. The value is now passed back in both cases.
3509 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
3510 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
3511 use login_cap.h, so on its own it isn't the distinguishing feature. The
3512 new name refers directly to the setclassresources() function.
3514 PH/65 Added configuration files for NetBSD3.
3516 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
3518 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
3519 is preferred over IPv4.
3521 PH/68 The bounce_return_message and bounce_return_body options were not being
3522 honoured for bounces generated during the reception of non-SMTP messages.
3523 In particular, this applied to messages rejected by the ACL. This bug has
3524 been fixed. However, if bounce_return_message is true and bounce_return_
3525 body is false, the headers that are returned for a non-SMTP message
3526 include only those that have been read before the error was detected.
3527 (In the case of an ACL rejection, they have all been read.)
3529 PH/69 The HTML version of the specification is now built in a directory called
3530 spec_html instead of spec.html, because the latter looks like a path with
3531 a MIME-type, and this confuses some software.
3533 PH/70 Catch two compiler warnings in sieve.c.
3535 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
3536 function verify_get_ident() calls ip_connect() to connect a socket, but
3537 if the "connect()" function timed out, ip_connect() used to close the
3538 socket. However, verify_get_ident() also closes the socket later, and in
3539 between Exim writes to the log, which may get opened at this point. When
3540 the socket was closed in ip_connect(), the log could get the same file
3541 descriptor number as the socket. This naturally causes chaos. The fix is
3542 not to close the socket in ip_connect(); the socket should be closed by
3543 the function that creates it. There was only one place in the code where
3544 this was missing, in the iplookup router, which I don't think anybody now
3545 uses, but I've fixed it anyway.
3547 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
3548 well as to direct DNS lookups. Otherwise the handling of names in host
3549 lists is inconsistent and therefore confusing.
3555 PH/01 Two changes to the default runtime configuration:
3557 (1) Move the checks for relay_from_hosts and authenticated clients from
3558 after to before the (commented out) DNS black list checks.
3560 (2) Add control=submission to the relay_from_hosts and authenticated
3561 clients checks, on the grounds that messages accepted by these
3562 statements are most likely to be submissions.
3564 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
3566 (1) Generate an error if the third argument for the ${prvs expansion is
3569 (2) Treat a missing third argument of ${prvscheck as if it were an empty
3572 (3) Reset the variables that are obtained from the first argument of
3573 ${prvscheck and used in the second argument before leaving the code,
3574 because their memory is reclaimed, so using them afterwards may do
3577 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
3578 one (it's much easier than Tom thought :-).
3580 (5) Because of (4), we can now allow for the use of $prvscheck_result
3581 inside the third argument.
3583 PH/03 For some reason, the default setting of PATH when running a command from
3584 a pipe transport was just "/usr/bin". I have changed it to
3587 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
3588 anything to be listed in the output from -bV.
3590 PH/05 When a filter generated an autoreply, the entire To: header line was
3591 quoted in the delivery log line, like this:
3593 => >A.N.Other <ano@some.domain> <original@ddress> ...
3595 This has been changed so that it extracts the operative address. There
3596 may be more than one such address. If so, they are comma-separated, like
3599 => >ano@some.domain,ona@other.domain <original@ddress> ...
3601 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
3602 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
3603 not being looked up in the rDNS to get a host name, Exim was showing the
3604 IP address twice in Received: lines, even though the IP addresses were
3605 identical. For example:
3607 Received: from [1.2.3.4] (helo=[1.2.3.4])
3609 However, if the real host name was known, it was omitting the HELO data
3610 if it matched the actual IP address. This has been tidied up so that it
3611 doesn't show the same IP address twice.
3613 PH/07 When both +timestamp and +memory debugging was on, the value given by
3614 $tod_xxx expansions could be wrong, because the tod_stamp() function was
3615 called by the debug printing, thereby overwriting the timestamp buffer.
3616 Debugging no longer uses the tod_stamp() function when +timestamp is set.
3618 PH/08 When the original message was included in an autoreply transport, it
3619 always said "this is a copy of the message, including all the headers",
3620 even if body_only or headers_only was set. It now gives an appropriate
3623 PH/09 Applied a patch from the Sieve maintainer which:
3625 o fixes some comments
3626 o adds the (disabled) notify extension core
3627 o adds some debug output for the result of if/elsif tests
3628 o points to the current vacation draft in the documentation
3629 and documents the missing references header update
3633 o fixes a bug in processing the envelope test (when testing
3634 multiple envelope elements, the last element determined the
3637 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
3638 Electronic Mail") by including:
3640 Auto-submitted: auto-generated
3642 in the messages that it generates (bounce messages and others, such as
3643 warnings). In the case of bounce messages for non-SMTP messages, there was
3644 also a typo: it was using "Auto_submitted" (underscore instead of
3645 hyphen). Since every message generated by Exim is necessarily in response
3646 to another message, thes have all been changed to:
3648 Auto-Submitted: auto-replied
3650 in accordance with these statements in the RFC:
3652 The auto-replied keyword:
3654 - SHOULD be used on messages sent in direct response to another
3655 message by an automatic process,
3657 - MUST NOT be used on manually-generated messages,
3659 - MAY be used on Delivery Status Notifications (DSNs) and Message
3660 Disposition Notifications (MDNs),
3662 - MUST NOT be used on messages generated by automatic or periodic
3663 processes, except for messages which are automatic responses to
3666 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
3667 to the default Received: header definition.
3669 PH/12 Added log selector acl_warn_skipped (default on).
3671 PH/13 After a successful wildlsearch lookup, discard the values of numeric
3672 variables because (a) they are in the wrong storage pool and (b) even if
3673 they were copied, it wouldn't work properly because of the caching.
3675 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
3676 checking when decoding. Apparently there are clients that generate
3677 overlong encoded strings. Why am I not surprised?
3679 PH/15 If the first argument of "${if match_address" was not empty, but did not
3680 contain an "@" character, Exim crashed. Now it writes a panic log message
3681 and treats the condition as false.
3683 PH/16 In autoreply, treat an empty string for "once" the same as unset.
3685 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
3686 extension "envelope-auth". The code is finished and in agreement with
3687 other implementations, but there is no documentation so far and in fact,
3688 nobody wrote the draft yet. This extension is currently #undef'ed, thus
3689 not changing the active code.
3691 Print executed "if" and "elsif" statements when debugging is used. This
3692 helps a great deal to understand what a filter does.
3694 Document more things not specified clearly in RFC3028. I had all this
3695 sorted out, when out of a sudden new issues came to my mind. Oops."
3697 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
3700 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
3701 canonical form (as documented). However, after a host name lookup from
3702 the IP address, check_host() was doing a simple string comparison with
3703 addresses acquired from the DNS when checking that the found name did
3704 have the original IP as one of its addresses. Since any found IPv6
3705 addresses are likely to be in abbreviated form, the comparison could
3706 fail. Luckily, there already exists a function for doing the comparison
3707 by converting both addresses to binary, so now that is used instead of
3708 the text comparison.
3710 PH/20 There was another similar case to PH/19, when a complete host name was
3711 given in a host list; looking up its IP address could give an abbreviated
3712 form, whereas the current host's name might or might not be abbreviated.
3713 The same fix has been applied.
3719 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
3720 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
3723 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
3724 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
3726 PH/03 Typo: missing ".o" in src/pcre/Makefile.
3728 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
3729 header line, restrict the check to what is listed in RFCs 2369 and 2929.
3730 Also, for "Auto-Submitted", treat anything other than "no" as
3731 non-personal, in accordance with RFC 3834. (Previously it treated
3732 anything starting "auto-" as non-personal.)
3734 TF/01 The control=submission/name=... option had a problem with syntax
3735 errors if the name included a slash character. The /name= option
3736 now slurps the rest of the string, so it can include any characters
3737 but it must come last in the list of options (after /sender_retain
3740 PH/05 Some modifications to the interface to the fake nameserver for the new
3748 TK/01 Added the "success_on_redirect" address verification option. See
3749 NewStuff for rationale and an example.
3751 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
3753 PH/02 Patch to exigrep to allow it to work on syslog lines.
3755 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
3756 fread() to skip over the body file's header line, because in Cygwin the
3757 header line is locked and is inaccessible.
3759 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
3760 co-exist for some time) to make it clear that it is the Exim ID that is
3761 referenced, not the Message-ID: header line.
3763 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
3764 string_format() function, because snprintf() does not exist on all
3767 PH/06 The use of forbid_filter_existstest now also locks out the use of the
3768 ${stat: expansion item.
3770 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
3771 protocol synchronization error", to keep the pedants happy.
3773 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
3774 well as for IRIX systems, when gcc is being used. See the host.c source
3777 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
3779 PH/10 Named domain lists were not working if used in a queue_smtp_domains
3782 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
3783 transport and to the smtp transport in LMTP mode.
3785 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
3787 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
3788 run a filter in a subprocess. This could lead to confusion in subsequent
3789 lookups in the parent process. There should also be a search_tidyup() at
3790 the end of the subprocess.
3792 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
3793 only if the host matched helo_try_verify_hosts, which caused the
3794 verification to occur when the EHLO/HELO command was issued. The ACL just
3795 tested the remembered result. Now, if a previous verification attempt has
3796 not happened, "verify = helo" does it there and then.
3798 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
3800 TK/03 Fix log output including CR from clamd.
3802 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
3803 exist provoked a memory error which could cause a segfault.
3805 PH/15 Installed PCRE 6.2
3807 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
3809 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
3810 of the problem. Specifically, suggested +O2 rather than +O1 for the
3813 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
3815 PH/20 If a delivery was routed to a non-standard port by means of an SRV
3816 record, the port was not correctly logged when the outgoing_port log
3817 selector was set (it logged the transort's default port).
3819 PH/21 Added support for host-specific ports to manualroute, queryprogram,
3820 fallback_hosts, and "hosts" in the smtp transport.
3822 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
3823 host errors such as "Connection refused".
3825 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
3826 authentication with radiusclient 0.4.9:
3828 - Error returned from rc_read_config was caught wrongly
3829 - Username/password not passed on to radius server due to wrong length.
3831 The presumption is that some radiusclient API changes for 4.51/PH/17
3832 were not taken care of correctly. The code is still untested by me (my
3833 Linux distribution still has 0.3.2 of radiusclient), but it was
3834 contributed by a Radius user.
3836 PH/24 When doing a callout, the value of $domain wasn't set correctly when
3837 expanding the "port" option of the smtp transport.
3839 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
3840 while reading a MIME header. Thanks to Tom Hughes for a patch.
3842 PH/24 Include config.h inside local_scan.h so that configuration settings are
3845 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
3846 that in an ACL for RCPT (for example), you can examine exactly what was
3849 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
3850 commands, but it was not correctly comparing the address with the actual
3851 client host address. Thus, it would show the EHLO address in Received:
3852 header lines when this was not necessary.
3854 PH/27 Added the % operator to ${eval:}.
3856 PH/28 Exim tries to create and chdir to its spool directory when it starts;
3857 it should be ignoring failures (because with -C, for example, it has lost
3858 privilege). It wasn't ignoring creation failures other than "already
3861 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
3864 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
3865 because an input line was too long, either on its own, or by virtue of
3866 too many continuations, the temporary file was not being removed, and the
3867 return code was incorrect.
3869 PH/31 Missing "BOOL" in function definition in filtertest.c.
3871 PH/32 Applied Sieve patches from the maintainer.
3873 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
3875 PH/33 Added "verify = not_blind".
3877 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
3878 Local/Makefile (with some defaults set). These are used in built scripts
3879 such as exicyclog, but they have never been used in the exim_install
3880 script (though there are many overriding facilities there). I have
3881 arranged that the exim_install script now takes note of these two
3884 PH/35 Installed configuration files for Dragonfly.
3886 PH/36 When a locally submitted message by a trusted user did not contain a
3887 From: header, and the sender address was obtained from -f or from an SMTP
3888 MAIL command, and the trusted user did not use -F to supply a sender
3889 name, $originator_name was incorrectly used when constructing a From:
3890 header. Furthermore, $originator_name was used for submission mode
3891 messages from external hosts without From: headers in a similar way,
3892 which is clearly wrong.
3894 PH/37 Added control=suppress_local_fixups.
3896 PH/38 When log_selector = +received_sender was set, and the addition of the
3897 sender made the log line's construction buffer exactly full, or one byte
3898 less than full, an overflow happened when the terminating "\n" was
3901 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
3902 when the result of a list match is failure because a DNS lookup failed.
3904 PH/40 RM_COMMAND is now used in the building process.
3906 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
3907 the "build-* directories that it finds.
3909 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
3910 address in a domain literal was a prefix of an interface address.
3912 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
3913 when verifying a sender address, unless rewrite_headers is false.
3915 PH/44 Wrote a long comment about why errors_to addresses are verified as
3916 recipients, not senders.
3918 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
3919 the ratelimit ACL was added.
3921 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
3923 PH/46 Added extra information about PostgreSQL errors to the error string.
3925 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
3926 avoiding the need to install special zones in a real server. This is
3927 backwards compatible; if it can't find the fake resolver, it drops back.
3928 Thus, both old and new test suites can be run.
3930 TF/02 Added util/ratelimit.pl
3932 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
3933 clock is set back in time.
3935 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
3936 Candler <B.Candler@pobox.com>.
3938 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
3939 OK for addresses that are the result of redirections.
3941 PH/48 A number of further additions for the benefit of the new test suite,
3942 including a fake gethostbyname() that interfaces to the fake DNS resolver
3945 TF/06 The fix for widen_domains has also been applied to qualify_single and
3946 search_parents which are the other dnslookup options that can cause
3949 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
3952 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
3954 TF/07 Exim produced the error message "an SRV record indicated no SMTP
3955 service" if it encountered an MX record with an empty target hostname.
3956 The message is now "an MX or SRV record indicated no SMTP service".
3958 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
3959 if the DNS of the sending site is misconfigured. This is quite a
3960 common situation. This change restores the behaviour of treating a
3961 helo verification defer as a failure.
3963 PH/51 If self=fail was set on a router, the bounce message did not include the
3964 actual error message.
3970 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
3972 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
3973 command itself ended in error, the underlying message about the transport
3974 filter timeout was being overwritten with the pipe command error. Now the
3975 underlying error message should be appended to the second error message.
3977 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
3979 PH/02 Reduce the amount of output that "make" produces by default. Full output
3980 can still be requested.
3982 PH/03 The warning log line about a condition test deferring for a "warn" verb
3983 was being output only once per connection, rather than after each
3984 occurrence (because it was using the same function as for successful
3985 "warn" verbs). This seems wrong, so I have changed it.
3987 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
3988 it should not have, which might have caused a crash in the right
3989 circumstances, but probably never did.
3991 PH/04 Installed a modified version of Tony Finch's patch to make submission
3992 mode fix the return path as well as the Sender: header line, and to
3993 add a /name= option so that you can make the user's friendly name appear
3996 TF/03 Added the control = fakedefer ACL modifier.
3998 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
3999 Mark Lowes for thorough testing.
4001 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
4003 TK/03 Merged latest SRS patch from Miles Wilton.
4005 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
4006 with the definition in sysexits.h (which is #included earlier).
4007 Fortunately, Exim does not actually use EX_OK. The code used to try to
4008 preserve the sysexits.h value, by assuming that macro definitions were
4009 scanned for macro replacements. I have been disabused of this notion,
4010 so now the code just undefines EX_OK before #including unistd.h.
4012 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
4013 in the smtp transport. When a block could not be written in a single
4014 write() function, the timeout was being re-applied to each part-write.
4015 This seems wrong - if the receiver was accepting one byte at a time it
4016 would take for ever. The timeout is now adjusted when this happens. It
4017 doesn't have to be particularly precise.
4019 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
4020 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
4022 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
4023 without a domain if the check to <postmaster@domain> fails.
4025 SC/01 Eximstats: added -xls and the ability to specify output files
4026 (patch written by Frank Heydlauf).
4028 SC/02 Eximstats: use FileHandles for outputting results.
4030 SC/03 Eximstats: allow any combination of xls, txt, and html output.
4032 SC/04 Eximstats: fixed display of large numbers with -nvr option
4034 SC/05 Eximstats: fixed merging of reports with empty tables.
4036 SC/06 Eximstats: added the -include_original_destination flag
4038 SC/07 Eximstats: removed tabs and trailing whitespace.
4040 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
4042 TK/06 MBOX spool code: Add real "From " MBOX separator line
4043 so the .eml file is really in mbox format (even though
4044 most programs do not really care). Patch from Alex Miller.
4046 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
4047 The latter is generated from $received_to and is only set if the
4048 message has one envelope recipient. SA can use these headers,
4049 obviously out-of-the-box. Patch from Alex Miller.
4051 PH/08 The ${def test on a variable was returning false if the variable's
4052 value was "0", contrary to what the specification has always said!
4053 The result should be true unless the variable is empty.
4055 PH/09 The syntax error of a character other than { following "${if
4056 def:variable_name" (after optional whitespace) was not being diagnosed.
4057 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
4058 accidental colon was present, for example, could give incorrect results.
4060 PH/10 Tidied the code in a number of places where the st_size field of a stat()
4061 result is used (not including appendfile, where other changes are about
4064 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
4065 This involved changing a lot of size variables from int to off_t. It
4066 should work with maildirs and everything.
4068 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
4069 spamd dying while we are connected to it.
4071 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
4074 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
4075 function for BDB 4.3.
4077 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
4079 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
4080 the sources for PCRE have been re-arranged and more files are now
4083 PH/15 The code I had for printing potentially long long variables in PH/11
4084 above was not the best (it lost precision). The length of off_t variables
4085 is now inspected at build time, and an appropriate printing format (%ld
4086 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
4087 to be "long long int" or "long int". This is needed for the internal
4088 formatting function string_vformat().
4090 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
4091 the configuration file to be ":syslog", then the script "guesses" where
4092 the logs files are, rather than using the compiled in default. In our
4093 case the guess is not the same as the compiled default, so the script
4094 suddenly stopped working when I started to use syslog. The patch checks
4095 to see if log_file_path is "". If so, it attempts to read it from exim
4096 with no configuration file to get the compiled in version, before it
4097 falls back to the previous guessing code."
4099 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
4100 implementing BATV in an Exim configuration. See NewStuff for the gory
4103 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
4104 Makefile that are specific to HP-UX.
4106 PH/18 If the "use_postmaster" option was set for a recipient callout together
4107 with the "random" option, the postmaster address was used as the MAIL
4108 FROM address for the random test, but not for the subsequent recipient
4109 test. It is now used for both.
4111 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
4112 patch removes a few documentation additions to RFC 3028, because the
4113 latest draft now contains them. It adds the new en;ascii-case comparator
4114 and a new error check for 8bit text in MIME parts. Comparator and
4115 require names are now matched exactly. I enabled the subaddress
4116 extension, but it is not well tested yet (read: it works for me)."
4118 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
4119 rework some of the code of TK/09 above to avoid the hardwired use of
4120 "%lld" and "long long". Replaced the call to snprintf() with a call to
4123 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
4124 records point to non-existent hosts", "retry timeout exceeded", and
4125 "retry time not reached for any host after a long failure period".
4127 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
4128 experimental DomainKeys support:
4130 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
4131 (2) On an error such as an illegally used "control", the wrong name for
4132 the control was given.
4134 These problems did NOT occur unless DomainKeys support was compiled.
4136 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
4138 PH/24 Added ${if match_ip condition.
4140 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
4141 databases so that it will be absolutely obvious if a crash occurs in the
4142 DB library. This is a regular occurrence (often caused by mis-matched
4145 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
4146 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
4147 picked up on a user's system that detects such things. There doesn't seem
4148 to be a gcc warning option for this - only an attribute that has to be
4149 put on the function's prototype. It seems that in Fedora Core 4 they have
4150 set this on a number of new functions. No doubt there will be more in due
4153 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
4154 specify a transport. However, if an address that was verified by such a
4155 router was the subject of a callout, Exim crashed because it tried to
4156 read the rcpt_include_affixes from the non-existent transport. Now it
4157 just assumes that the setting of that option is false. This bug was
4158 introduced by 4.51/PH/31.
4160 PH/28 Changed -d+all to exclude +memory, because that information is very
4161 rarely of interest, but it makes the output a lot bigger. People tend to
4162 do -d+all out of habit.
4164 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
4165 code in os-type was giving problems when libc.so lives in lib64, like on
4168 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
4169 aren't the modern standard, and it seems that some systems' include files
4170 don't always have them. Exim was already checking for some of the newer
4171 ones like T_AAAA, and defining it itself. I've added checks for all the
4172 record types that Exim uses.
4174 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
4175 not automatically generating a new one, as it is supposed to. This
4176 prevented TLS from working. If the file did exist, but contained invalid
4177 data, a new version was generated, as expected. It was only the case of a
4178 non-existent file that was broken.
4180 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
4181 with a change in libdomainkeys > 0.64.
4183 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
4184 from DNS. If the selector record carries the flag, it now has
4185 precedence over the domain-wide flag.
4187 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
4189 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
4190 the use of a port name that isn't defined in /etc/services) occurred, the
4191 message was deferred as in a normal delivery, and thus remained on the
4192 spool, instead of being failed because of the mua_wrapper setting. This
4193 is now fixed, and I tidied up some of the mua_wrapper messages at the
4196 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
4197 the messages in a hash of arrays rather than using individual hashes.
4198 This is a bit cleaner and results in dramatic memory savings, albeit
4199 at a slight CPU cost.
4201 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
4202 as requested by Marc Sherman.
4204 SC/10 Eximstats: added histograms for user specified patterns as requested
4207 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
4209 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
4210 fopen() in the content-scanning modules that did not already have it.
4216 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
4217 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
4219 TK/02 Fix ACL "control" statement not being available in MIME ACL.
4221 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
4223 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
4224 to test Sieve filters that use "vacation".
4226 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
4227 that changes the way the GnuTLS parameters are stored in the cache file.
4228 The new format can be generated externally. For backward compatibility,
4229 if the data in the cache doesn't make sense, Exim assumes it has read an
4230 old-format file, and it generates new data and writes a new file. This
4231 means that you can't go back to an older release without removing the
4234 PH/03 A redirect router that has both "unseen" and "one_time" set does not
4235 work if there are any delivery delays because "one_time" forces the
4236 parent to be marked "delivered", so its unseen clone is never tried
4237 again. For this reason, Exim now forbids the simultaneous setting of
4240 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
4241 redirected to themselves ("homonym" addresses). Read the long ChangeLog
4242 entry if you want to know the details. The fix, however, neglected to
4243 consider the case when local delivery batching is involved. The test for
4244 "previously delivered" was not happening when checking to see if an
4245 address could be batched with a previous (undelivered) one; under
4246 certain circumstances this could lead to multiple deliveries to the same
4249 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
4250 in its include files, and this causes problems building Exim.
4252 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
4253 header_syntax) but Exim was just ignoring anything given after a slash.
4254 In particular, this caused confusion with an attempt to use "verify =
4255 reverse_host_lookup/defer_ok". An error is now given when options are
4256 supplied for verify items that do not have them. (Maybe reverse_host_
4257 lookup should have a defer_ok option, but that's a different point.)
4259 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
4260 defined by RFC 821) to 2048, because there were problems with some AUTH
4261 commands, and RFC 1869 says the size should be increased for extended
4262 SMTP commands that take arguments.
4264 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
4267 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
4268 "unknown" error; now it says that the functionality isn't in the binary.
4270 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
4271 an address' error message when a string expansion fails (syntax or
4272 whatever). Otherwise the password may appear in the log. Following change
4273 PH/42 below, there is no longer a chance of it appearing in a bounce
4276 PH/11 Installed exipick version 20050225.0 from John Jetmore.
4278 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
4279 of its addresses was ever tried. (Bugzilla bug #2.)
4281 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
4282 the result incorrectly in the debug output. (It correctly added a newline
4283 to what was transported.)
4285 TF/01 Added $received_time.
4287 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
4288 commented out examples of how to interface to a virus scanner and to
4289 SpamAssassin. Also added commented examples of av_scanner and
4290 spamd_address settings.
4292 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
4293 and controls are allowed in which ACLs. There were a couple of minor
4294 errors. Some of the entries in the conditions table (which is a table of
4295 where they are NOT allowed) were getting very unwieldy; rewrote them as a
4296 negation of where the condition IS allowed.
4298 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
4300 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
4301 header file does not have a version number, so I've had to invent a new
4302 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
4303 API. The code is untested by me (my Linux distribution still has 0.3.2 of
4304 radiusclient), but it was contributed by a Radius user.
4306 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
4307 files or directories, for OpenSSL.
4309 PH/19 When an Exim process that is running as root has to create an Exim log
4310 file, it does so in a subprocess that runs as exim:exim so as to get the
4311 ownership right at creation (otherwise, other Exim processes might see
4312 the file with the wrong ownership). There was no test for failure of this
4313 fork() call, which would lead to the process getting stuck as it waited
4314 for a non-existent subprocess. Forks do occasionally fail when resources
4315 run out. I reviewed all the other calls to fork(); they all seem to check
4318 PH/20 When checking for unexpected SMTP input at connect time (before writing
4319 the banner), Exim was not dealing correctly with a non-positive return
4320 from the read() function. If the client had disconnected by this time,
4321 the result was a log entry for a synchronization error with an empty
4322 string after "input=" when read() returned zero. If read() returned -1
4323 (an event I could not check), uninitialized data bytes were printed.
4324 There were reports of junk text (parts of files, etc) appearing after
4327 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
4329 PH/22 Added support for macro redefinition, and (re)definition in between
4330 driver and ACL definitions.
4332 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
4333 forgetting to use the resulting value; it was using the unexpanded value.
4335 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
4336 hadn't been configured. The fix is from Juergen Kreileder, who
4337 understands it better than I do:
4339 "Here's what I see happening with three configured cyrus_sasl
4340 authenticators configured (plain, login, cram-md5):
4342 On startup auth_cyrus_sasl_init() gets called for each of these.
4343 This means three calls to sasl_listmech() without a specified mech_list.
4344 => SASL tests which mechs of all available mechs actually work
4345 => three warnings about OTP not working
4346 => the returned list contains: plain, login, cram-md5, digest-md5, ...
4348 With the patch, sasl_listmech() also gets called three times. But now
4349 SASL's mech_list option is set to the server_mech specified in the the
4350 authenticator. Or in other words, the answer from sasl_listmech()
4351 gets limited to just the mech you're testing for (which is different
4353 => the return list contains just 'plain' or 'login', 'cram-md5' or
4354 nothing depending on the value of ob->server_mech.
4356 I've just tested the patch: Authentication still works fine,
4357 unavailable mechs specified in the exim configuration are still
4358 caught, and the auth.log warnings about OTP are gone."
4360 PH/25 When debugging is enabled, the contents of the command line are added
4361 to the debugging output, even when log_selector=+arguments is not
4364 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
4365 answer is "GNU", and only if the return is "GNU/something" is the answer
4368 PH/27 $acl_verify_message is now set immediately after the failure of a
4369 verification in an ACL, and so is available in subsequent modifiers. In
4370 particular, the message can be preserved by coding like this:
4372 warn !verify = sender
4373 set acl_m0 = $acl_verify_message
4375 Previously, $acl_verify_message was set only while expanding "message"
4376 and "log_message" when a very denied access.
4378 PH/28 Modified OS/os.c-Linux with
4380 -#ifndef OS_LOAD_AVERAGE
4381 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
4383 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
4384 nomenclature these days.)
4386 PH/29 Installed patch from the Sieve maintainer that adds the options
4387 sieve_useraddress and sieve_subaddress to the redirect router.
4389 PH/30 In these circumstances:
4390 . Two addresses routed to the same list of hosts;
4391 . First host does not offer TLS;
4392 . First host accepts first address;
4393 . First host gives temporary error to second address;
4394 . Second host offers TLS and a TLS session is established;
4395 . Second host accepts second address.
4396 Exim incorrectly logged both deliveries with the TLS parameters (cipher
4397 and peerdn, if requested) that were in fact used only for the second
4400 PH/31 When doing a callout as part of verifying an address, Exim was not paying
4401 attention to any local part prefix or suffix that was matched by the
4402 router that accepted the address. It now behaves in the same way as it
4403 does for delivery: the affixes are removed from the local part unless
4404 rcpt_include_affixes is set on the transport.
4406 PH/32 Add the sender address, as F=<...>, to the log line when logging a
4407 timeout during the DATA phase of an incoming message.
4409 PH/33 Sieve envelope tests were broken for match types other than :is. I have
4410 applied a patch sanctioned by the Sieve maintainer.
4412 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
4413 the uid or gid is negative. A case of a negative gid caused this to be
4414 noticed. The fix allows for either to be negative.
4416 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
4417 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
4420 PH/36 The change PH/12 above was broken. Fixed it.
4422 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
4423 the grounds that routing the same address twice would always produce the
4424 same answer. This might have been true once, but it is certainly no
4425 longer true now. Routing a child address may depend on the previous
4426 routing that produced that child. Some complicated redirection strategies
4427 went wrong when messages had multiple recipients, and made Exim's
4428 behaviour dependent on the order in which the addresses were given.
4430 I have moved the duplicate checking until after the routing is complete.
4431 Exim scans the addresses that are assigned to local and remote
4432 transports, and removes any duplicates. This means that more work will be
4433 done, as duplicates will always all be routed, but duplicates are
4434 presumably rare, so I don't expect this is of any significance.
4436 For deliveries to pipes, files, and autoreplies, the duplicate checking
4437 still happens during the routing process, since they are not going to be
4440 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
4441 It corrects a timeout issue with spamd. This is Ian's comment: "The
4442 background is that sometimes spamd either never reads data from a
4443 connection it has accepted, or it never writes response data. The exiscan
4444 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
4445 blindly assumes that writes won't block so it may never time out."
4447 PH/39 Allow G after quota size as well as K and M.
4449 PH/40 The value set for $authenticated_id in an authenticator may not contain
4450 binary zeroes or newlines because the value is written to log lines and
4451 to spool files. There was no check on this. Now the value is run through
4452 the string_printing() function so that such characters are converted to
4453 printable escape sequences.
4455 PH/41 $message_linecount is a new variable that contains the total number of
4456 lines in the message. Compare $body_linecount, which is the count for the
4459 PH/42 Exim no longer gives details of delivery errors for specific addresses in
4460 bounce and delay warning messages, except in certain special cases, which
4463 (a) An SMTP error message from a remote host;
4464 (b) A message specified in a :fail: redirection;
4465 (c) A message specified in a "fail" command in a system filter;
4466 (d) A message specified in a FAIL return from the queryprogram router;
4467 (e) A message specified by the cannot_route_message router option.
4469 In these cases only, Exim does include the error details in bounce and
4470 warning messages. There are also a few cases where bland messages such
4471 as "unrouteable address" or "local delivery error" are given.
4473 PH/43 $value is now also set for the "else" part of a ${run expansion.
4475 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
4476 being worked on, but at least Exim now implements the latest version to
4479 PH/45 In a pipe transport, although a timeout while waiting for the pipe
4480 process to complete was treated as a delivery failure, a timeout while
4481 writing the message to the pipe was logged, but erroneously treated as a
4482 successful delivery. Such timeouts include transport filter timeouts. For
4483 consistency with the overall process timeout, these timeouts are now
4484 treated as errors, giving rise to delivery failures by default. However,
4485 there is now a new Boolean option for the pipe transport called
4486 timeout_defer, which, if set TRUE, converts the failures into defers for
4487 both kinds of timeout. A transport filter timeout is now identified in
4490 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
4491 systems where "make" and "gmake" are different, calling "gmake" at top
4492 level broke things. I've arranged for the value of $(MAKE) to be passed
4493 from the Makefile to this script so that it can call the same version of
4497 A note about Exim versions 4.44 and 4.50
4498 ----------------------------------------
4500 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
4501 changes of various kinds. As a consequence, a big documentation update was
4502 needed. This delayed the release for rather longer than seemed good, especially
4503 in the light of a couple of (minor) security issues. Therefore, the changes
4504 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
4505 release. So 4.44 and 4.50 are in effect two different branches that both start
4508 I have left the 4.50 change log unchanged; it contains all the changes since
4509 4.43. The change log for 4.44 is below; many of its items are identical to
4510 those for 4.50. This seems to be the most sensible way to preserve the
4511 historical information.
4517 1. Minor wording change to the doc/README.SIEVE file.
4519 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
4520 computation of the current number of files was incorrect.
4522 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4523 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4526 4. Give more explanation in the error message when the command for a transport
4527 filter fails to execute.
4529 5. There are several places where Exim runs a non-Exim command in a
4530 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4531 was being done only for the command run by the queryprogram router. It is
4532 now done for all such subprocesses. The other cases are: ${run, transport
4533 filters, and the commands run by the lmtp and pipe transports.
4535 6. Added CONFIGURE_GROUP build-time option.
4537 7. Some older OS have a limit of 256 on the maximum number of file
4538 descriptors. Exim was using setrlimit() to set 1000 as a large value
4539 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4540 systems. I've change it so that if it can't get 1000, it tries for 256.
4542 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4543 was an oversight, and furthermore, ever since the addition of extra
4544 controls (e.g. 4.43/32), the checks on when to allow different forms of
4545 "control" were broken. There should now be diagnostics for all cases when a
4546 control that does not make sense is encountered.
4548 9. Added the /retain_sender option to "control=submission".
4550 10. $recipients is now available in the predata ACL (oversight).
4552 11. Tidy the search cache before the fork to do a delivery from a message
4553 received from the command line. Otherwise the child will trigger a lookup
4554 failure and thereby defer the delivery if it tries to use (for example) a
4555 cached ldap connection that the parent has called unbind on.
4557 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4558 of $address_data from the recipient verification was clobbered by the
4559 sender verification.
4561 13. The value of address_data from a sender verification is now available in
4562 $sender_address_data in subsequent conditions in the ACL statement.
4564 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
4566 15. Added a new option "connect=<time>" to callout options, to set a different
4569 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4570 was its contents. (It was OK if the option was not defined at all.)
4572 17. A "Completed" log line is now written for messages that are removed from
4573 the spool by the -Mrm option.
4575 18. New variables $sender_verify_failure and $recipient_verify_failure contain
4576 information about exactly what failed.
4578 19. Added -dd to debug only the daemon process.
4580 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
4581 handles timeouts, both on the server side and network timeouts. Renamed the
4582 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
4584 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
4585 It is now set to "smtps".
4587 22. $host_address is now set to the target address during the checking of
4588 ignore_target_hosts.
4590 23. When checking ignore_target_hosts for an ipliteral router, no host name was
4591 being passed; this would have caused $sender_host_name to have been used if
4592 matching the list had actually called for a host name (not very likely,
4593 since this list is usually IP addresses). A host name is now passed as
4596 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4597 code that specifies a non-restarting handler (typically sigaction() in
4598 modern systems) in an attempt to fix a rare and obscure crash bug.
4600 25. Narrowed the window for a race in the daemon that could cause it to ignore
4601 SIGCHLD signals. This is not a major problem, because they are used only to
4602 wake it up if nothing else does.
4604 26. A malformed maildirsize file could cause Exim to calculate negative values
4605 for the mailbox size or file count. Odd effects could occur as a result.
4606 The maildirsize information is now recalculated if the size or filecount
4609 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4610 support for a long time. Removed HAVE_SYS_VFS_H.
4612 28. Installed the latest version of exipick from John Jetmore.
4614 29. In an address list, if the pattern was not a regular expression, an empty
4615 subject address (from a bounce message) matched only if the pattern was an
4616 empty string. Non-empty patterns were not even tested. This was the wrong
4617 because it is perfectly reasonable to use an empty address as part of a
4618 database query. An empty address is now tested by patterns that are
4619 lookups. However, all the other forms of pattern expect the subject to
4620 contain a local part and a domain, and therefore, for them, an empty
4621 address still always fails if the pattern is not itself empty.
4623 30. Exim went into a mad DNS loop when attempting to do a callout where the
4624 host was specified on an smtp transport, and looking it up yielded more
4625 than one IP address.
4627 31. Re-factored the code for checking spool and log partition space into a
4628 function that finds that data and another that does the check. The former
4629 is then used to implement four new variables: $spool_space, $log_space,
4630 $spool_inodes, and $log_inodes.
4632 32. The RFC2047 encoding function was originally intended for short strings
4633 such as real names; it was not keeping to the 75-character limit for
4634 encoded words that the RFC imposes. It now respects the limit, and
4635 generates multiple encoded words if necessary. To be on the safe side, I
4636 have increased the buffer size for the ${rfc2047: expansion operator from
4639 33. It is now permitted to omit both strings after an "if" condition; if the
4640 condition is true, the result is "true". As before, when the second string
4641 is omitted, a false condition yields an empty string. This makes it less
4642 cumbersome to write custom ACL and router conditions.
4644 34. Failure to deliver a bounce message always caused it to be frozen, even if
4645 there was an errors_to setting on the router. The errors_to setting is now
4648 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4649 canonical form (fully expanded) before being placed in
4650 $sender_host_address.
4652 36. The table in the code that translates DNS record types into text (T_A to
4653 "A" for instance) was missing entries for NS and CNAME. It is just possible
4654 that this could have caused confusion if both these types were looked up
4655 for the same domain, because the text type is used as part of Exim's
4656 per-process caching. But the chance of anyone hitting this buglet seems
4659 37. The dnsdb lookup has been extended in a number of ways.
4661 (1) There is a new type, "zns", which walks up the domain tree until it
4662 finds some nameserver records. It should be used with care.
4664 (2) There is a new type, "mxh", which is like "mx" except that it returns
4665 just the host names, not the priorities.
4667 (3) It is now possible to give a list of domains (or IP addresses) to be
4668 looked up. The behaviour when one of the lookups defers can be
4669 controlled by a keyword.
4671 (4) It is now possible to specify the separator character for use when
4672 multiple records are returned.
4674 38. The dnslists ACL condition has been extended: it is now possible to supply
4675 a list of IP addresses and/or domains to be looked up in a particular DNS
4678 39. Added log_selector=+queue_time_overall.
4680 40. When running the queue in the test harness, wait just a tad after forking a
4681 delivery process, to get repeatability of debugging output.
4683 41. Include certificate and key file names in error message when GnuTLS fails
4684 to set them up, because the GnuTLS error message doesn't include the name
4685 of the failing file when there is a problem reading it.
4687 42. Allow both -bf and -bF in the same test run.
4689 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
4691 44. The "Exiscan patch" is now merged into the mainline Exim source.
4693 45. Sometimes the final signoff response after QUIT could fail to get
4694 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4695 < 0 before doing a fflush(). This bug looks as though it goes back to the
4696 introduction of TLS in release 3.20, but "sometimes" must have been rare
4697 because the tests only now provoked it.
4699 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
4700 (this can affect the format of dates).
4702 47. exim_tidydb, when checking for the continued existence of a message for
4703 which it has found a message-specific retry record, was not finding
4704 messages that were in split spool directories. Consequently, it was
4705 deleting retry records that should have stayed in existence.
4707 48. Steve fixed some bugs in eximstats.
4709 49. The SPA authentication driver was not abandoning authentication and moving
4710 on to the next authenticator when an expansion was forced to fail,
4711 contradicting the general specification for all authenticators. Instead it
4712 was generating a temporary error. It now behaves as specified.
4714 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
4715 (the order specifies the preference for clients). The order is now AES256,
4716 AES128, 3DES, ARCFOUR128.
4718 51. Small patch to Sieve code - explicitly set From: when generating an
4721 52. Exim crashed if a remote delivery caused a very long error message to be
4722 recorded - for instance if somebody sent an entire SpamAssassin report back
4723 as a large number of 550 error lines. This bug was coincidentally fixed by
4724 increasing the size of one of Exim's internal buffers (big_buffer) that
4725 happened as part of the Exiscan merge. However, to be on the safe side, I
4726 have made the code more robust (and fixed the comments that describe what
4729 53. Now that there can be additional text after "Completed" in log lines (if
4730 the queue_time_overall log selector is set), a one-byte patch to exigrep
4731 was needed to allow it to recognize "Completed" as not the last thing in
4734 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
4735 patch that reportedly fixes this has been added. I am not expert enough to
4736 create a test for it. This is what the patch creator wrote:
4738 "I found a little strange behaviour of ldap code when working with
4739 Windows 2003 AD Domain, where users was placed in more than one
4740 Organization Units. When I tried to give exim partial DN, the exit code
4741 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
4742 But simultaneously result of request was absolutely normal ldap result,
4743 so I produce this patch..."
4745 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
4746 REFERENCE, so I have modified the code to exclude the patch when that macro
4749 55. Some experimental protocols are using DNS PTR records for new purposes. The
4750 keys for these records are domain names, not reversed IP addresses. The
4751 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4752 leaves it alone. Component reversal etc. now happens only for IP addresses.
4755 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
4757 57. Double the size of the debug message buffer (to 2048) so that more of very
4758 long debug lines gets shown.
4760 58. The exicyclog utility now does better if the number of log files to keep
4761 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
4763 59. Two changes related to the smtp_active_hostname option:
4765 (1) $smtp_active_hostname is now available as a variable.
4766 (2) The default for smtp_banner uses $smtp_active_hostname instead
4767 of $primary_hostname.
4769 60. The host_aton() function is supposed to be passed a string that is known
4770 to be a valid IP address. However, in the case of IPv6 addresses, it was
4771 not checking this. This is a hostage to fortune. Exim now panics and dies
4772 if the condition is not met. A case was found where this could be provoked
4773 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4774 components; fortuitously, this particular loophole had already been fixed
4775 by change 4.50/55 above.
4777 If there are any other similar loopholes, the new check in host_aton()
4778 itself should stop them being exploited. The report I received stated that
4779 data on the command line could provoke the exploit when Exim was running as
4780 exim, but did not say which command line option was involved. All I could
4781 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4782 running as the user.
4785 61. There was a buffer overflow vulnerability in the SPA authentication code
4786 (which came originally from the Samba project). I have added a test to the
4787 spa_base64_to_bits() function which I hope fixes it.
4790 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
4791 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
4793 63. The daemon start-up calls getloadavg() while still root for those OS that
4794 need the first call to be done as root, but it missed one case: when
4795 deliver_queue_load_max is set with deliver_drop_privilege. This is
4796 necessary for the benefit of the queue runner, because there is no re-exec
4797 when deliver_drop_privilege is set.
4799 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
4800 This has been fixed.
4802 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4803 was in use, was not putting the data itself into the right store pool;
4804 consequently, it could be overwritten for a subsequent message in the same
4805 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4808 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
4810 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
4811 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
4812 the function were treating the return as a boolean value, which happened to
4813 work because 0=false and not-0=true, but is not correct code.
4815 68. The host_aton() function was not handling scoped IPv6 addresses (those
4816 with, for example, "%eth0" on the end) correctly.
4818 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
4819 negated items (that is, ~something) in unsigned ints. Some compilers
4820 apparently mutter when there is no cast.
4822 70. If an address verification called from an ACL failed, and did not produce a
4823 user-specific message (i.e. there was only a "system" message), nothing was
4824 put in $acl_verify_message. In this situation, it now puts the system
4827 71. Change 4.23/11 added synchronization checking at the start of an SMTP
4828 session; change 4.31/43 added the unwanted input to the log line - except
4829 that it did not do this in the start of session case. It now does.
4831 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
4832 This is wrong and can cause the other end to generate a synchronization
4833 error if it is another Exim or anything else that does the synchronization
4834 check. A QUIT command is no longer sent after a timeout.
4836 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
4837 during host lookups.
4839 74. The defer_ok option of callout verification was not working if it was used
4840 when verifying addresses in header lines, that is, for this case:
4842 verify = header_sender/callout=defer_ok
4844 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
4845 those file descriptors could be used for SMTP connections. If anything
4846 wrote to stderr (the example that came up was "warn" in embedded Perl), it
4847 could be sent to the SMTP client, causing chaos. The daemon now opens
4848 stdin, stdout, and stderr to /dev/null when it puts itself into the
4851 76. Arrange for output from Perl's "warn" command to be written to Exim's main
4852 log by default. The user can override this with suitable Perl magic.
4854 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
4855 the log message when discard triggers, was not working for the DATA ACL or
4856 for the non-SMTP ACL.
4858 78. Error message wording change in sieve.c.
4860 79. If smtp_accept_max_per_host was set, the number of connections could be
4861 restricted to fewer than expected, because the daemon was trying to set up
4862 a new connection before checking whether the processes handling previous
4863 connections had finished. The check for completed processes is now done
4864 earlier. On busy systems, this bug wouldn't be noticed because something
4865 else would have woken the daemon, and it would have reaped the completed
4868 80. If a message was submitted locally by a user whose login name contained one
4869 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
4870 It caused a spool format error. I have fixed the spool reading code. A
4871 related problem was that the "from" clause in the Received: line became
4872 illegal because of the space(s). It is now covered by ${quote_local_part.
4874 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
4877 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
4879 83. Patch from Sieve maintainer to fix unterminated string problem in
4880 "vacation" handling.
4882 84. Some minor changes to the Linux configuration files to help with other
4883 OS variants using glibc.
4885 85. One more patch for Sieve to update vacation handling to latest spec.
4888 ----------------------------------------------------
4889 See the note above about the 4.44 and 4.50 releases.
4890 ----------------------------------------------------
4896 1. Change 4.43/35 introduced a bug that caused file counts to be
4897 incorrectly computed when quota_filecount was set in an appendfile
4900 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4901 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4904 3. Give more explanation in the error message when the command for a transport
4905 filter fails to execute.
4907 4. There are several places where Exim runs a non-Exim command in a
4908 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4909 was being done only for the command run by the queryprogram router. It is
4910 now done for all such subprocesses. The other cases are: ${run, transport
4911 filters, and the commands run by the lmtp and pipe transports.
4913 5. Some older OS have a limit of 256 on the maximum number of file
4914 descriptors. Exim was using setrlimit() to set 1000 as a large value
4915 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4916 systems. I've change it so that if it can't get 1000, it tries for 256.
4918 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4919 was an oversight, and furthermore, ever since the addition of extra
4920 controls (e.g. 4.43/32), the checks on when to allow different forms of
4921 "control" were broken. There should now be diagnostics for all cases when a
4922 control that does not make sense is encountered.
4924 7. $recipients is now available in the predata ACL (oversight).
4926 8. Tidy the search cache before the fork to do a delivery from a message
4927 received from the command line. Otherwise the child will trigger a lookup
4928 failure and thereby defer the delivery if it tries to use (for example) a
4929 cached ldap connection that the parent has called unbind on.
4931 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4932 of $address_data from the recipient verification was clobbered by the
4933 sender verification.
4935 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4936 was its contents. (It was OK if the option was not defined at all.)
4938 11. A "Completed" log line is now written for messages that are removed from
4939 the spool by the -Mrm option.
4941 12. $host_address is now set to the target address during the checking of
4942 ignore_target_hosts.
4944 13. When checking ignore_target_hosts for an ipliteral router, no host name was
4945 being passed; this would have caused $sender_host_name to have been used if
4946 matching the list had actually called for a host name (not very likely,
4947 since this list is usually IP addresses). A host name is now passed as
4950 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4951 code that specifies a non-restarting handler (typically sigaction() in
4952 modern systems) in an attempt to fix a rare and obscure crash bug.
4954 15. Narrowed the window for a race in the daemon that could cause it to ignore
4955 SIGCHLD signals. This is not a major problem, because they are used only to
4956 wake it up if nothing else does.
4958 16. A malformed maildirsize file could cause Exim to calculate negative values
4959 for the mailbox size or file count. Odd effects could occur as a result.
4960 The maildirsize information is now recalculated if the size or filecount
4963 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4964 support for a long time. Removed HAVE_SYS_VFS_H.
4966 18. Updated exipick to current release from John Jetmore.
4968 19. Allow an empty sender to be matched against a lookup in an address list.
4969 Previously the only cases considered were a regular expression, or an
4972 20. Exim went into a mad DNS lookup loop when doing a callout where the
4973 host was specified on the transport, if the DNS lookup yielded more than
4976 21. The RFC2047 encoding function was originally intended for short strings
4977 such as real names; it was not keeping to the 75-character limit for
4978 encoded words that the RFC imposes. It now respects the limit, and
4979 generates multiple encoded words if necessary. To be on the safe side, I
4980 have increased the buffer size for the ${rfc2047: expansion operator from
4983 22. Failure to deliver a bounce message always caused it to be frozen, even if
4984 there was an errors_to setting on the router. The errors_to setting is now
4987 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4988 canonical form (fully expanded) before being placed in
4989 $sender_host_address.
4991 24. Updated eximstats to version 1.33
4993 25. Include certificate and key file names in error message when GnuTLS fails
4994 to set them up, because the GnuTLS error message doesn't include the name
4995 of the failing file when there is a problem reading it.
4997 26. Expand error message when OpenSSL has problems setting up cert/key files.
5000 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
5001 (this can affect the format of dates).
5003 28. exim_tidydb, when checking for the continued existence of a message for
5004 which it has found a message-specific retry record, was not finding
5005 messages that were in split spool directories. Consequently, it was
5006 deleting retry records that should have stayed in existence.
5008 29. eximstats updated to version 1.35
5009 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
5010 1.35 - bugfix such that pie charts by volume are generated correctly
5012 30. The SPA authentication driver was not abandoning authentication and moving
5013 on to the next authenticator when an expansion was forced to fail,
5014 contradicting the general specification for all authenticators. Instead it
5015 was generating a temporary error. It now behaves as specified.
5017 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
5018 (the order specifies the preference for clients). The order is now AES256,
5019 AES128, 3DES, ARCFOUR128.
5021 31. Small patch to Sieve code - explicitly set From: when generating an
5024 32. Exim crashed if a remote delivery caused a very long error message to be
5025 recorded - for instance if somebody sent an entire SpamAssassin report back
5026 as a large number of 550 error lines. This bug was coincidentally fixed by
5027 increasing the size of one of Exim's internal buffers (big_buffer) that
5028 happened as part of the Exiscan merge. However, to be on the safe side, I
5029 have made the code more robust (and fixed the comments that describe what
5032 33. Some experimental protocols are using DNS PTR records for new purposes. The
5033 keys for these records are domain names, not reversed IP addresses. The
5034 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
5035 leaves it alone. Component reversal etc. now happens only for IP addresses.
5038 34. The host_aton() function is supposed to be passed a string that is known
5039 to be a valid IP address. However, in the case of IPv6 addresses, it was
5040 not checking this. This is a hostage to fortune. Exim now panics and dies
5041 if the condition is not met. A case was found where this could be provoked
5042 from a dnsdb PTR lookup with an IPv6 address that had more than 8
5043 components; fortuitously, this particular loophole had already been fixed
5044 by change 4.50/55 or 4.44/33 above.
5046 If there are any other similar loopholes, the new check in host_aton()
5047 itself should stop them being exploited. The report I received stated that
5048 data on the command line could provoke the exploit when Exim was running as
5049 exim, but did not say which command line option was involved. All I could
5050 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
5051 running as the user.
5054 35. There was a buffer overflow vulnerability in the SPA authentication code
5055 (which came originally from the Samba project). I have added a test to the
5056 spa_base64_to_bits() function which I hope fixes it.
5059 36. The daemon start-up calls getloadavg() while still root for those OS that
5060 need the first call to be done as root, but it missed one case: when
5061 deliver_queue_load_max is set with deliver_drop_privilege. This is
5062 necessary for the benefit of the queue runner, because there is no re-exec
5063 when deliver_drop_privilege is set.
5065 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
5066 was in use, was not putting the data itself into the right store pool;
5067 consequently, it could be overwritten for a subsequent message in the same
5068 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
5071 38. Sometimes the final signoff response after QUIT could fail to get
5072 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
5073 < 0 before doing a fflush(). This bug looks as though it goes back to the
5074 introduction of TLS in release 3.20, but "sometimes" must have been rare
5075 because the tests only now provoked it.
5081 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
5082 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
5083 Now it returns an integer. A number of calls were still expecting a T/F
5084 return. Fortuitously, in all cases, the tests worked in OK situations,
5085 which is the norm. However, things would have gone wrong on any write
5086 failures on the smtp file descriptor. This function is used when sending
5087 messages over SMTP and also when doing verify callouts.
5089 2. When Exim is called to do synchronous delivery of a locally submitted
5090 message (the -odf or -odi options), it no longer closes stderr before doing
5093 3. Implemented the mua_wrapper option.
5095 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
5097 5. Implemented the functions header_remove(), header_testname(),
5098 header_add_at_position(), and receive_remove_recipient(), and exported them
5101 6. If an ACL "warn" statement specified the addition of headers, Exim already
5102 inserted X-ACL-Warn: at the start if there was no header name. However, it
5103 was not making this test for the second and subsequent header lines if
5104 there were newlines in the string. This meant that an invalid header could
5105 be inserted if Exim was badly configured.
5107 7. Allow an ACL "warn" statement to add header lines at the start or after all
5108 the Received: headers, as well as at the end.
5110 8. Added the rcpt_4xx retry error code.
5112 9. Added postmaster_mailfrom=xxx to callout verification option.
5114 10. Added mailfrom=xxxx to the callout verification option, for verify=
5117 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
5118 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
5120 12. Inserted some casts to stop certain compilers warning when using pointer
5121 differences as field lengths or precisions in printf-type calls (mostly
5122 affecting debugging statements).
5124 13. Added optional readline() support for -be (dynamically loaded).
5126 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
5127 same clock tick as a message's arrival, so that its received time was the
5128 same as the "first fail" time on the retry record, and that message
5129 remained on the queue past the ultimate address timeout, every queue runner
5130 would try a delivery (because it was past the ultimate address timeout) but
5131 after another failure, the ultimate address timeout, which should have then
5132 bounced the address, did not kick in. This was a "< instead of <=" error;
5133 in most cases the first failure would have been in the next clock tick
5134 after the received time, and all would be well.
5136 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
5137 being recognized when the domain list was tested by the match_domain
5138 condition in an expansion string.
5140 16. Added the ${str2b64: operator.
5142 17. Exim was always calling setrlimit() to set a large limit for the number of
5143 processes, without checking whether the existing limit was already
5144 adequate. (It did check for the limit on file descriptors.) Furthermore,
5145 errors from getrlimit() and setrlimit() were being ignored. Now they are
5146 logged to the main and panic logs, but Exim does carry on, to try to do its
5147 job under whatever limits there are.
5149 18. Imported PCRE 5.0.
5151 19. Trivial typo in log message " temporarily refused connection" (the leading
5154 20. If the log selector return_path_on_delivery was set and an address was
5155 redirected to /dev/null, the delivery process crashed because it assumed
5156 that a return path would always be set for a "successful" delivery. In this
5157 case, the whole delivery is bypassed as an optimization, and therefore no
5160 21. Internal re-arrangement: the function for sending a challenge and reading
5161 a response while authentication was assuming a zero-terminated challenge
5162 string. It's now changed to take a pointer and a length, to allow for
5163 binary data in such strings.
5165 22. Added the cyrus_sasl authenticator (code supplied by MBM).
5167 23. Exim was not respecting finduser_retries when seeking the login of the
5168 uid under which it was called; it was always trying 10 times. (The default
5169 setting of finduser_retries is zero.) Also, it was sleeping after the final
5170 failure, which is pointless.
5172 24. Implemented tls_on_connect_ports.
5174 25. Implemented acl_smtp_predata.
5176 26. If the domain in control=submission is set empty, Exim assumes that the
5177 authenticated id is a complete email address when it generates From: or
5178 Sender: header lines.
5180 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
5181 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
5182 chown and chgrp in /bin and hostname in /usr/bin.
5184 28. Exim was keeping the "process log" file open after each use, just as it
5185 does for the main log. This opens the possibility of it remaining open for
5186 long periods when the USR1 signal hits a daemon. Occasional processlog
5187 errors were reported, that could have been caused by this. Anyway, it seems
5188 much more sensible not to leave this file open at all, so that is what now
5191 29. The long-running daemon process does not normally write to the log once it
5192 has entered its main loop, and it closes the log before doing so. This is
5193 so that log files can straightforwardly be renamed and moved. However,
5194 there are a couple of unusual error situations where the daemon does write
5195 log entries, and I had neglected to close the log afterwards.
5197 30. The text of an SMTP error response that was received during a remote
5198 delivery was being truncated at 512 bytes. This is too short for some of
5199 the long messages that one sometimes sees. I've increased the limit to
5202 31. It is now possible to make retry rules that apply only when a message has a
5203 specific sender, in particular, an empty sender.
5205 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
5206 it possible to be selective about when SMTP synchronization is enforced.
5208 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
5210 32. Implemented hosts_connection_nolog.
5212 33. Added an ACL for QUIT.
5214 34. Setting "delay_warning=" to disable warnings was not working; it gave a
5217 35. Added mailbox_size and mailbox_filecount to appendfile.
5219 36. Added control = no_multiline_responses to ACLs.
5221 37. There was a bug in the logic of the code that waits for the clock to tick
5222 in the case where the clock went backwards by a substantial amount such
5223 that the microsecond fraction of "now" was more than the microsecond
5224 fraction of "then" (but the whole seconds number was less).
5226 38. Added support for the libradius Radius client library this is found on
5227 FreeBSD (previously only the radiusclient library was supported).
5233 1. When certain lookups returned multiple values in the form name=value, the
5234 quoting of the values was not always being done properly. Specifically:
5235 (a) If the value started with a double quote, but contained no whitespace,
5237 (b) If the value contained whitespace other than a space character (i.e.
5238 tabs or newlines or carriage returns) it was not quoted.
5239 This fix has been applied to the mysql and pgsql lookups by writing a
5240 separate quoting function and calling it from the lookup code. The fix
5241 should probably also be applied to nisplus, ibase and oracle lookups, but
5242 since I cannot test any of those, I have not disturbed their existing code.
5244 2. A hit in the callout cache for a specific address caused a log line with no
5245 reason for rejecting RCPT. Now it says "Previous (cached) callout
5246 verification failure".
5248 3. There was an off-by-one bug in the queryprogram router. An over-long
5249 return line was truncated at 256 instead of 255 characters, thereby
5250 overflowing its buffer with the terminating zero. As well as fixing this, I
5251 have increased the buffer size to 1024 (and made a note to document this).
5253 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
5254 when Exim is waiting for an SMTP response from a remote server, Exim
5255 restarts its select() call on the socket, thereby resetting its timeout.
5256 This is not a problem when such interrupts are rare. Somebody set up a cron
5257 job to run exiwhat every 2 minutes, which is less than the normal select()
5258 timeout (5 or 10 minutes). This meant that the select() timeout never
5259 kicked in because it was always reset. I have fixed this by comparing the
5260 time when an interrupt arrives with the time at the start of the first call
5261 to select(). If more time than the timeout has elapsed, the interrupt is
5262 treated as a timeout.
5264 5. Some internal re-factoring in preparation for the addition of Sieve
5265 extensions (by MH). In particular, the "personal" test is moved to a
5266 separate function, and given an option for scanning Cc: and Bcc: (which is
5267 not set for Exim filters).
5269 6. When Exim created an email address using the login of the caller as the
5270 local part (e.g. when creating a From: or Sender: header line), it was not
5271 quoting the local part when it contained special characters such as @.
5273 7. Installed new OpenBSD configuration files.
5275 8. Reworded some messages for syntax errors in "and" and "or" conditions to
5276 try to make them clearer.
5278 9. Callout options, other than the timeout value, were being ignored when
5279 verifying sender addresses in header lines. For example, when using
5281 verify = header_sender/callout=no_cache
5283 the cache was (incorrectly) being used.
5285 10. Added a missing instance of ${EXE} to the exim_install script; this affects
5286 only the Cygwin environment.
5288 11. When return_path_on_delivery was set as a log selector, if different remote
5289 addresses in the same message used different return paths and parallel
5290 remote delivery occurred, the wrong values would sometimes be logged.
5291 (Whenever a remote delivery process finished, the return path value from
5292 the most recently started remote delivery process was logged.)
5294 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
5295 lines when AUTH and/or TLS are in use. This is the "received protocol"
5296 field. Exim used to use "asmtp" for authenticated SMTP, without any
5297 indication (in the protocol name) for TLS use. Now it follows the RFC and
5298 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
5299 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
5300 names appear in log lines as well as in Received: header lines.
5302 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
5303 extensions, and comparison tests, and to fix some bugs.
5305 14. Changes to the "personal" filter test:
5307 (1) The test was buggy in that it was just doing the equivalent of
5308 "contains" tests on header lines. For example, if a user's address was
5309 anne@some.where, the "personal" test would incorrectly be true for
5311 To: susanne@some.where
5313 This test is now done by extracting each address from the header in turn,
5314 and checking the entire address. Other tests that are part of "personal"
5315 are now done using regular expressions (for example, to check local parts
5316 of addresses in From: header lines).
5318 (2) The list of non-personal local parts in From: addresses has been
5319 extended to include "listserv", "majordomo", "*-request", and "owner-*",
5320 taken from the Sieve specification recommendations.
5322 (3) If the message contains any header line starting with "List-" it is
5323 treated as non-personal.
5325 (4) The test for "circular" in the Subject: header line has been removed
5326 because it now seems ill-conceived.
5328 15. Minor typos in src/EDITME comments corrected.
5330 16. Installed latest exipick from John Jetmore.
5332 17. If headers_add on a router specified a text string that was too long for
5333 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
5334 of string_sprintf() is now avoided.
5336 18. $message_body_size was not set (it was always zero) when running the DATA
5337 ACL and the local_scan() function.
5339 19. For the "mail" command in an Exim filter, no default was being set for
5340 the once_repeat time, causing a random time value to be used if "once" was
5341 specified. (If the value happened to be <= 0, no repeat happened.) The
5342 default is now 0s, meaning "never repeat". The "vacation" command was OK
5343 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
5344 (I found it when inspecting the code).
5346 20. There is now an overall timeout for performing a callout verification. It
5347 defaults to 4 times the callout timeout, which applies to individual SMTP
5348 commands during the callout. The overall timeout applies when there is more
5349 than one host that can be tried. The timeout is checked before trying the
5350 next host. This prevents very long delays if there are a large number of
5351 hosts and all are timing out (e.g. when the network connections are timing
5352 out). The value of the overall timeout can be changed by specifying an
5353 additional sub-option for "callout", called "maxwait". For example:
5355 verify = sender/callout=5s,maxwait=20s
5357 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
5358 to the end before writing, but this should make it even safer).
5360 22. Exim was forgetting that it had advertised PIPELINING for the second and
5361 subsequent messages on an SMTP connection. It was also not resetting its
5362 memory on STARTTLS and an internal HELO.
5364 23. When Exim logs an SMTP synchronization error within a session, it now
5365 records whether PIPELINING has been advertised or not.
5367 24. Added 3 instances of "(long int)" casts to time_t variables that were being
5368 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
5369 rather than long int.
5371 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
5373 26. Added the never_mail option to autoreply.
5379 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
5380 crash if the getsockname() call failed; this can happen if a connection is
5381 closed very soon after it is established. The problem was simply in the
5382 order in which certain operations were done, causing Exim to try to write
5383 to the SMTP stream before it had set up the file descriptor. The bug has
5384 been fixed by making things happen in the correct order.
5390 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
5391 before the connection was closed, thus losing the rejection response.
5393 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
5394 some early Solaris releases, but causes trouble in current releases where
5395 socklen_t is defined.
5397 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
5400 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
5403 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
5404 that releasing the top of it at the end releases what was used for sub-
5405 expansions (unless the block got too big). However, discard this block if
5406 the first thing is a variable or header, so that we can use its block when
5407 it is dynamic (useful for very large $message_headers, for example).
5409 6. Lookups now cache *every* query, not just the most recent. A new, separate
5410 store pool is used for this. It can be recovered when all lookup caches are
5411 flushed. Lookups now release memory at the end of their result strings.
5412 This has involved some general refactoring of the lookup sources.
5414 7. Some code has been added to the store_xxx() functions to reduce the amount
5415 of flapping under certain conditions.
5417 8. log_incoming_interface used to affect only the <= reception log lines. Now
5418 it causes the local interface and port to be added to several more SMTP log
5419 lines, for example "SMTP connection from", and rejection lines.
5421 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
5423 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
5425 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
5426 could be overwritten at the end of the current message (or the start of a
5427 new message if it was set in a HELO ACL). The value is now preserved for
5428 the duration of the SMTP connection.
5430 12. If a transport had a headers_rewrite setting, and a matching header line
5431 contained an unqualified address, that address was qualified, even if it
5432 did not match any rewriting rules. The underlying bug was that the values
5433 of the flags that permit the existence of unqualified sender and recipient
5434 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
5435 non-local messages, and by -bnq for local messages) were not being
5436 preserved with the message after it was received.
5438 13. When Exim was logging an SMTP synchronization error, it could sometimes log
5439 "next input=" as part of the text comprising the host identity instead of
5440 the correct text. The code was using the same buffer for two different
5441 strings. However, depending on which order the printing function evaluated
5442 its arguments, the bug did not always show up. Under Linux, for example, my
5443 test suite worked just fine.
5445 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
5446 doesn't work with some older versions of Perl. It has been changed to "my",
5447 which in any case is probably the better facility to use.
5449 15. A really picky compiler found some instances of statements for creating
5450 error messages that either had too many or two few arguments for the format
5453 16. The size of the buffer for calls to the DNS resolver has been increased
5454 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
5455 for addresses that have a lot of PTR records. This alleviates a problem; it
5456 does not fully solve it.
5458 17. A dnsdb lookup for PTR records that receives more data than will fit in the
5459 buffer now truncates the list and logs the incident, which is the same
5460 action as happens when Exim is looking up a host name and its aliases.
5461 Previously in this situation something unpredictable would happen;
5462 sometimes it was "internal error: store_reset failed".
5464 18. If a server dropped the connection unexpectedly when an Exim client was
5465 using GnuTLS and trying to read a response, the client delivery process
5466 crashed while trying to generate an error log message.
5468 19. If a "warn" verb in an ACL added multiple headers to a message in a single
5469 string, for example:
5471 warn message = H1: something\nH2: something
5473 the text was added as a single header line from Exim's point of view
5474 though it ended up OK in the delivered message. However, searching for the
5475 second and subsequent header lines using $h_h2: did not work. This has been
5476 fixed. Similarly, if a system filter added multiple headers in this way,
5477 the routers could not see them.
5479 20. Expanded the error message when iplsearch is called with an invalid key to
5480 suggest using net-iplsearch in a host list.
5482 21. When running tests using -bh, any delays imposed by "delay" modifiers in
5483 ACLs are no longer actually imposed (and a message to that effect is
5486 22. If a "gecos" field in a passwd entry contained escaped characters, in
5487 particular, if it contained a \" sequence, Exim got it wrong when building
5488 a From: or a Sender: header from that name. A second bug also caused
5489 incorrect handling when an unquoted " was present following a character
5490 that needed quoting.
5492 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
5493 was not being matched caselessly.
5495 24. Arranged for all hyphens in the exim.8 source to be escaped with
5498 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
5499 properly. Recipient callout cache records were still being keyed to include
5500 the sender, even when use_sender was set false. This led to far more
5501 callouts that were necessary. The sender is no longer included in the key
5502 when use_sender is false.
5504 26. Added "control = submission" modifier to ACLs.
5506 27. Added the ${base62d: operator to decode base 62 numbers.
5508 28. dnsdb lookups can now access SRV records.
5510 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
5511 the configuration file.
5513 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
5514 (-v) mode. This makes the output for a verbose queue run more intelligible.
5516 31. Added a use_postmaster feature to recipient callouts.
5518 32. Added the $body_zerocount variable, containing the number of binary zero
5519 bytes in the message body.
5521 33. The time of last modification of the "new" subdirectory is now used as the
5522 "mailbox time last read" when there is a quota error for a maildir
5525 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
5527 35. Added +ignore_unknown as a special item in host lists.
5529 36. Code for decoding IPv6 addresses in host lists is now included, even if
5530 IPv6 support is not being compiled. This fixes a bug in which an IPv6
5531 address was recognized as an IP address, but was then not correctly decoded
5532 into binary, causing unexpected and incorrect effects when compared with
5539 1. Very minor rewording of debugging text in manualroute to say "list of
5540 hosts" instead of "hostlist".
5542 2. If verify=header_syntax was set, and a header line with an unqualified
5543 address (no domain) and a large number of spaces between the end of the
5544 name and the colon was received, the reception process suffered a buffer
5545 overflow, and (when I tested it) crashed. This was caused by some obsolete
5546 code that should have been removed. The fix is to remove it!
5548 3. When running in the test harness, delay a bit after writing a bounce
5549 message to get a bit more predictability in the log output.
5551 4. Added a call to search_tidyup() just before forking a reception process. In
5552 theory, someone could use a lookup in the expansion of smtp_accept_max_
5553 per_host which, without the tidyup, could leave open a database connection.
5555 5. Added the variables $recipient_data and $sender_data which get set from a
5556 lookup success in an ACL "recipients" or "senders" condition, or a router
5557 "senders" option, similar to $domain_data and $local_part_data.
5559 6. Moved the writing of debug_print from before to after the "senders" test
5562 7. Change 4.31/66 (moving the time when the Received: is generated) caused
5563 problems for message scanning, either using a data ACL, or using
5564 local_scan() because the Received: header was not generated till after they
5565 were called (in order to set the time as the time of reception completion).
5566 I have revised the way this works. The header is now generated after the
5567 body is received, but before the ACL or local_scan() are called. After they
5568 are run, the timestamp in the header is updated.
5574 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
5575 before starting a queue runner without re-exec. This happened only when
5576 deliver_drop_privilege was set or when the Exim user was set to root. The
5577 effect of the bug was that timeouts during subsequent deliveries caused
5578 crashes instead of being properly handled. The handler is now left at its
5579 default (and expected) setting.
5581 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
5582 message, again when deliver_drop_privilege is set or Exim is run as root.
5583 The bug described in (1) was not present in this case, but the tidying up
5584 of the other signals was missing. I have made the two cases consistent.
5586 3. The ignore_target_hosts setting on a manualroute router was being ignored
5587 for hosts that were looked up using the /MX notation.
5589 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
5592 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
5593 operated on the sender address. After changing the $sender_address to <>
5594 for the sender address verify, Exim was re-instated it as the original
5595 (before rewriting) address, but remembering that it had rewritten it, so it
5596 wasn't rewriting it again. This bug also had the effect of breaking the
5597 sender address verification caching when the sender address was rewritten.
5599 6. The ignore_target_hosts option was being ignored by the ipliteral router.
5600 This has been changed so that if the ip literal address matches
5601 ignore_target_hosts, the router declines.
5603 7. Added expansion conditions match_domain, match_address, and match_local_
5604 part (NOT match_host).
5606 8. The placeholder for the Received: header didn't have a length field set.
5608 9. Added code to Exim itself and to exim_lock to test for a specific race
5609 condition that could lead to file corruption when using MBX delivery. The
5610 issue is with the lockfile that is created in /tmp. If this file is removed
5611 after a process has opened it but before that process has acquired a lock,
5612 there is the potential for a second process to recreate the file and also
5613 acquire a lock. This could lead to two Exim processes writing to the file
5614 at the same time. The added code performs the same test as UW imapd; it
5615 checks after acquiring the lock that its file descriptor still refers to
5616 the same named file.
5618 10. The buffer for building added header lines was of fixed size, 8192 bytes.
5619 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
5622 11. Added the smtp_active_hostname option. If used, this will typically be made
5623 to depend on the incoming interface address. Because $interface_address is
5624 not set up until the daemon has forked a reception process, error responses
5625 that can happen earlier (such as "too many connections") no longer contain
5628 12. If an expansion in a condition on a "warn" statement fails because a lookup
5629 defers, the "warn" statement is abandoned, and the next ACL statement is
5630 processed. Previously this caused the whole ACL to be aborted.
5632 13. Added the iplsearch lookup type.
5634 14. Added ident_timeout as a log selector.
5636 15. Added tls_certificate_verified as a log selector.
5638 16. Added a global option tls_require_ciphers (compare the smtp transport
5639 option of the same name). This controls incoming TLS connections.
5641 17. I finally figured out how to make tls_require_ciphers do a similar thing
5642 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
5643 before starting the TLS session.
5645 18. Tabs are now shown as \t in -bP output.
5647 19. If the log selector return_path_on_delivery was set, Exim crashed when
5648 bouncing a message because it had too many Received: header lines.
5650 20. If two routers both had headers_remove settings, and the first one included
5651 a superfluous trailing colon, the final name in the first list and the
5652 first name in the second list were incorrectly joined into one item (with a
5653 colon in the middle).
5659 1. Added -C and -D options to the exinext utility, mainly to make it easier
5660 to include in the automated testing, but these could be helpful when
5661 multiple configurations are in use.
5663 2. The exinext utility was not formatting the output nicely when there was
5664 an alternate port involved in the retry record key, nor when there was a
5665 message id as well (for retries that were specific to a specific message
5666 and a specific host). It was also confused by IPv6 addresses, because of
5667 the additional colons they contain. I have fixed the IPv4 problem, and
5668 patched it up to do a reasonable job for IPv6.
5670 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
5671 delivery, the log line now contains "pipelined" if PIPELINING was used.
5673 4. An SMTP transport process used to panic and die if the bind() call to set
5674 an explicit outgoing interface failed. This has been changed; it is now
5675 treated in the same way as a connect() failure.
5677 5. A reference to $sender_host_name in the part of a conditional expansion
5678 that was being skipped was still causing a DNS lookup. This no longer
5681 6. The def: expansion condition was not recognizing references to header lines
5682 that used bh_ and bheader_.
5684 7. Added the _cache feature to named lists.
5686 8. The code for checking quota_filecount in the appendfile transport was
5687 allowing one more file than it should have been.
5689 9. For compatibility with Sendmail, the command line option
5697 and sets the incoming protocol and host name (for trusted callers). The
5698 host name and its colon can be omitted when only the protocol is to be set.
5699 Note the Exim already has two private options, -pd and -ps, that refer to
5700 embedded Perl. It is therefore impossible to set a protocol value of "d" or
5701 "s", but I don't think that's a major issue.
5703 10. A number of refactoring changes to the code, none of which should affect
5706 (a) The number of logging options was getting close to filling up the
5707 32-bit word that was used as a bit map. I have split them into two classes:
5708 those that are passed in the argument to log_write(), and those that are
5709 only ever tested independently outside of that function. These are now in
5710 separate 32-bit words, so there is plenty of room for expansion again.
5711 There is no change in the user interface or the logging behaviour.
5713 (b) When building, for example, log lines, the code previously used a
5714 macro that called string_cat() twice, in order to add two strings. This is
5715 not really sufficiently general. Furthermore, there was one instance where
5716 it was actually wrong because one of the argument was used twice, and in
5717 one call a function was used. (As it happened, calling the function twice
5718 did not affect the overall behaviour.) The macro has been replaced by a
5719 function that can join an arbitrary number of extra strings onto a growing
5722 (c) The code for expansion conditions now uses a table and a binary chop
5723 instead of a serial search (which was left over from when there were very
5724 few conditions). Also, it now recognizes conditions like "pam" even when
5725 the relevant support is not compiled in: a suitably worded error message is
5726 given if an attempt is made to use such a condition.
5728 11. Added ${time_interval:xxxxx}.
5730 12. A bug was causing one of the ddress fields not to be passed back correctly
5731 from remote delivery subprocesses. The field in question was not being
5732 subsequently used, so this caused to problems in practice.
5734 13. Added new log selectors queue_time and deliver_time.
5736 14. Might have fixed a bug in maildirsizefile handling that threw up
5737 "unexpected character" debug warnings, and recalculated the data
5738 unnecessarily. In any case, I expanded the warning message to give more
5741 15. Added the message "Restricted characters in address" to the statements in
5742 the default ACL that block characters like @ and % in local parts.
5744 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
5745 Three changes have been made:
5747 (a) There was a serious bug; a negative response to MAIL caused the whole
5748 recipient domain to be cached as invalid, thereby blocking all messages
5749 to all local parts at the same domain, from all senders. This bug has
5750 been fixed. The domain is no longer cached after a negative response to
5751 MAIL if the sender used is not empty.
5753 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
5756 (c) A new callout option, "use_sender" has been added for people who want
5757 the modified behaviour.
5763 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
5766 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
5767 indeed breaks things for older releases.
5769 3. Added additional logging to the case where there is a problem reading data
5770 from a filter that is running in a subprocess using a pipe, in order to
5771 try to track down a specific problem.
5773 4. Testing facility fudge: when running in the test harness and attempting
5774 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
5775 getting "No route to host". Convert this to a timeout.
5777 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
5780 6. Some OS don't have socklen_t but use size_t instead. This affects the
5781 fifth argument of getsockopt() amongst other things. This is now
5782 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
5783 can be set for individual OS. I have set it for SunOS5, OSF1, and
5784 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
5785 some earlier ones do not.
5787 7. Change 4.30/15 was not doing the test caselessly.
5789 8. The standard form for an IPv6 address literal was being rejected by address
5790 parsing in, for example, MAIL and RCPT commands. An example of this kind of
5791 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5792 this, as well as the form without the "IPv6" on the front (but only when
5793 address literals are enabled, of course).
5795 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
5797 10. Exim crashed if a message with an empty sender address specified by -f
5798 encountered a router with an errors_to setting. This could be provoked only
5799 by a command such as
5803 where an empty string was supplied; "<>" did not hit this bug.
5805 11. Installed PCRE release 4.5.
5807 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
5808 remained set. It is now erased.
5810 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
5811 times from message ids (which are base 36 rather than the normal 62).
5813 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
5814 were being counted as actual protocol errors, and logged if the log
5815 selector +smtp_protocol_error was set. One cannot be perfect in this test,
5816 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
5817 and DATA following a set of rejected RCPTs do not count as protocol errors.
5818 In other words, Exim assumes they were pipelined, though this may not
5819 actually be the case. Of course, in all cases the client gets an
5820 appropriate error code.
5822 15. If a lookup fails in an ACL condition, a message about the failure may
5823 be available; it is used if testing the ACL cannot continue, because most
5824 such messages specify what the cause of the deferral is. However, some
5825 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
5826 that caused an old message to be retained and used if a later statement
5827 caused a defer, replacing the real cause of the deferral.
5829 16. If an IP address had so many PTR records that the DNS lookup buffer
5830 was not large enough to hold them, Exim could crash while trying to process
5831 the truncated data. It now detects and logs this case.
5833 17. Further to 4.21/58, another change has been made: if (and only if) the
5834 first line of a message (the first header line) ends with CRLF, a bare LF
5835 in a subsequent header line has a space inserted after it, so as not to
5836 terminate the header.
5838 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
5839 unnecessarily, used atoi() instead of strtol(), and didn't check the
5840 termination when getting file sizes from file names by regex.
5842 19. Completely re-implemented the support for maildirsize files, in the light
5843 of a number of problems with the previous contributed implementation
5844 (4.30/29). In particular:
5846 . If the quota is zero, the maildirsize file is maintained, but no quota is
5849 . If the maildir directory does not exist, it is created before any attempt
5850 to write a maildirsize file.
5852 . The quota value in the file is just a cache; if the quota is changed in
5853 the transport, the new value overrides.
5855 . A regular expression is available for excluding directories from the
5858 20. The autoreply transport checks the characters in options that define the
5859 message's headers; it allows continued headers, but it was checking with
5860 isspace() after an embedded newline instead of explicitly looking for a
5863 21. If all the "regular" hosts to which an address was routed had passed their
5864 expiry times, and had not reached their retry times, the address was
5865 bounced, even if fallback hosts were defined. Now Exim should go on to try
5868 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
5869 equivalent code in the SMTP transport. Some hosts send humungous responses
5870 to HELO/EHLO, more than 1024 it seems.
5872 23. Refactoring: code in filter.c used (void *) for "any old type" but this
5873 gives compiler warnings in some environments. I've now done it "properly",
5876 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
5877 (because of problems with the built-in one) was declared to return uschar *
5878 instead of char *, causing compiler failure.
5880 25. Fixed a file descriptor leak when processing alias/forward files.
5882 26. Fixed a minor format string issue in dbfn.c.
5884 27. Typo in exim.c: ("dmbnz" for "dbmnz").
5886 28. If a filter file refered to $h_xxx or $message_headers, and the headers
5887 contained RFC 2047 "words", Exim's memory could, under certain conditions,
5890 29. When a sender address is verified, it is cached, to save repeating the test
5891 when there is more than one recipient in a message. However, when the
5892 verification involves a callout, it is possible for different callout
5893 options to be set for different recipients. It is too complicated to keep
5894 track of this in the cache, so now Exim always runs a verification when a
5895 callout is required, relying on the callout cache for the optimization.
5896 The overhead is duplication of the address routing, but this should not be
5899 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
5900 to be verified with callout=postmaster, and the main callout worked but the
5901 postmaster check failed, the verification correctly failed. However, if a
5902 subsequent RCPT command asked for sender verification *without* the
5903 postmaster check, incorrect caching caused this verification also to fail,
5906 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
5907 it was not caching the DNS options (qualify_single, search_parents) that
5908 were used when the lookup failed. A subsequent lookup with different
5909 options therefore always gave the same answer, though there were cases
5910 where it should not have. (Example: a "domains = !$mx_any" option on a
5911 dnslookup router: the "domains" option is always processed without any
5912 widening, but the router might have qualify_single set.) Now Exim uses the
5913 cached value only when the same options are set.
5915 32. Added John Jetmore's "exipick" utility to the distribution.
5917 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
5918 than a timeout (e.g. a certificate is required, and is not provided), an
5919 Exim server now closes the connection immediately. Previously it waited for
5920 the client to close - but if the client is SSL, it seems that they each
5921 wait for each other, leading to a delay before one of them times out.
5923 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
5924 maintained 0.8.x compatibility because I don't think many are using it, and
5925 it is clearly obsolete.
5927 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
5930 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
5931 client certificate was expired. A simple patch fixes this, though I don't
5932 understand the full logic of why the verify callback is called multiple
5935 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
5936 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
5937 which causes problems with some clients (such as the Certicom SSL Plus
5938 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
5939 disables the coutermeasure allowing Eudora to connect."
5941 38. Exim was not checking that a write() to a log file succeeded. This could
5942 lead to Bad Things if a log got too big, in particular if it hit a file
5943 size limit. Exim now panics and dies if it cannot write to a log file, just
5944 as it does if it cannot open a log file.
5946 39. Modified OS/Makefile-Linux so that it now contains
5948 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
5950 The two -D definitions ensure that Exim is compiled with large file
5951 support, which makes it possible to handle log files that are bigger than
5954 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
5955 instance) a domain was checked against a named list that involved a lookup,
5956 causing $domain_data to be set, then another domain was checked against the
5957 same list, then the first domain was re-checked, the value of $domain_data
5958 after the final check could be wrong. In particular, if the second check
5959 failed, it could be set empty. This bug probably also applied to
5962 41. The strip_trailing_dot option was not being applied to the address given
5963 with the -f command-line option.
5965 42. The code for reading a message's header from the spool was incrementing
5966 $received_count, but never initializing it. This meant that the value was
5967 incorrect (doubled) while delivering a message in the same process in which
5968 it was received. In the most common configuration of Exim, this never
5969 happens - a fresh exec is done - but it can happen when
5970 deliver_drop_privilege is set.
5972 43. When Exim logs an SMTP synchronization error - client data sent too soon -
5973 it now includes up to 150 characters of the unexpected data in the log
5976 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
5977 and building data strings. The size of both of these buffers was 10 000
5978 bytes - far larger than anybody would *ever* want, thought I. Needless to
5979 say, somebody hit the limit. I have increased the maximum line length to
5980 20 000 and the maximum data length of concatenated lines to 100 000. I have
5981 also fixed two bugs, because there was no checking on these buffers. Tsk,
5982 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
5983 buffer is too small.
5985 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
5986 lsearch lookups. Now it does.
5988 46. When parsing a route_list item in a manualroute router, a fixed-length
5989 buffer was used for the list of hosts. I made this 1024 bytes long,
5990 thinking that nobody would ever have a list of hosts that long. Wrong.
5991 Somebody had a whole pile of complicated expansion conditions, and the
5992 string was silently truncated, leading to an expansion error. It turns out
5993 that it is easier to change to an unlimited length (owing to other changes
5994 that have happened since this code was originally written) than to build
5995 structure for giving a limitation error. The length of the item that
5996 expands into the list of hosts is now unlimited.
5998 47. The lsearch lookup could not handle data where the length of text line was
5999 more than 4095 characters. Such lines were truncated, leading to shortened
6000 data being returned. It should now handle lines of any length.
6002 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
6003 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
6006 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
6007 replacing strings such as BIN_DIRECTORY in a source file: the replacement
6008 no longer happens in comment lines. A list of replacements is now placed
6009 at the head of all of the source files, except those whose only change is
6010 to replace PERL_COMMAND in the very first #! line.
6012 50. Replaced the slow insertion sort in queue.c, for sorting the list of
6013 messages on the queue, with a bottom-up merge sort, using code contributed
6014 by Michael Haardt. This should make operations like -bp somewhat faster on
6015 large queues. It won't affect queue runners, except when queue_run_in_order
6018 51. Installed eximstats 1.31 in the distribution.
6020 52. Added support for SRV lookups to the dnslookup router.
6022 53. If an ACL referred to $message_body or $message_body_end, the value was not
6023 reset for any messages that followed in the same SMTP session.
6025 54. The store-handling optimization for building very long strings was not
6026 differentiating between the different store pools. I don't think this
6027 actually made any difference in practice, but I've tidied it.
6029 55. While running the routers to verify a sender address, $sender_address
6030 was still set to the sender address. This is wrong, because when routing to
6031 send a bounce to the sender, it would be empty. Therefore, I have changed
6032 it so that, while verifying a sender address, $sender_address is set to <>.
6033 (There is no change to what happens when verifying a recipient address.)
6035 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
6036 target A or AAAA records (if not already returned) without resetting the
6037 qualify_single or search_parents options of the DNS resolver. These are
6038 inappropriate in this case because the targets of MX and SRV records must
6039 be FQDNs. A broken DNS record could cause trouble if it happened to have a
6040 target that, when qualified, matched something in the local domain. These
6041 two options are now turned off when doing these lookups.
6043 57. It seems that at least some releases of Reiserfs (which does not have the
6044 concept of a fixed number of inodes) returns zero and not -1 for the
6045 number of available inodes. This interacted badly with check_spool_inodes,
6046 which assumed that -1 was the "no such thing" setting. What I have done is
6047 to check that the total number of inodes is greater than zero before doing
6048 the test of how many are available.
6050 58. When a "warn" ACL statement has a log_message modifier, the message is
6051 remembered, and not repeated. This is to avoid a lot of repetition when a
6052 message has many recipients that cause the same warning to be written.
6053 However, Exim was preserving the list of already written lines for an
6054 entire SMTP session, which doesn't seem right. The memory is now reset if a
6055 new message is started.
6057 59. The "rewrite" debugging flag was not showing the result of rewriting in the
6058 debugging output unless log_rewrite was also set.
6060 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
6061 of (int)(handle) when we know that handle contains (void *)(-1).
6063 61. The Exim daemon panic-logs an error return when it closes the incoming
6064 connection. However "connection reset by peer" seems to be common, and
6065 isn't really an error worthy of noting specially, so that particular error
6068 62. When Exim is trying to find all the local interfaces, it used to panic and
6069 die if the ioctl to get the interface flags failed. However, it seems that
6070 on at least one OS (Solaris 9) it is possible to have an interface that is
6071 included in the list of interfaces, but for which you get a failure error
6072 for this call. This happens when the interface is not "plumbed" into a
6073 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
6074 failure of the "get flags" call assumes that the interface is down.
6076 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
6077 makes life easier for people who are doing arithmetic on fields extracted
6078 from dates, where you often get leading zeros that should not be
6079 interpreted as octal.
6081 64. Added qualify_domain to the redirect router, to override the global
6084 65. If a pathologically long header line contained very many addresses (the
6085 report of this problem mentioned 10 000) and each of them was rewritten,
6086 Exim could use up a very large amount of memory. (It kept on making new
6087 copies of the header line as it rewrote, and never released the old ones.)
6088 At the expense of a bit more processing, the header rewriting function has
6089 been changed so that it no longer eats memory in this way.
6091 66. The generation of the Received: header has been moved from the time that a
6092 message starts to be received, to the time that it finishes. The timestamp
6093 in the Received: header should now be very close to that of the <= log
6094 line. There are two side-effects of this change:
6096 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
6097 logged header lines no longer include the local Received: line, because
6098 it has not yet been created. The same applies to a copy of the message
6099 that is returned to a non-SMTP sender when a message is rejected.
6101 (b) When a filter file is tested using -bf, no additional Received: header
6102 is added to the test message. After some thought, I decided that this
6105 This change does not affect the value of $received_for. It is still set
6106 after address rewriting, but before local_scan() is called.
6108 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
6110 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
6111 gave an unhelpful panic error message, and a defer error. I have managed to
6112 change this behaviour so that it now rejects any supplied certificate,
6113 which seems right, as the list of acceptable certificates is empty.
6115 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
6116 gave an unhelpful defer error. I have not managed to make this reject any
6117 supplied certificates, but the error message it gives is "no certificate
6118 supplied", which is not helpful.
6120 70. exigrep's output now also includes lines that are not associated with any
6121 message, but which match the given pattern. Implemented by a patch from
6122 Martin Sluka, which also tidied up the Perl a bit.
6124 71. Recipient callout verification, like sender verification, was using <> in
6125 the MAIL FROM command. This isn't really the right thing, since the actual
6126 sender may affect whether the remote host accepts the recipient or not. I
6127 have changed it to use the actual sender in the callout; this means that
6128 the cache record is now keyed on a recipient/sender pair, not just the
6129 recipient address. There doesn't seem to be a real danger of callout loops,
6130 since a callout by the remote host to check the sender would use <>.
6131 [SEE ABOVE: changed after hitting problems.]
6133 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
6134 temporary errors. However, in the case of such a code being given after
6135 the end of a data transmission (i.e. after ".") Exim was failing to write
6136 a retry record for the message. (Yes, there was some broken host that was
6137 actually sending 8xx at this point.)
6139 73. An unknown lookup type in a host list could cause Exim to panic-die when
6140 the list was checked. (An example that provoked this was putting <; in the
6141 middle of a list instead of at the start.) If this happened during a DATA
6142 ACL check, a -D file could be left lying around. This kind of configuration
6143 error no longer causes Exim to die; instead it causes a defer error. The
6144 incident is still logged to the main and panic logs.
6146 74. Buglet left over from Exim 3 conversion. The message "too many messages
6147 in one connection" was written to the rejectlog but not the mainlog, except
6148 when address rewriting (yes!) was being logged.
6150 75. Added write_rejectlog option.
6152 76. When a system filter was run not as root (that is, when system_filter_user
6153 was set), the values of the $n variables were not being returned to the
6154 main process; thus, they were not subsequently available in the $sn
6157 77. Added +return_path_on_delivery log selector.
6159 78. A connection timeout was being treated differently from recipients deferred
6160 when testing hosts_max_try with a message that was older than the host's
6161 retry timeout. (The host should not be counted, thus allowing all hosts to
6162 be tried at least once before bouncing.) This may have been the cause of an
6163 occasionally reported bug whereby a message would remain on the queue
6164 longer than the retry timeout, but would be bounced if a delivery was
6165 forced. I say "may" because I never totally pinned down the problem;
6166 setting up timeout/retry tests is difficult. See also the next item.
6168 79. The ultimate address timeout was not being applied to errors that involved
6169 a combination of host plus message (for example, a timeout on a MAIL
6170 command). When an address resolved to a number of possible hosts, and they
6171 were not all tried for each delivery (e.g. because of hosts_max_try), a
6172 message could remain on the queue longer than the retry timeout.
6174 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
6177 81. Fixed an obscure SMTP outgoing bug which required at least the following
6178 conditions: (a) there was another message waiting for the same server;
6179 (b) the server returned 5xx to all RCPT commands in the first message so
6180 that the message was not completed; (c) the server dropped the connection
6181 or gave a negative response to the RSET that Exim sends to abort the
6182 transaction. The observed case was a dropped connection after DATA that had
6183 been sent in pipelining mode. That is, the server had advertised PIPELINING
6184 but was not implementing it correctly. The effect of the bug was incorrect
6185 behaviour, such as trying another host, and this could lead to a crash.
6191 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
6192 and daemon.c were passed as pointers to ints; they should have been
6193 pointers to socklen_t variables (which are typically unsigned ints).
6195 2. Some signed/unsigned type warnings in the os.c file for Linux have been
6198 3. Fixed a really odd bug that affected only the testing scheme; patching a
6199 certain fixed string in the binary changed the value of another string that
6200 happened to be identical to the end of the original first string.
6202 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
6203 name", it returns that address as the IP address. On some operating
6204 systems (e.g. Solaris), it also passes back the IP address string as the
6205 "host name". However, on others (e.g. Linux), it passes back an empty
6206 string. Exim wasn't checking for this, and was changing the host name to an
6207 empty string, assuming it had been canonicalized.
6209 5. Although rare, it is permitted to have more than one PTR record for a given
6210 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
6211 all the names associated with an address, because they do in Solaris.
6212 However, it seems that they do not in Linux for data that comes from the
6213 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
6214 I found this out when I moved to a new Linux workstation and tried to run
6215 the Exim test suite.
6217 To get round this problem I have changed the code so that it now does its
6218 own call to the DNS to look up PTR records when searching for a host name.
6219 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
6220 addresses that are only in /etc/hosts are still found.
6222 This behaviour is, however, controlled by an option called host_lookup_
6223 order, which defaults to "bydns:byaddr". If people want to use the other
6224 order, or indeed, just use one or the other means of lookup, they can
6225 specify it in this variable.
6227 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
6228 some operating systems, this comes back from gethostbyaddr() as an empty
6229 string, and this is what Exim used to test for. However, it seems that in
6230 other systems, "." is yielded. Exim now tests for this case too.
6232 7. The values of check_spool_space and check_log_space are now held internally
6233 as a number of kilobytes instead of an absolute number of bytes. If a
6234 numbers is specified without 'K' or 'M', it is rounded up to the nearest
6235 kilobyte. This means that much larger values can be stored.
6237 8. Exim monitor: an attempt to get the action menu when not actually pointing
6238 at a message produces an empty menu entitled "No message selected". This
6239 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
6240 no entries in it ("Shell widget menu has zero width and/or height"). So I
6241 have added a single, blank menu entry in this case.
6243 9. Added ${quote_local_part.
6245 10. MIME decoding is now applied to the contents of Subject: header lines when
6248 11. Now that a reference to $sender_host_address automatically causes a reverse
6249 lookup to occur if necessary (4.13/18), there is no need to arrange for a
6250 host lookup before query-style lookups in lists that might use this
6251 variable. This has therefore been abolished, and the "net-" prefix is no
6252 longer necessary for query-style lookups.
6254 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
6255 have been a typo for LFLAGS, so it has been changed.
6257 13. The install script calls Exim with "-C /dev/null" in order to find the
6258 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
6259 to be output. However, since Exim outputs its version number before the
6260 error, it didn't break the script. It just looked ugly. I fixed this by
6261 always allowing "-C /dev/null" if the caller is root.
6263 14. Ignore overlarge ACL variable number when reading spool file - insurance
6264 against a later release with more variables having written the file.
6266 15. The standard form for an IPv6 address literal was being rejected by EHLO.
6267 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
6268 this, as well as the form without the "IPv6" on the front.
6270 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
6271 OS/Makefile-Darwin file.
6273 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
6274 only to LDAP libraries that do not have LDAP_OPT_DEREF.
6276 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
6277 variable. However, some OS use ints for inodes. I've added cast to long int
6278 to get rid of the compiler warning.
6280 19. I had forgotten to lock out "/../" in configuration file names when
6281 ALT_CONFIG_PREFIX was set.
6283 20. Routers used for verification do not need to specify transports. However,
6284 if such a router generated a host list, and callout was configured, Exim
6285 crashed, because it could not find a port number from the (non-existent)
6286 transport. It now assumes port 25 in this circumstance.
6288 21. Added the -t option to exigrep.
6290 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
6291 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
6292 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
6293 them is set, LOOKUP_LSEARCH is forced.
6295 23. "exim -bV" now outputs a list of lookups that are included in the binary.
6297 24. Added sender and host information to the "rejected by local_scan()" log
6298 line; previously there was no indication of these.
6300 25. Added .include_if_exists.
6302 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
6303 Linux. It turns out that not all file systems support this. Apparently some
6304 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
6305 do it.) To cope with this, the error EINVAL, which means that sync-ing is
6306 not supported on the file descriptor, is now ignored when Exim is trying to
6307 sync a directory. This applies only to Linux.
6309 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
6311 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
6312 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
6315 29. Added support for maildirsize files from supplied patch (modified a bit).
6317 30. The use of :fail: followed by an empty string could lead Exim to respond to
6318 sender verification failures with (e.g.):
6320 550 Verification failed for <xxx>
6321 550 Sender verify failed
6323 where the first response line was missing the '-' that indicates it is not
6324 the final line of the response.
6326 31. The loop for finding the name of the user that called Exim had a hardwired
6327 limit of 10; it now uses the value of finduser_retries, which is used for
6328 all other user lookups.
6330 32. Added $received_count variable, available in data and not_smtp ACLs, and at
6333 33. Exim was neglecting to zero errno before one call of strtol() when
6334 expanding a string and expecting an integer value. On some systems this
6335 resulted in spurious "integer overflow" errors. Also, it was casting the
6336 result into an int without checking.
6338 34. Testing for a connection timeout using "timeout_connect" in the retry rules
6339 did not work. The code looks as if it has *never* worked, though it appears
6340 to have been documented since at least release 1.62. I have made it work.
6342 35. The "timeout_DNS" error in retry rules, also documented since at least
6343 1.62, also never worked. As it isn't clear exactly what this means, and
6344 clearly it isn't a major issue, I have abolished the feature by treating it
6345 as "timeout", and writing a warning to the main and panic logs.
6347 36. The display of retry rules for -brt wasn't always showing the error code
6350 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
6351 timeout_connect_A, timeout_connect_MX.
6353 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
6354 to the empty sender.
6356 39. The daemon was not analysing the content of -oX till after it had closed
6357 stderr and disconnected from the controlling terminal. This meant that any
6358 syntax errors were only noted on the panic log, and the return code from
6359 the command was 0. By re-arranging the code a little, I've made the
6360 decoding happen first, so such errors now appear on stderr, and the return
6361 code is 1. However, the actual setting up of the sockets still happens in
6362 the disconnected process, so errors there are still only recorded on the
6365 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
6366 connections (as happens on some IP stacks) was logged at start up time as
6367 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
6368 it from "IPv6 and IPv4", which means that two separate sockets are being
6371 41. The debug output for gethostbyname2() or getipnodebyname() failures now
6372 says whether AF_INET or AF_INET6 was passed as an argument.
6374 42. Exiwhat output was messed up when time zones were included in log
6377 43. Exiwhat now gives more information about the daemon's listening ports,
6378 and whether -tls-on-connect was used.
6380 44. The "port" option of the smtp transport is now expanded.
6382 45. A "message" modifier in a "warn" statement in a non-message ACL was being
6383 silently ignored. Now an error message is written to the main and panic
6386 46. There's a new ACL modifier called "logwrite" which writes to a log file
6387 as soon as it is encountered.
6389 47. Added $local_user_uid and $local_user_gid at routing time.
6391 48. Exim crashed when trying to verify a sender address that was being
6394 49. Exim was recognizing only a space character after ".include". It now also
6395 recognizes a tab character.
6397 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
6398 extracting the relevant information from the specification. The man page no
6399 longer contains scrambled data for the -d option, and I've added a section
6400 at the front about calling Exim under different names.
6402 51. Added "extra_headers" argument to the "mail" command in filter files.
6404 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
6407 53. Installed eximstats 1.29.
6409 54. Added transport_filter_timeout as a generic transport option.
6411 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
6412 Cc: header lines. This was required by RFC 822, but it not required by RFC
6415 56. Exim used to add From:, Date:, and Message-Id: header lines to any
6416 incoming messages that did not have them. Now it does so only if the
6417 message originates locally, that is, if there is no associated remote host
6418 address. When Resent- header lines are present, this applies to the Resent-
6419 lines rather than the non-Resent- lines.
6421 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
6422 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
6424 58. Messages for configuration errors now include the name of the main
6425 configuration file - useful now that there may be more than one file in a
6426 list (.included file names were always shown).
6428 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
6429 for those rare installations that do not start the daemon as root or run it
6430 setuid root. I've cut out the call to initgroups() if the daemon is not
6433 60. The Exim user and group can now be bound into the binary as text strings
6434 that are looked up at the start of Exim's processing.
6436 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
6438 62. Added $mailstore_basename variable.
6440 63. Installed patch to sieve.c from Michael Haardt.
6442 64. When Exim failed to open the panic log after failing to open the main log,
6443 the original message it was trying to log was written to stderr and debug
6444 output, but if they were not available (the usual case in production), it
6445 was lost. Now it is written to syslog before the two lines that record the
6446 failures to open the logs.
6448 65. Users' Exim filters run in subprocesses under the user's uid. It is
6449 possible for a "deliver" command or an alias in a "personal" command to
6450 provoke an address rewrite. If logging of address rewriting is configured,
6451 this fails because the process is not running as root or exim. There may be
6452 a better way of dealing with this, but for the moment (because 4.30 needs
6453 to be released), I have disabled address rewrite logging when running a
6454 filter in a non-root, non-exim process.
6460 1. The buildconfig auxiliary program wasn't quoting the value set for
6461 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
6462 not defined. This bug was masked in 4.22 by the effect that was fixed in
6465 2. Some messages that were rejected after a message id was allocated were
6466 shown as "incomplete" by exigrep. It no longer does this for messages that
6467 are rejected by local_scan() or the DATA or non-SMTP ACLs.
6469 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
6470 have allow_domain_literals set, the ID did not get logged in the <= line.
6471 Domain literals are now always recognized in Message-ID: header lines.
6473 4. The first argument for a ${extract expansion item is the key name or field
6474 number. Leading and trailing spaces in this item were not being ignored,
6475 causing some misleading effects.
6477 5. When deliver_drop_privilege was set, single queue runner processes started
6478 manually (i.e. by the command "exim -q") or by the daemon (which uses the
6479 same command in the process it spins off) were not dropping privilege.
6481 6. When the daemon running as "exim" started a queue runner, it always
6482 re-executed Exim in the spun-off process. This is a waste of effort when
6483 deliver_drop_privilege is set. The new process now just calls the
6484 queue-runner function directly.
6490 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
6493 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
6494 ignored. Though the use of -oP was forcing the writing of a pid file, it
6495 was always written to the default place.
6497 3. If the message "no IP address found for host xxxx" is generated during
6498 incoming verification, it is now followed by identification of the incoming
6499 connection (so you can more easily find what provoked it).
6501 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
6503 5. Added some features to "harden" Exim a bit more against certain attacks:
6505 (a) There is now a build-time option called FIXED_NEVER_USERS that can
6506 be put in Local/Makefile. This is like the never_users runtime option,
6507 but it cannot be overridden. The default setting is "root".
6509 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
6510 prefix string with which any file named in a -C command line option
6513 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
6514 is retained for -C and -D only if the caller of Exim is root. Without
6515 it, the exim user may also use -C and -D and retain privilege.
6517 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
6518 command line option is disabled.
6520 6. Macro names set by the -D option must start with an upper case letter, just
6521 like macro names defined in the configuration file.
6523 7. Added "dereference=" facility to LDAP.
6525 8. Two instances of the typo "uknown" in the source files are fixed.
6527 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
6528 the Configure-Makefile script screwed up while processing it.
6530 10. Incorporated PCRE 4.4.
6532 11. The SMTP synchronization check was not operating right at the start of an
6533 SMTP session. For example, it could not catch a HELO sent before the client
6534 waited for the greeting. There is now a check for outstanding input at the
6535 point when the greeting is written. Because of the duplex, asynchronous
6536 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
6537 way, but not yet received, when the check is performed.
6539 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
6540 on TCP/IP sockets, because this apparently causes some broken clients to
6543 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
6544 unchanged) from the Cygwin maintainer.
6546 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
6547 was supported instead of testing for mbx. Effectively a typo.
6549 15. The spa authenticator server code was not checking that the input it
6550 received was valid base64.
6552 16. The debug output line for the "set" modifier in ACLs was not showing the
6553 name of the variable that was being set.
6555 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
6557 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
6558 lookup on its own. Something else was needed to trigger the lookup. For
6559 example, a match in host_lookup or the need for a host name in a host list.
6560 Now, if $sender_host_name is referenced and the host name has not yet been
6561 looked up, a lookup is performed. If the lookup fails, the variable remains
6562 empty, and $host_lookup_failed is set to "1".
6564 19. Added "eqi" as a case-independent comparison operator.
6566 20. The saslauthd authentication condition could segfault if neither service
6567 nor realm was specified.
6569 21. If an overflowing value such as "2048M" was set for message_size_limit, the
6570 error message that was logged was misleading, and incoming SMTP
6571 connections were dropped. The message is now more accurate, and temporary
6572 errors are given to SMTP connections.
6574 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
6575 (except RSET) with a 421 error, until QUIT is received. However, it was
6576 failing to send a response to QUIT.
6578 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
6579 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
6580 ACL is now run after the helo_try_verify_hosts code.
6582 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
6583 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
6584 case-independent, so other case variants are also recognized). Apparently
6585 some systems use these upper case variants.
6587 25. If more than two messages were waiting for the same host, and a transport
6588 filter was specified for the transport, Exim sent two messages over the
6589 same TCP/IP connection, and then failed with "socket operation on non-
6590 socket" when it tried to send the third.
6592 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
6594 27. The extern definition of crypt16() in expand.c was not being excluded when
6595 the OS had its own crypt16() function.
6597 28. Added bounce_return_body as a new option, and bounce_return_size_limit
6598 as a preferred synonym for return_size_limit, both as an option and as an
6601 29. Added LIBS=-liconv to OS/Makefile-OSF1.
6603 30. Changed the default configuration ACL to relax the local part checking rule
6604 for addresses that are not in any local domains. For these addresses,
6605 slashes and pipe symbols are allowed within local parts, but the sequence
6606 /../ is explicitly forbidden.
6608 31. SPA server authentication was not clearing the challenge buffer before
6611 32. log_message in a "warn" ACL statement was writing to the reject log as
6612 well as to the main log, which contradicts the documentation and doesn't
6613 seem right (because no rejection is happening). So I have stopped it.
6615 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
6616 However, I am unable to do any testing of this.
6618 34. Fixed an infelicity in the appendfile transport. When checking directories
6619 for a mailbox, to see if any needed to be created, it was accidentally
6620 using path names with one or more superfluous leading slashes; tracing
6621 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
6623 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
6624 contents are added to the log line that is written for every discarded
6625 recipient. (Previously a log_message setting was ignored.)
6627 36. The ${quote: operator now quotes the string if it is empty.
6629 37. The install script runs exim in order to find its version number. If for
6630 some reason other than non-existence or emptiness, which it checks, it
6631 could not run './exim', it was installing it with an empty version number,
6632 i.e. as "exim-". This error state is now caught, and the installation is
6635 38. An argument was missing from the function that creates an error message
6636 when Exim fails to connect to the socket for saslauthd authentication.
6637 This could cause Exim to crash, or give a corrupted message.
6639 39. Added isip, isip4, and isip6 to ${if conditions.
6641 40. The ACL variables $acl_xx are now saved with the message, and can be
6642 accessed later in routers, transports, and filters.
6644 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
6645 strings in the file are not string-expanded.
6647 42. If a MAIL command specified a SIZE value that was too large to fit into an
6648 int variable, the check against message_size_limit failed. Such values are
6649 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
6650 day this will have to be increased, but I don't think I want to be around
6651 when emails are that large.
6658 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
6659 iconv() is not standard in FreeBSD.
6661 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
6662 IPv6 enabled. The observed symptom was a segmentation fault on return from
6663 the function os_common_find_running_interfaces() in src/os.c.
6665 3. In the check_special_case() function in daemon.c I had used "errno" as an
6666 argument name, which causes warnings on some systems. This was basically a
6667 typo, since it was named "eno" in the comments!
6669 4. The code that waits for the clock to tick (at a resolution of some fraction
6670 of a second) so as to ensure message-id uniqueness was always waiting for
6671 at least one whole tick, when it could have waited for less. [This is
6672 almost certainly not relevant at current processor speeds, where it is
6673 unlikely to ever wait at all. But we try to future-proof.]
6675 5. The function that sleeps for a time interval that includes fractions of a
6676 second contained a race. It did not block SIGALRM between setting the
6677 timer, and suspending (a couple of lines later). If the interval was short
6678 and the sigsuspend() was delayed until after it had expired, the suspension
6679 never ended. On busy systems this could lead to processes getting stuck for
6682 6. Some uncommon configurations may cause a lookup to happen in a queue runner
6683 process, before it forks any delivery processes. The open lookup caching
6684 mechanism meant that the open file or database connection was passed into
6685 the delivery process. The problem was that delivery processes always tidy
6686 up cached lookup data. This could cause a problem for the next delivery
6687 process started by the queue runner, because the external queue runner
6688 process does not know about the closure. So the next delivery process
6689 still has data in the lookup cache. In the case of a file lookup, there was
6690 no problem because closing a file descriptor in a subprocess doesn't affect
6691 the parent. However, if the lookup was caching a connection to a database,
6692 the connection was closed, and the second delivery process was likely to
6693 see errors such as "PGSQL: query failed: server closed the connection
6694 unexpectedly". The problem has been fixed by closing all cached lookups
6695 in a queue runner before running a delivery process.
6697 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
6698 seem to have the "const" qualifier which it has on other OS. I've
6701 8. Change 4.21/2 was too strict. It is only if there are two authenticators
6702 *of the same type* (client or server) with the same public name that an
6703 error should be diagnosed.
6705 9. When Exim looked up a host name for an IP address, but failed to find the
6706 original IP address when looking up the host name (a safety check), it
6707 output the message "<ip address> does not match any IP for NULL", which was
6708 confusing, to say the least. The bug was that the host name should have
6709 appeared instead of "NULL".
6711 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
6712 user that is built into the binary, and the -C or -D options is used, root
6713 privilege is dropped before the configuration file is read. In addition,
6714 logging is switched to stderr instead of the normal log files. If the
6715 configuration then re-defines the Exim user, the unprivileged environment
6716 is probably not what is expected, so Exim logs a panic warning message (but
6719 However, if deliver_drop_privilege is set, the unprivileged state may well
6720 be exactly what is intended, so the warning has been cut out in that case,
6721 and Exim is allowed to try to write to its normal log files.
6727 1. smtp_return_error_details was not giving details for temporary sender
6728 or receiver verification errors.
6730 2. Diagnose a configuration error if two authenticators have the same public
6733 3. Exim used not to create the message log file for a message until the first
6734 delivery attempt. This could be confusing when incoming messages were held
6735 for policy or load reasons. The message log file is now created at the time
6736 the message is received, and an initial "Received" line is written to it.
6738 4. The automatically generated man page for command line options had a minor
6739 bug that caused no ill effects; however, a more serious problem was that
6740 the procedure for building the man page automatically didn't always
6741 operate. Consequently, release 4.20 contains an out-of-date version. This
6742 shouldn't happen again.
6744 5. When building Exim with embedded Perl support, the script that builds the
6745 Makefile was calling 'perl' to find its compile-time parameters, ignoring
6746 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
6748 6. The freeze_tell option was not being used for messages that were frozen on
6749 arrival, either by an ACL or by local_scan().
6751 7. Added the smtp_incomplete_transaction log selector.
6753 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
6754 was accepting AUTH without a new EHLO.
6756 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
6757 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
6760 10. Logging of TCP/IP connections (when configured) now happens in the main
6761 daemon process instead of the child process, so that the TCP/IP connection
6762 count is more accurate (but it can never be perfect).
6764 11. The use of "drop" in a nested ACL was not being handled correctly in the
6765 outer ACL. Now, if condition failure induced by the nested "drop" causes
6766 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
6767 or "require"), the connection is dropped.
6769 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
6770 that yield "discard" can now be used with an "accept" or a "discard" verb,
6771 but an error is generated for any others (because I can't see a useful way
6772 to define what should happen).
6774 13. When an ACL is read dynamically from a file (or anywhere else), the lines
6775 are now processed in the same way as lines in the Exim configuration file.
6776 In particular, continuation lines are supported.
6778 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
6780 15. Added -ti meaning -t -i.
6782 16. Check for letters, digits, hyphens, and dots in the names of dnslist
6783 domains, and warn by logging if others are found.
6785 17. At least on BSD, alignment is not guaranteed for the array of ifreq's
6786 returned from GIFCONF when Exim is trying to find the list of interfaces on
6787 a host. The code in os.c has been modified to copy each ifreq to an aligned
6788 structure in all cases.
6790 Also, in some cases, the returned ifreq's were being copied to a 'struct
6791 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
6792 means the last couple of bytes of an IPv6 address could be chopped if the
6793 ifreq contained only a normal sockaddr (14 bytes storage).
6795 18. Named domain lists were not supported in the hosts_treat_as_local option.
6796 An entry such as +xxxx was not recognized, and was treated as a literal
6799 19. Ensure that header lines added by a DATA ACL are included in the reject log
6800 if the ACL subsequently rejects the message.
6802 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
6803 MD5 (which is deprecated).
6805 21. When testing a filter file using -bf, Exim was writing a message when it
6806 took the sender from a "From " line in the message, but it was not doing so
6807 when it took $return_path from a Return-Path: header line. It now does.
6809 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
6810 with a valid header line field name (a series of printing characters
6811 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
6813 23. Changed "disc" in the source to "disk" to conform to the documentation and
6814 the book and for uniformity.
6816 24. Ignore Sendmail's -Ooption=value command line item.
6818 25. When execve() failed while trying to run a command in a pipe transport,
6819 Exim was returning EX_UNAVAILABLE (69) from the subprocess. However, this
6820 could be confused with a return value of 69 from the command itself. This
6821 has been changed to 127, the value the shell returns if it is asked to run
6822 a non-existent command. The wording for the related log line suggests a
6823 non-existent command as the problem.
6825 26. If received_header_text expands to an empty string, do not add a Received:
6826 header line to the message. (Well, it adds a token one on the spool, but
6827 marks it "old" so that it doesn't get used or transmitted.)
6829 27. Installed eximstats 1.28 (addition of -nt option).
6831 28. There was no check for failure on the call to getsockname() in the daemon
6832 code. This can fail if there is a shortage of resources on the system, with
6833 ENOMEM, for example. A temporary error is now given on failure.
6835 29. Contrary to the C standard, it seems that in some environments, the
6836 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
6837 program. Exim now does this explicitly; it affects the formatting of
6838 timestamps using strftime().
6840 30. If exiqsumm was given junk data, it threw up some uninitialized variable
6841 complaints. I've now initialized all the variables, to avoid this.
6843 32. Header lines added by a system filter were not being "seen" during
6844 transport-time rewrites.
6846 33. The info_callback() function passed to OpenSSL is set up with type void
6847 (*)(SSL *, int, int), as described somewhere. However, when calling the
6848 function (actually a macro) that sets it up, the type void(*)() is
6849 expected. I've put in a cast to prevent warnings from picky compilers.
6851 34. If a DNS black list lookup found a CNAME record, but there were no A
6852 records associated with the domain it pointed at, Exim crashed.
6854 35. If a DNS black list lookup returned more than one A record, Exim ignored
6855 all but the first. It now scans all returned addresses if a particular IP
6856 value is being sought. In this situation, the contents of the
6857 $dnslist_value variable are a list of all the addresses, separated by a
6860 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
6861 to accept a host name and all its aliases if the forward lookup for any of
6862 them yielded the IP address of the incoming connection. Now it accepts only
6863 those names whose forward lookup yields the correct IP address. Any other
6864 names are discarded. This closes a loophole whereby a rogue DNS
6865 administrator could create reverse DNS records to break through a
6866 wildcarded host restriction in an ACL.
6868 37. If a user filter or a system filter that ran in a subprocess used any of
6869 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
6870 the wrong values were passed to the pipe command ($thisaddress had the
6871 value of $0, $0 had the value of $1, etc). This bug was introduced by
6872 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
6874 38. Improved the line breaking for long SMTP error messages from ACLs.
6875 Previously, if there was no break point between 40 and 75 characters, Exim
6876 left the rest of the message alone. Two changes have been made: (a) I've
6877 reduced the minimum length to 35 characters; (b) if it can't find a break
6878 point between 35 and 75 characters, it looks ahead and uses the first one
6879 that it finds. This may give the occasional overlong line, but at least the
6880 remaining text gets split now.
6882 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
6883 file descriptors might be low, and that setting 1000 would always raise it.
6884 It turns out that in some environments, the limit is already over 1000 and
6885 that lowering it causes trouble. So now Exim takes care not to decrease it.
6887 40. When delivering a message, the value of $return_path is set to $sender_
6888 address at the start of routing (routers may change the value). By an
6889 oversight, this default was not being set up when an address was tested by
6890 -bt or -bv, which affected the outcome if any router or filter referred to
6893 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
6894 log only when "message" or "log_message" are set. However, if one of the
6895 conditions was an address verification, or a call to a nested ACL, the
6896 messages generated by the underlying test were being passed through. This
6897 no longer happens. The underlying message is available in $acl_verify_
6898 message for both "message" and "log_message" expansions, so it can be
6899 passed through if needed.
6901 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
6902 new expansion $bh_ to give the encoded byte string without charset
6903 translation. Translation happens only if iconv() is available; HAVE_ICONV
6904 indicates this at build time. HEADERS_CHARSET gives the charset to
6905 translate to; headers_charset can change it in the configuration, and
6906 "headers charset" can change it in an individual filter file.
6908 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
6909 that creates RFC 2047 encoded "words" labels them as that charset instead
6910 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
6911 expansion operator; (ii) when Exim creates a From: line for a local
6912 message; (iii) when a header line is rewritten to include a "phrase" part.
6914 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
6915 buggy, causing it to skip the first lines of messages whose message ID
6916 ended in 'D'. This would not have bitten before Exim release 4.14, because
6917 message IDs were unlikely to end in 'D' before then. The effect was to have
6918 incorrect size information for certain domains.
6920 45. #include "config.h" was missing at the start of the crypt16.c module. This
6921 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
6924 46. If there was a timeout during a "random" callout check, Exim treated it as
6925 a failure of the random address, and carried on sending RSET and the real
6926 address. If the delay was just some slowness somewhere, the response to the
6927 original RCPT would be taken as a response to RSET and so on, causing
6928 mayhem of various kinds.
6930 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
6931 when I implemented it. It didn't allow for the fact that some option values
6932 may legitimately be negative (e.g. size_addition), and it didn't even do
6933 the right test for positive values.
6935 48. Domain names in DNS records are case-independent. Exim always looks them up
6936 in lower case. Some resolvers return domain names in exactly the case they
6937 appear in the zone file, that is, they may contain uppercase letters. Not
6938 all resolvers do this - some return always lower case. Exim was treating a
6939 change of case by a resolver as a change of domain, similar to a widening
6940 of a domain abbreviation. This triggered its re-routing code and so it was
6941 trying to route what was effectively the same domain again. This normally
6942 caused routing to fail (because the router wouldn't handle the domain
6943 twice). Now Exim checks for this case specially, and just changes the
6944 casing of the domain that it ultimately uses when it transmits the message
6947 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
6950 50. If a filter generated a file delivery with a non-absolute name (possible if
6951 no home directory exists for the router), the forbid_file option was not
6954 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
6955 the existing equality matching.
6957 52. Exim was using ints instead of ino_t variables in some places where it was
6958 dealing with inode numbers.
6960 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
6961 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
6962 called TMPDIR, and if it finds it is different, it changes its value.
6964 54. The smtp_printf() function is now made available to local_scan() so
6965 additional output lines can be written before returning. There is also an
6966 smtp_fflush() function to enable the detection of a dropped connection.
6967 The variables smtp_input and smtp_batched_input are exported to
6970 55. Changed the default runtime configuration: the message "Unknown user"
6971 has been removed from the ACL, and instead placed on the localuser router,
6972 using the cannot_route_message feature. This means that any verification
6973 failures that generate their own messages won't get overridden. Similarly,
6974 the "Unrouteable address" message that was in the ACL for unverifiable
6975 relay addresses has also been removed.
6977 56. Added hosts_avoid_esmtp to the smtp transport.
6979 57. The exicyclog script was not checking for the esoteric option
6980 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
6981 will work only if exicyclog is run under the appropriate euid.
6983 58. Following a discussion on the list, the rules by which Exim recognises line
6984 endings on incoming messages have been changed. The -dropcr and drop_cr
6985 options are now no-ops, retained only for backwards compatibility. The
6986 following line terminators are recognized: LF CRLF CR. However, special
6987 processing applies to CR:
6989 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
6990 nor a local message in the state where . is a terminator.
6992 (ii) If a bare CR is encountered in a header line, an extra space is added
6993 after the line terminator so as not to end the header. The reasoning
6994 behind this is that bare CRs in header lines are most likely either
6995 to be mistakes, or people trying to play silly games.
6997 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
6998 was being incorrectly given as 18 bytes larger than it should have been.
6999 This is a VOB (very old bug).
7001 60. This may never have affected anything current, but just in case it has:
7002 When the local host is found other than at the start of a list of hosts,
7003 the local host, those with the same MX, and any that follow, are discarded.
7004 When the list in question was part of a longer list of hosts, the following
7005 hosts (not currently being processed) were also being discarded. This no
7006 longer happens. I'm not sure if this situation could ever has previously
7009 61. Added the "/MX" feature to lists of hosts in the manualroute and query
7012 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
7013 header. This is something that is recommended in a new Internet Draft, and
7014 is something that is documented as being done by Sendmail. There are two
7015 possible values. For messages generated by the autoreply transport, Exim
7018 Auto-Submitted: auto-replied
7020 whereas for all other generated messages (e.g. bounces) it adds
7022 Auto-Submitted: auto-generated
7024 63. The "personal" condition in filters now includes a test for the
7025 Auto-Submitted: header. If it contains the string "auto-" the message it
7026 not considered personal.
7028 64. Added rcpt_include_affixes as a generic transport option.
7030 65. Added queue_only_override (default true).
7032 66. Added the syslog_duplication option.
7034 67. If what should have been the first header line of a message consisted of
7035 a space followed by a colon, Exim was mis-interpreting it as a header line.
7036 It isn't of course - it is syntactically invalid and should therefore be
7037 treated as the start of the message body. The misbehaviour could have
7038 caused a number of strange effects, including loss of data in subsequent
7039 header lines, and spool format errors.
7041 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
7042 client host had authenticated. This control can now be exercised by an ACL
7043 for more flexibility.
7045 69. By default, callouts do not happen when testing with -bh. There is now a
7046 variant, -bhc, which does actually run the callout code, including
7047 consulting and updating the callout cache.
7049 70. Added support for saslauthd authentication, courtesy of Alexander
7052 71. If statvfs() failed on the spool or log directories while checking their
7053 size for availability, Exim confusingly gave the error "space shortage".
7054 Furthermore, in debugging mode it crashed with a floating point exception.
7055 These checks are done if check_{spool,log}_{space,inodes} are set, and when
7056 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
7057 serious problem, Exim now writes to the main and panic logs when this
7058 happens, with details of the failure. It then refuses to accept the
7059 incoming message, giving the message "spool directory problem" or "log
7060 directory problem" with a 421 code for SMTP messages.
7062 72. When Exim is about to re-exec itself, it ensures that the file descriptors
7063 0, 1, and 2 exist, because some OS complain for execs without them (see
7064 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
7065 descriptors. However, the code omitted to check that the open succeeded,
7066 causing mysterious errors if for some reason the permissions on /dev/null
7067 got screwed. Now Exim writes a message to the main and panic logs, and
7068 bombs out if it can't open /dev/null.
7070 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
7071 interact so that it is all more flexible. It is supposed to remain
7072 backwards compatible. Also added extra_local_interfaces.
7074 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
7075 to bomb out with an assertion failure - to the client this appears as a
7076 connection drop. This problem occurs in the part of the code that was taken
7077 from the Samba project. Fortunately, the assertion is in a very simple
7078 function, so I have fixed this by reproducing the function inline in the
7079 one place where it is called, and arranging for authentication to fail
7080 instead of killing the process with assert().
7082 75. The SPA client code was not working when the server requested OEM rather
7083 than Unicode encoding.
7085 76. Added code to make require_files with a specific uid setting more usable in
7086 the case where statting the file as root fails - usually a non-root-mounted
7087 NFS file system. When this happens and the failure is EACCES, Exim now
7088 forks a subprocess and does the per-uid checking as the relevant uid.
7090 77. Added process_log_path.
7092 78. If log_file_path was not explicitly set, a setting of check_log_space or
7093 check_log_inodes was ignored.
7095 79. If a space check for the spool or log partitions fails, the incident is now
7096 logged. Of course, in the latter case the data may get lost...
7098 80. Added the %p formatting code to string_format() so that it can be used to
7099 print addresses in debug_print(). Adjusted all the address printing in the
7100 debugging in store.c to use %p rather than %d.
7102 81. There was a concern that a line of code in smtp_in.c could overflow a
7103 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
7104 initially expressed, the concern was not well-founded, because trailing
7105 spaces are removed early. However, if the trailing spaces were followed by
7106 a NULL, they did not get removed, so the overflow was possible. Two fixes
7109 (a) I re-wrote the offending code in a cleaner fashion.
7110 (b) If an incoming SMTP command contains a NULL character, it is rejected
7113 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
7114 runs initgroups(), so that if the Exim user is in any additional groups,
7115 they will be used during message reception.
7121 The change log for 4.20 and earlier releases has been archived.
projects / exim.git / blob